Kp-Mini Self-Destruction [Build] [Archive]

deadwood83

11-01-2020, 11:37 PM

Alright, following the death of my hellcat project, (http://www.chinariders.raptorrefuge.com/showthread.php?t=26046) I needed to heal my heart.

My younger brother has a KP mini, and we tricked it out with BRT parts. 36/31mm ported head, hardened valves, stiffer springs, upgraded cam, 70mm piston and matched cylinder, PWK32. It is one dangerous machine.

I... really like the counterbalanced engines. After coming from the vibrations and inevitable transmission failures in the ZS190, I was ready for a change.

Enter, 2021 KP Mini! In white, of course.

23222

Anyway, the 223cc is awesome. No question. That bike has about 20-21hp and gobs of torque. But... is it really enough? No, no it is not. While mini shopping, I found the Tianda TDR300. It's a little small displacement affordable race bike. That, of course, lead me to researching the engine. Apparently it's a Zongshen 300 with 6 forward gears pushing 25-28hp.

Six gears... 20% power increase... mini bike? More than double the stock power? What? And it (in theory) fits? Whaaaat?
23223

I sat, and I stared, and I soul-searched for a good 28 hours. Then I ordered the engine. What is impulse control? Idk, still trying to figure that out.

But there are some complications.

This engine is water cooled. Where do you even put suitably sized radiators on a KP mini? I started by looking at CBR rads and KTM CX-W rads. Fitting those in the front fairings seemed... complicated.
The intake points the 'wrong' way. This ended up being a non-issue. The stock airbox had to come out for the 223cc big bore, I have no illusions of it being adequate for this engine.
Temperature control. Typically, this would go on an EFI install which has a fan controller. I do not have the patience nor dyno to properly tune, and the tank does not really accommodate a pump.
I cannot weld. I do not have the tools, nor have I ever welded (unless you count plumbing and electronics soldering. However, I can bend sheet steel with a 20-ton press and I can make molds up to 12"x12"x12" with my 3D printers.
Sprocket offset? ALso a non-issue. The cover bolts are your bog-standard 8mm hex heads. By using photogrammetry, I was able to approximate the sprocket distance from mount. It should be identical to the pre-installed 157FMJ engine.
Footpegs. The KP mini mounts the footpegs and kick stand on the bottom of the engine. Luckily, this engine case has those mounts present (or so it looks from the pictures).
Height? This engine is actually shorter.
Front engine mount. I will have to fab at least a new lower half. THis shouldn't be too bad. I have a friend who can weld the top and bottom together, and I can bend/press from a 3d printed mold I design in SolidWorks.
Can the rear hub take that power? Ehhhhhhhhhhh. Probably not. if it breaks, I will look into the solid Grom CBR conversion hubs. That said, I do not go hard off the line. I prefer clutches to last, so I go hard after the line. Probably only marginally better.

Okay, so now to the solutions.

I found a build where a gentleman cools a 1000CC bike with an under-seat radiator. This seems nifty, but it is also where the battery lives. 23224
But where would this fit on a KP? I recalled a shot from Small ENgine Velocity's youtube channel. 23225
Look at all that space! It's occupied by a single wire (and an airbox which will no longer be in the bike). Bracketry and radiator will be where the plastic currently sits. SPAL push fan (5.2") will provide the airflow when not in motion, and a cheap temperature controlled relay will handle the fan and provide an output to a small gauge/datalogger.

I have looked at SO MANY radiators in the past 24 hours. Think full page of google images. Multiply by 18, then search for dimensions. I have settled on a Gen1 KLR650 radiator. The fill port should be under the seat, but I will have to bend the ports to point the opposite direction from where they are. The original rad is known for letting the bike overheat... on a 650cc engine. If it runs too hot, I can look back into dual rads under the fairings, or just get a Gen2 rad which is thiccer. 23226

Riding season just ended here, so I will not be rushing this project. I cannot find any evidence of anybody swapping this engine into a KP Mini (or any bike, from cursory searches). This is unknown territory. I have only my wits, the mechanical skills of a drunk simian, and my smooth, smooth brain.

So far I have ordered:
- Engine
- Bike (thanks to Pete @ Kronik, who it would appear threw in free rearsets)
- MT-15 rear shock. It is a bit short, but thanks to a gentleman on facebook, it can easily be raised by altering the wishbone geometry in the suspension linkage. Drill 2 new holes, attach T-piece there instead of lower original hole, gain 2 inches and preload reduction.

I need to get the bike into the kitchen (where it's warm and good things are made) as opposed to a 3rd party garage where it currently resides ("Hey fam, can I use up about 2 foot by 5 foot in your garage?").

Engine comes with a PWK34. Will have to see what the quality looks like, might spring for a Nibbi or faux-Sudco.

Gearing is similar to the stock KP. One exception is first. The 300CC has a 2.56 ratio as opposed to the 3.077 in the lifan. Primary reduction is also reduced on the 300cc at 3.09 compared to the Lifan's 3.35. Final drive is .938 Lifan and .857 on the Zong. Shouldn't be gearing limited at all.

It looks like many of the top end parts are related to the NC250. This makes me wonder if some of the performance parts from one of the guys around here would work. I am trying to find specifically a worked head or high comp piston (though at 10.5-10.7 it is already a tad elevated). My goal is to break the 30hp barrier in a Lifan KP Mini. Is it smart? No. Can I recommend it? Not yet.

But it should be a fun challenge.

PhildoScaggins

11-02-2020, 07:56 AM

We are watching you with great interest...

franque

11-02-2020, 08:09 AM

Yeah, what he said... Is that motor basically a liquid-cooled CB bottom end with an NC250 top end? If so, that's really cool!

deadwood83

11-02-2020, 03:48 PM

Yeah, what he said... Is that motor basically a liquid-cooled CB bottom end with an NC250 top end? If so, that's really cool!

I would love for this to be totally top-end compatible (NC250 with a different cam grind) but the Tianda diagram 'Chinese Name' column shows CBS300 in the name for the head and cylinder. Their parts diagrams tend to be really good about calling out NC250 parts by having NC250 in the Chinese name.

In my dreams, an 82mm or 84mm big bore kit from an NC250 would fit; making a 343/360CC total displacement engine. Why? Because stupid. The one BIG issue I can foresee which would break compatibility is increased stroke. The NC260 is about 56mm while this has a 65mm stroke. If the kits that nzbrakelathes sells can easily accommodate the extra stroke though?.... Maybe I should figure out where to put an auxiliary fuel tank.

Also, I will not post a link to the seller on Ali until I get the engine in my hands, but so far they have been insanely helpful and have had excellent communication. Their engine listing apparently includes a bunch of goodies. They mention the carb being part of the package, but I got a message asking if I would like to wait 4-6 days or get a partial refund because they were out of wiring harnesses. I was not expecting a bike wiring harness to come with it.

franque

11-02-2020, 04:55 PM

Sounds great! Yeah, I didn't think about the stroke differences, you're right about that. Where are you looking at parts diagrams? I would guess that they are quite different, even though the cylinder head looks very similar. My guess is that the case mouth is quite a bit smaller than an NC250.

Also, just out of curiosity, did the 167FMM drop right in? If so, very interesting! Do you have details for the build of your brother's bike?

deadwood83

11-02-2020, 10:16 PM

Maintenance manual: https://tiandaracing.com/wp-content/uploads/2019/06/CBS300-Service-Manual.pdf

Did you say, "Parts?": https://tiandaracing.com/wp-content/uploads/2019/06/20190510-CBS300-Engine-Parts-Catalogue.pdf

Unfortunately, I was just the idea man, part-financier, and free labor (80% of it) on my brother's build. I did not document it well.

Parts were sourced from Plug N Play Performance in Indonesia. Short list:
BRT 70mm forged piston w/ rings and wrist pin

BRT ceramic coated Super head (36/31mm) with BRT valves, springs, OE Honda CRF150F wipers(the head is GORGEOUS)
OE Honda 2006 CRF150F cam carrier
OE Honda CRF150F valve cover
Ceramic coated BRT 70mm cylinder
EBC CR85 clutch fibers
MasterCam (Indonesian brand, not the software) Stage [unsure] racing cam
Nibbi PWK34 carburetor
Misc mods (intake plenum, throttle cable, couple bolts)
Harbor Freight "catch Can"

Initially, the exhaust valve seized within 3 miles. Very (the gentleman in Indonesia) reached out to BRT and did some digging. Found the guide was under-honed, sent a new guide, spring set, seat, retainer, wipers, and 5mm flex hone via DHL free of charge. The man is a legend.

Supposedly BRT is developing a 4v head for the motor, with an ETA of Feb-March (barring any delays). With the big head (2v) the bike already RIPS. If you dump the clutch with any sort of frequency, it will break the cush drive within 1 season. I am a huge fan of the guy and want to support him, but one of my passions is troubleshooting combined with ample stupidity/silliness. I wanted to do something new, novel, and push the limits of Lifan's mini.

PnP/Very/BRT do offer a 300CC kit with the big valve head, but case boring is required. If I am going to drop a brand new engine from a bike, I do not really want to risk not having it. Plus by the time you pay for all the bits, plus machinist time, the new engine is cheaper? (It's not once you factor in time spent researching, radiators, etc).

I cannot say for sure if the new engine drops in, as it is leaving China today. I can say with absolute certainty that the front motor mount bottom half will not line up. The rest of the measurements seem to line up with the Lifan casings. I could not find exact drawings of the 157FMJ, but the 162FMM, 162FMJ, and 157FMI all shared the same lower castings so I felt confident using those drawings as a basis. By trhese drawings, the front lower engine mount from the KP mini will have to be remade. It should not be very difficult since the top hole (on the motor) seems to line up with the top hole (on the mount).

The 174M-3 (CBS300 engine) comes standard with a 520 sprocket, but my searching on TaoBao seems to indicate a 20mm output shaft commonality across the listings. This is the same shaft dimension on the KP Mini from factory (Zs190 is 17mm) and even if my memory is failing me, sprockets are cheap and plentiful.

Limited pics of the other build:
23236
23230
23231

And the seller provided images of the exact items I will receive. As expected, 4-pin DC CDI, 2-pin trigger, bog-standard Honda-style rectifier, and the wire colors even look similar. All connectors appear to be identical style and shape. At worst, the pinouts will need to be redone.

23232

Comes with a PWK34 clone. No idea what jets are inside, but they will likely need to be changed due to living 4500+' asl.

In my searching, I discovered the Kayo EFI setup that comes on the PitsterPro LS250. Since one of my favorite rides involves climbing 6000'+ elevation in 40 minutes, I would be lying if I claimed it was not intriguing. I couldn't find concrete evidence that CHondaChondaChonda ever got the tuning process to 'take.' I'll see how it runs with the PWK before going totally crazy, but my brother said he is very interested in EFI for his Lifan.

You'd think I would be the one clamoring for EFI after bringing a 40-year-old Yamaha Xj back to life but here I am telling him "carb is fine." For being called unreliable, picky headaches those old Hitachi CV carbs sure do seem to work incredibly well after you: ultrasonic every single individual part, match the TB flaps to bores, put in new diaphragms, throttle valve seals, springs, mixture needles, mixture o-rings, fuel delivery o-rings, reassemble, synch, colortune, synch, colortune, synch, colortune, synch, colortune, and then finally adjust idle.

Obligatory XJ pic:
(nvm, forum fails to upload every version of my XJ photo)

Even after riding his souped-up KP Mini; when he played the "pursuit vehicle" on my Xj his feedback was, "you should have warned me the throttle is so sensitive." Man I love old bikes.

Bruces

11-02-2020, 10:44 PM

just wondering how the border will handle that engine ,is it epa approved ?

deadwood83

11-02-2020, 11:48 PM

just wondering how the border will handle that engine ,is it epa approved ?

WOW. That question got me digging HARD.

Okay, near as I can tell, yes, it is EPA certified. Engine debuted in 2018, and from the EPA's 2018 publication, here is a Sanyang 278cc (Marketed as a 300) listing.

Oh, I'm dumb. The Tianda is sold in the USA. Tianda is where I got the idea for this engine from. So yes, it is most certainly EPA approved.

Heck, the Tianda is sold in California. So it;s probably CARB approved too I would guess?

Falkon45

11-03-2020, 08:01 AM

Man... this has me wanting to do a 300 build on my KPM. Hmm.... I wonder if they make parts for the NBf2.

deadwood83

11-06-2020, 02:26 PM

Okay, a couple more hurdles discovered.

The new engine comes with a 3-phase magneto. Luckily, this is likely an 18 or 21 coil unit, so it should have no problem running the LED lights and the fan. UNluckily, Lifan uses a 2-phase rectifier in the mini. To counter that, the new engine comes with CDI and recumfrier. I will have to run at least one wire.
Wiring diagrams for these engines are literally impossible to find. Tianda throws an 8-pole stator in and uses a 5-wire setup since they run carb and 0 lights. Their only concern is running the gauge, and keeping the lipo topped-up. They also do not provide a wiring diagram, just photos. Cross-referencing Russian MX bikes with the engine (Zuum CX300 NC and BSE Z7 and Racer SR-X2) they do not offer any manuals with wiring diagrams. This might get interesting because the lifan 2021 wiring diagram shows two 12V outputs. The second output drives horn, headlights, running lights.

On point two:
23267
Interestingly, the gear indicator shows 1-6. Turns out this engine DOES come in 5 and 6 speed varieties. No idea which one I will receive. Honestly, both are fine. Final drive should be similar either way. 5 speed will be easier to wire. 6 speed will be neater.

The Zong rectifier is 6 wire.

Yellow X3 - AC in
Red - 12 V Out
Green - Ground
Purple - ? Hopefully switched 12v out. This seems to be the most likely scenario since vsense id really ancient. Engine can run on battery while I measure. Alternately, I just hook it up and if I get no running lights then I know it is vsense. Many 6-wire rectifiers available, so that's just a time thing.

I got a final measure on the planned radiator loc. It is 7.25 wide inside to inside. Both KLR gen rads will fit horizontally, but the gen 1 is about 1" shorter so I ordered that.

My MCO.MSO was not in the box, and I have been having trouble reaching Pete @ Kronik. This is odd, since he has always been incredibly responsive in the past. I believe things are probably happening and it's just a matter of poor timing with USPS load.

NC250 cylinders are def not interchangeable. Gaskets are totally different, coolant outlet/inlet is totally different. The head really does look like an NC250 though, and it uses NC250 valves.

deadwood83

11-15-2020, 07:10 PM

Okay, the engine is here.
The MCO is here.
Vin inspection done.
DMV appointment set.
Still making room in my living room / kitchen to perform the surgery.
SV Racing stopped responding to me once I disclosed I was not shopping for parts to replace broken pieces on a Tianda. I guess they are either too busy, or don’t care about projects.
The engines available to order are 5 speed versions. I cannot find any gear ratios for them. The geartrain bears the same casting numbers as CB250 (ZS170MM-2) transmissions. I am having incredible difficulty finding a 6-speed star. Drum, forks, pawl, countershaft/main shaft are all available on Taobao.

The 6 speed is probably not at all necessary. Realistically, it just splits 3rd/4th gear and removes a tenth or two from final drive. However, in my quest to make a minibike viable for both city and highway without spending $7000 on a Grom + CBR swap, this is my punishment.
If anybody has a hook-up on a Hawk 6 speed shift star…. Hit me up. The fourth image on this post looks like EXACTLY what I need. http://www.chinariders.net/showpost.php?p=324612&postcount=196
Maybe a CRF230F shift star would work? The gearset will not, as the primary reduction is wrong. I also need 6th to be an overdrive.

Meanwhile I have a theory. I found another bike this is used in. It is used in the United Motors Commander/Sport/Classic line of bikes. Indian reviews of the bike indicate it gets a bit flat feeling above 70. This is probably due to gearing combined with Zongshen’s choice of cam grind.
Let’s look at the venerable NC250, and SOHC cam geometry. It has a 77mm bore from factory. Examining the heads, we see that the combustion chamber itself is wider.
We know from the Tianda parts fiche that the CBS300 uses the same rocker arms as the NC250, same exhaust gaskets, and the same intake manifold. Cross-examination of scaled photos shows that the NC250 and CBS300 use the same rocker<->cam locations relative to the centerline of the cam, and they use the same valves. Therefore, the only way to bring the valve heads physically closer is to pivot them about the stem head locations. Even pivoted, the valve lift relative to the face of the combustion chamber would remain unchanged if both engines had a similar cam grind.
That said, the cam grinds must be rather different due to Zongshen’s own output charts showing such contrasting curve shapes. So, I am doing an experiment.
Hypothesis: The CBS300, having identical valve are as the NC250 can flow a hypothetical equal amount of air through the engine. The CBS has a torque curve which declines at a more rapid pace at the higher RPM ranges after peaking around 8200 RPM. The primary differences between the two engines base configurations are cams (potentially) and bore x stroke. The CBS, having a longer stroke, exhibits a “block outrunning the head” scenario where piston velocity exceeds airflow capability.
Process:
1.) I have ordered an NC250 cam. I plan to remove the rear bearing, measure the solid rear lobe on the CBS cam, and press a new bearing onto the NC250 camshaft at the rear (left) location to both:
a.) remove frictional losses of a steel bearing round against the head
b.) try out the NC250 camshaft.
By using an extra camshaft, the option of returning to CBS300 is maintained. Furthermore, the cam end on the CBS300 can be examined for bearing swap at the same time. Likely bearing candidates are sealed units with one side removed to maintain oiling system efficacy.
2.) Porting. This is a minibike, not a cruiser. Power should be upper midrange for maximum stupidity. TO this effect, I will work on valve guide humps on the exhaust and intake, without touching the guides themselves, and widen/flatten the floor a little. That’s it. I don’t want to go to 10k, just lift and shift the curve a little to the right.

Fueling: I am torn. FCR35? PWK34? Lectron (SSR 300 flavor b/c similarity)? Kayo/Rojo EFI? I am strongly leaning Lectron after taking a slide-valve carburetor up the canyon. I thought my Yam was bad, but it has CV carbs which masked the problem more than I realized. If I go Lectron or EFI, I will be sacrificing some max power… but climbing 7000 feet in elevation within 35 minutes or less is also something I do every two weeks. FCR35 would hands-down win on the street but has a track record of being a touch picky about large elevation swings. PWK I just listed because the engine came with one, and they’re dead-nuts simple to jet. They get really awful for rapid altitude changes though.

I get plates on Tuesday, and that’s when the bike will move into my kitchen for surgery. It’s really tempting to just rush to cram everything in, but I am trying to take my time and think everything through beforehand. It’s a lot easier to have a plan and wind up scrapping 50% of it than it is to formulate 3-4 plans one after another.
23343

deadwood83

11-19-2020, 12:32 AM

I wish I could re-title this thread because I just thought of a zinger.

Cooking With Gas: A Culinary Cycle Story

NOTE: Engine supplier is yitonmoto on AliBaba. Awesome folks to work with, highly recommended, small selection but they spent DAYS with me in chat before I bought a single item, and then continued to deliver impeccable service. I would go drinking with these guys. They're awesome.

The bike is titled, plated, in my home; and I have begun disassembly and analysis.

The radiator... is going to be tight.I used Amazon boxes to protect my hardwood and tiles floor. Yes I use a kegerator, yes I homebrew, and yes, one of the brews got really high in alc content.
23428
23429

I Don;t know why Lifan got rid of the color on the side. It looks really nice. I also have some ideas for somewhat inappropriate stickers. With a new engine, it is a new model of bike, and that gives me opportunity to name it.
23430

Paint quality has declined in the 2021 model year. Sadly it is weak, and eager to rub off. I may vinyl the whole thing (with the current scheme plus some extras)

I ordered an FCR28 as found on the Honda CRF250. I am researching if the 2009 CRF450 upgraded water pump (Boyesen) will fit. I also found the original supplier of the speedo. This one is useless. When going 35 it shows 43-44mph. Seems dangerous.

Oh.... and the six-speed transmission is on the way. Cost $57 + $74 shipping to USA for main and countershaft (loaded with cogs) + drum + forks + retainer + retainer spring.

Still going forward with my NC250 cam swap. Zongshen marketed the CBS300 as a budget alternative to the NC250. In marketing, it is bad when your "budget" engine outperforms your "flagship" thumper. I am more convinced than ever that they detuned this .

There is a local shop who offered to do milk headwork but no flow bench. The internal routing of the head and layout just SCREAMS, "I'm a TRX300EX derivative!" so I will be using proven TRX porting as reference.

I THINK properly tuned FCR, Cam, and Headwork will get me crazy close to 30BHP. Tianda measures 27.5 with a free flowing exhaust and Nibbi PWK34 (compared to the 25.4 Zongshen publishes) so these other bits should get me there.

In stock form, this new engine has more power than I know what to do with. The real challenges will be gearing and carb tuning. FCR guys say that going PWK or CV -> FCR gives about an extra 10% in the top end which at 25hp would be 2.5. Add half that to the 27.5 Tianda got and we're already at 28.25. Hoping that porting with FCR adds that missing 1.5 to hit my goal.

Otherwise this will be my only vehicle with less than 100bhp/L. (2003 M3, 1982 Yamaha 650, 2019 Hyundai Veloster Turbo). At least the Yam and BMW do it honestly (no blower).

More to come as I progress.

Falkon45

11-19-2020, 07:39 AM

That model name... baaahahahahahaha!!!!

Also, KP-MOTO club has an exhaust for the KP-mini. So, you can factor that in also, unless you plan on building your own.

culcune

11-19-2020, 07:53 AM

I am guessing paint quality was down due to reopening after COVID shutdown. I wouldn't be surprised at other quality control issues you might run into.

deadwood83

11-19-2020, 09:59 AM

That model name... baaahahahahahaha!!!!

Also, KP-MOTO club has an exhaust for the KP-mini. So, you can factor that in also, unless you plan on building your own.

ROger that, but sold out nationwide with next expected delivery sometime around January. I neither weld, nor plan to learn welding in my kitchen so the KP-Moto CLub exhaust is what I am looking at. There are other possible alternatives (Kymco Spade MMNTHBX exhaust, Yamaha YZF-R15 exhaust, Kymco A.I.R.150 Exhaust) which all look like they would work but only the Kymco/Lifan offerings are of an appropriate diameter (at least 33MM I.D at the head) and the MNNTHBX and Taiwanese offerings are really expensive.

I am guessing paint quality was down due to reopening after COVID shutdown. I wouldn't be surprised at other quality control issues you might run into.

Hoping the paint is all I encounter. So far, other aspects look OK. It;s a shame about the paint because on closer inspection it is actually a pearl white, but you can't see the pearl because of the micro scratching all over the clear which came with the bike.

Maybe 3m Ghost Pearl, but the gloss instead of satin. Satin is a pain to take care of. I wish Lifan sold the decals so I could just redo it.
23431

Ugh. Just realized I forgot to drain the carb before bringing it in.

deadwood83

11-19-2020, 10:23 PM

KLR rad is a no-go. I had my brother measure his bike before I ordered it. I am beginning to seriously doubt his measuring skills.

Exploring other possible options for a rear fit. The pusher fan (SPAL) is the perfect size but that KLR boi is too thicc (width).
Can see it is about 1/2 inch too wide (after adjusting for my lousy positioning for the photo)
23442

In all likelihood, I will move the horn and put some other rad up front like so:
23444

I will not be able to use my fancy pusher fan, and will have to locate a fancy puller fan. The head also needs porting. There's something to be said for velocity vs area, but this is pushing the smaller ports argument too far.
23445

I was going to have the frame totally stripped today and the new (rear) suspension installed but no. WHoever at Lifan had the 15mm ugga-dugga decided they were in a bad mood. I can literally lift the bike with the socket and not have the nuts break loose. My brother has a battery powered milwaukee that I will be borrowing. Throw in a little bit of my own Knipex and it's a party. We'll see how my cat continues to react to foul language.

deadwood83

11-21-2020, 11:25 PM

Progress can be summed up in one picture. I introduce:
Kp-Mini 300 BUCA-K.
23460

More to come as I progress. The secret to breaking those nuts was a 200lb man and a 3'+ breaker bar.....

grumpyunk

11-22-2020, 11:39 AM

Look at it this way.. you now know a 15mm can handle 200X3 ft-lb of tightening torque. That is if you were standing on the breaker bar..
I came in late to the build, but am making the assumption you are replacing the powerplant with a water cooled version that has more displacement. And you are changing the gears inside. Speaking of, did you find that they fit or are they still "Turned over to airline" or "Left shipping center"? A whole lot of 'magic names' to cover the fact that they are on a ricksha somewhere between Shenzen and HK. Maybe.
tom

Added
Just brain fading, but it they had put the inlet/outlet of the radiator at opposite corners, you could have used 'thermal convection' to cause the coolant flow. Colder coolant will flow downward, and into the engine, hotter upward and out to the upper inlet. The design you have depends 100% on pump flow to cause coolant movement. If they had placed the fittings at opposite diagonal corners you could have had thermal assist at least. FWIW. You may not need a fan, unless you drive slowly or in heavy traffic. Is there provision on the engine/head for a coolant temperature sensor?
tom

deadwood83

11-22-2020, 02:41 PM

Look at it this way.. you now know a 15mm can handle 200X3 ft-lb of tightening torque. That is if you were standing on the breaker bar..

I wasn't standing on it, but rather sat on the bike and used it like a kickstarter. I do not b elieve they can really take that much torque. I suspect if I set up a jig and examined roundness, stretch would rear its ugly head. I will be using new fasteners for safety sake.

I came in late to the build, but am making the assumption you are replacing the powerplant with a water cooled version that has more displacement. And you are changing the gears inside. Speaking of, did you find that they fit or are they still "Turned over to airline" or "Left shipping center"? A whole lot of 'magic names' to cover the fact that they are on a ricksha somewhere between Shenzen and HK. Maybe.
tom

Sort of! It's hard to call this engine a liquid cooled version since the only common parts are the mounting bolts, drain plug, and sprocket.

In short:
Displacement: 149cc -> 278.5cc
Valvetrain: SOHC-2 -> SOHC-4
Cooling: Air -> liquid
Compression Ratio: 9.1:1 -> 11.1:1
Trans: 5-speed -> 6-speed (engine is offered in both 5 and 6 speed, but only 5 speed is avail on alibaba)
Bore X Stroke: 57.3X57.8 -> 74X65
Highest gear ratio: .938 -> .854
Power: 12.8hp -> ~27.5-30 (after head, exhaust, cam)

For the parts from taobao, I used an agent (Sugargoo). They are coming via DHL express. Label tracking does show that it has not been received yet, but reddit posts and youtube videos do show people receiving their goods from the agent. They also took photos of all parts before repacking.

Added
Just brain fading, but it they had put the inlet/outlet of the radiator at opposite corners, you could have used 'thermal convection' to cause the coolant flow. Colder coolant will flow downward, and into the engine, hotter upward and out to the upper inlet. The design you have depends 100% on pump flow to cause coolant movement. If they had placed the fittings at opposite diagonal corners you could have had thermal assist at least. FWIW. You may not need a fan, unless you drive slowly or in heavy traffic. Is there provision on the engine/head for a coolant temperature sensor?
tom

The new planned rad does have in/out at opposite corners, but will still rely on pumping since the optimal outlet position is higher than inlet for bleeding and pressure purposes. No good way to fit a t-stat on this motor nor this frame. Too small. No provision for a temp sensor on rad nor engine. If I get really antsy about it, I can put one inline. I'll do initial testing with a 185F fan switch and gauge things by how much the fan comes on.

23462
23463

This will be largely an around-town bike, so traffic lights will be a thing. I will also be doing some canyon runs, and I would prefer having a rad fan rather than dumping from the overflow bottle :D

___________________________

In fact, (due to space constraints) I had to abandon my FCR idea. Held the carb to the frame.... it is taller than the top of the starter to the upper frame member. Looks like PWK or EFI will be the way forward.

23464

To fit the engine, I did have to bend the coil mounts (red). I do not feel overly comfortable with the coil pretty much touching the fuel tank (which is grounded to the frame). GSXR coils are a coil on plug design and have a primary resistance of 1.1-1.4-Ohm. They are also found very cheaply on ebay and GSXR riders don't seem to swap them due to failures; just for mods. This may be a possible way forward.

I'll do some highly scienticious clearance testing with the front cowling and some TP/PT tubes.

I will have to shorten the petcock tube. It is literally just a threaded pipe, so figuring out the pitch and doing the work should not be overly complicated. Currently the petcock outlet would be below the carb fuel inlet, and on the wrong side (blue).

grumpyunk

11-23-2020, 09:27 AM

If you don't go COP, moving the coils is somewhat a simple thing, requiring only longer primary and/or secondary wires.
I don't understand what difference purging air from the radiator would make choosing the inlet or outlet side. If you were to do a cross mount rather than having the tubes oriented vertically, the coolant pump would require a bit more flow than the other way.
EFI is available on aliexpress and likely alibaba and the vendors seem to understand their product and what it takes to make it work effectively. I think I have read some posts about installing, tuning and use that were positive. Once you get them set, they sure can be a lot less 'fussy' than carburetors.
Just off the top of my head, why not rotate the radiator 90 clockwise from the image, locating the 'out' relatively near the inlet to the pump(given IN marked in blue is the pump inlet). A longer hose from the hot outlet from the engine to the radiator inlet, but you would get convection movement.
The fan switch fitting could also be a temperature sensor fitting, and you could use a 'tape on' temperature sensor wrapped onto the HOT hose to regulate the fan. I think one can be made to work reliably, and if you really want it permanent, they can be installed 'in-line'.
tom

Falkon45

11-23-2020, 09:38 AM

If you want, see if you can find a radiator for a lifan KPM 200 or kp200. It's two of them, and they're pretty small. Another is the RC 390 radiator. It's tiny in itself. Too small for the ktm, but should be perfect for that.

JerryHawk250

11-24-2020, 10:59 AM

Hmmm! Reading all this is putting all kind of ideas in my head for the X22R. I got all kind of room for a NC250.

deadwood83

11-24-2020, 11:33 AM

Rectumfriers! Arrgh!

One thing I have wondered is how the unit shipped with the new engine will wire in to the current harness. The new engine uses a ~220W 12-pole stator pushing 3-phase AC. Lifan equipped an 8-pole ~60-80 watt stator from factory. SPAL rates their 5.2” fans at about 4A (48W); Headlight is 15W max; tails are what, 10W? Blinkers, maybe 3W each for a total of 12W. So… my draw should be about 100-150W under the worst possible conditions. That leaves me ~75W to play with should I elect to go EFI. Assuming a 4A fuel pump, there should be no problem.

But that wiring.

New Reg/Rect:
Black = Ignition/VSense
Red = DC+
Green = GND
Yel X3 – AC 3 phase
23482

Current setup:
Yel/Pnk– A/C 2-phase
Red – DC+
Green - GND
Red/White – DC+

So, the Red/White powers rear markers, headlight, and license plate light. This means the Lifan rectifier is just that, a rectifier only. It has no regulator capabilities. The new rectifier/regulator (as diagrammed above and verified with my Fluke multimeter) requires an input containing current battery voltage and control clamping on S1-S3 so as not to overcharge the battery.

There are two ways forward:
1. Try to find a new regulator/rectifier that takes 3-phase AC and outputs properly with double DC out, risk overcharging the battery, and leaking acid.
2. Add a splice to feed ignition voltage to the regulator tap in addition to powering the lights. Looking at the ignition contacts, they are Sumitomo 2.8mm style on maybe 18ga wiring (20ga worst case). For chassis wiring, that has a 12V rating of around 11A for copper. MORE than enough.Let's splice!
23483

The other wiring work required related to connectors. The new engine came with Furukawa RFW sealed connectors. These ARE superior to the Sumitomo MTW style that Lifan ships with. However, I was faced with the choice of buying expensive connectors and rewiring everything… or just convert engine harness to Sumitomo MTW. I went the Sumitomo route so I wouldn’t have to buy new crimpers.

Hmmm! Reading all this is putting all kind of ideas in my head for the X22R. I got all kind of room for a NC250.

Do it. NC250 + Big bore. I believe you may have the fastest X22R, but it could be faster.

If you don't go COP, moving the coils is somewhat a simple thing, requiring only longer primary and/or secondary wires.

This is true. My primary motivation for COP is to not require welding any more things to the frame since I do not have the tools nor experience to do so.

I don't understand what difference purging air from the radiator would make choosing the inlet or outlet side. If you were to do a cross mount rather than having the tubes oriented vertically, the coolant pump would require a bit more flow than the other way.

This is exactly why. I do not have any specs on the pump, so I want to make its job as easy as possible. My leads on a Boyesen unit died out, because Zong moved one fastener to the other side of the water channel on an 04-08 CRF250R/X pump design so I am locked into their design unless I want to pay $$$$$$$ to have one CNC'ed custom.

EFI is available on aliexpress and likely alibaba and the vendors seem to understand their product and what it takes to make it work effectively. I think I have read some posts about installing, tuning and use that were positive. Once you get them set, they sure can be a lot less 'fussy' than carburetors.

Available? Yes. General consensus online is that you buy the kit, get it running (poorly) then wind up swapping most components over time with Ecotrons or MS. If I go EFI, it will probably be the Kayo unit for the NC250 since they engines are rather similar in operation, and the Kayo seems to respond well to the 82/84mm big bore kits on the NC motors (283CC and 297CC; respectively). If I can make the NC250 cam swap work, the motor would be even more similar in operation (barring gear ratios).

Just off the top of my head, why not rotate the radiator 90 clockwise from the image, locating the 'out' relatively near the inlet to the pump(given IN marked in blue is the pump inlet). A longer hose from the hot outlet from the engine to the radiator inlet, but you would get convection movement.
The fan switch fitting could also be a temperature sensor fitting, and you could use a 'tape on' temperature sensor wrapped onto the HOT hose to regulate the fan. I think one can be made to work reliably, and if you really want it permanent, they can be installed 'in-line'.
tom

Nowhere near enough space. For reference, the distance from the reinforcement plates at the top of the front frame member to the top of the front mount is about 6.5". It would either collide with the bottom of the triple tree, or be the recipient of everything the front tire throws up and bend/break when going over speed bumps.

I am also not convinced convection forces will do anything with the coolants used in an engine. In the PC space, one overclocking group did make an effective convective cooler, but it required filling with either a very low flash point coolant, or having an extremely high pressure. In building heating systems using convective principles, there is typically a steam pump or other high pressure application. I think it might be bad if I let my bike get to the point where it is making a steam pump inside the engine.

If you want, see if you can find a radiator for a lifan KPM 200 or kp200. It's two of them, and they're pretty small. Another is the RC 390 radiator. It's tiny in itself. Too small for the ktm, but should be perfect for that.

KPM and KPR radiators did cross my mind. However, this engine will have 40% more displacement and make ~53% more power. I do not believe the KP radiators would be sufficient. The KTM might! But my CB250 rad is already on order. If it does not work out, I will try and dig around more for RC390 radiator dimensions. Once immediate thought is that if the rad is too narrow, I would be unable to get a pull fan behind it next to the center support while being able to clear coolant lines and exhaust piping. Fans will not fit in front, because the back of the forks have about a 29mm clearance (longitudinally along the chassis) between the front frame member. A pusher fan might fit in the middle between the forks at the front, but then you lose effective area to the frame behind the rad.

It is truly a challenging balancing act, and I am thoroughly enjoying the challenge (even if I do get frustrated at times!)

JerryHawk250

11-24-2020, 11:53 AM

I can get a complete NC250 engine kit with EFI that includes everything i need minus the radiator for under $1500. It's very tempting.
https://gpxmoto.com/image/cache/catalog/ENGINES/250cc/250%20nc%202-1200x1200.jpg

culcune

11-24-2020, 12:37 PM

JerryHawk250

11-24-2020, 12:41 PM

what would you throw it in?

x22r

deadwood83

11-24-2020, 02:29 PM

x22r

Hmmm. This looks a whole lot like CG mounting pattern (like the KP Mini). Not sure if the honking kicker mech and back of the NC250 would fit.

23484

But you could do the CBS swap and we could explore this engine in tandem :hi:

That said, the only real thing you would gain is liquid cooling and 4 valves, with a loss of some Honda parts interchange.

Also, DHL texted me. They said I'm going to have a 6-speed transmission soon.

23485

Falkon45

11-25-2020, 01:37 PM

wheelbender6

11-25-2020, 07:15 PM

grumpyunk

11-25-2020, 07:21 PM

The PC cooling is not a real comparison in that the fluid boils inside as you note, while the coolant in a bike cooling system should not get to the vapor state.
Model T cooling used no pump. It was all done by convection(perhaps the wrong word) where hotter coolant was less dense, colder more dense, so a thermal circulation of liquid coolant occurred.
If you heat the coolant in a radiator the hot coolant will attempt to migrate to the top, the colder coolant to the bottom. That is the effect that would reduce the load on the coolant pump and be a more natural effect. Free circulation...
A cross flow radiator back about 1967 was a big deal, and allowed lower hoods on cars. There was less convective circulation, so there was concern about overheating in lower speed traffic. I cannot verify, but think larger pumps were installed 'just to be sure'. I don't know if I was clear, but I was suggesting mounting the radiator in a 'vertical' orientation, with the tubes going up and down. There is a post about mounting an oil cooler, actually more than one, in a vertical mode on site. I was thinking off-center mounting, FWIW, but I do not have a machine to futz around with attempting of different spots.
tom

Falkon45

11-25-2020, 07:40 PM

Well, the problem with the 390s are there were trying to cut weight. They also said bad machine work, so the heads were uneven, and would vote head gaskets.

But, how the kp radiators are mounted, there's one on each side. That gives you some space, if you have somewhere to put them. The fluid does green the lowest point up to the first radiator, across, and down through the second radiator into the head. Same principal, just split in two. Combined, they should have approximately the same surface area as your radiator. The side closest to the head has the fan mounted. It's also the fill and radiator cap. I haven't looked for the high pressure caps that can fit, but this just have me the idea to look for them.

deadwood83

11-25-2020, 11:53 PM

I mentioned the KP radiators, because it's actually two radiators. With the twin radiator confuguration, you could set them up like the VFR 800 rads were. Small, yes, but I believe those would be sufficient, unless you're running very high compression, and fueling on the lean side. My fan very rarely runs on the KPM. My KTM, on the other hand, was always at, or just below the "hot" line. They ran hot from the factory. The radiator is a decent size. I have mine off the bike. I can try and take some measurements for you. The fan is still attached also, so those will be close enough for you. I never did the FAL upgrade. I just used engine ice to keep overheating at bay.

Innnnnteresting. The shroud makes it look like one big unit. My CB rad is scheduled for delivery on Saturday. We'll see how close that size is and I'll see if Lifan is willing to sell rads to me should it prove to be another dead end. I do worry about exhaust clearance. It is already going to be tight, and setting the rads inboard of the frame's leading edge may prove to make it too tight. We'll see!

The PC cooling is not a real comparison in that the fluid boils inside as you note, while the coolant in a bike cooling system should not get to the vapor state.
Model T cooling used no pump. It was all done by convection(perhaps the wrong word) where hotter coolant was less dense, colder more dense, so a thermal circulation of liquid coolant occurred.
If you heat the coolant in a radiator the hot coolant will attempt to migrate to the top, the colder coolant to the bottom. That is the effect that would reduce the load on the coolant pump and be a more natural effect. Free circulation...
A cross flow radiator back about 1967 was a big deal, and allowed lower hoods on cars. There was less convective circulation, so there was concern about overheating in lower speed traffic. I cannot verify, but think larger pumps were installed 'just to be sure'. I don't know if I was clear, but I was suggesting mounting the radiator in a 'vertical' orientation, with the tubes going up and down. There is a post about mounting an oil cooler, actually more than one, in a vertical mode on site. I was thinking off-center mounting, FWIW, but I do not have a machine to futz around with attempting of different spots.
tom

Okay, I think I follow. Unfortunately the KP is just a wee little thing. Lifan left more or less 0 room behind the cowling so my original thought of vertically oriented KTM XC-W rads went out the window. I don't think it can host a radiator large enough to work without a fan in traffic. I may be underestimating the engine's heat tolerance (My XJ650 seems to do Okay without a rad and all 4 of its cylinders) but I also don't have a ton of radiators laying around. This is my only liquid cooled bike so supply is slim.

23502

What is the HP rating on that NC250?
Should be about 25 hp atock at the crank. There's an RX3 dyno floating around here showing 22.2something wheel.

Speaking of exhausts, I have had a hell of a time looking for one. The KP exhaust 100% will not fit. Not even close. THis engine has more of the ole up-n-down and when I tried to see how close the stock exhaust was.. well it just wasn't.

So I browsed Ali. Lots of photos, and sellers frequently include measurements.

Using SOlidWorks' sketch trace feature and the technical drawing from Zongshen, I got the exhaust port to be 108* off upward centerline. Sadly no top-down drawings so not super sure on the angle out, but would wager 30-45* off the front face. THis lead me to several candidates, with some being rules out quite quickly. All had the "looking up" exhaust port, with some showing the desired bend towards the centerline of the bike at the head to interface with the diagonal-facing exhaust port.

Top down is Yamaha MT-15, CBR150R (2010-2016), and Benelli Leoncino 250.

23505
So here's what's wrong with all of them:
- Yamaha: Too Small. 28mm inner diameter. May as well cut off one of the rocker legs and only run on one exhaust valve.

- Honda CBR: Angles are too shallow, and at minimum would need to be cut and re-welded to turn the header towards the exhaust port.

- Benelli: That bend in the head pipe causes it to cross the centerline of the Benelli motor. I can;t have that because that's where one of my engine mounts lives. Benelli uses a front mount, but the exhaust goes forward of it. That's where the radiator will live.

But then, like a beam of light from the heavens... CFMoto 250NK exhaust. From photos, it looked PERFECT. SO, I went to the fleabay and found a seller. Fleabay sellers are WAY better about responding to messages, even if its the same seller as on Ali.

23506

So I messaged:
"Hello friend, I am but one sad man building a happy motorcycle. Can you please share A and B sizes?"

With my heart soaring, thinking I had discovered the key, the perfect factory made system that would bolt on with only minor cutting and grinding; I awaited the reply with baited breath. I fought, and distracted myself by watching Long Way Up, hardly able to resist impulse-slamming that big blue "Buy it now" button. Then, not one hour ago...

23507

I want Gina to be wrong. No, I NEED Gina to be wrong. Gina... why? </3

I felt let down. I felt hopeless. I felt helpless.
But then I recalled that I have a very powerful set of tools: several 3d printers and SolidWorks. I could prototype a tool, made to alarming accuracy, to sand down the exhaust flange. So sketching just happened. Something I can mount on a drill, with sandpaper on the inside, which can evenly reduce an outer diameter....
23509

I forgot I can make my own tools. Eureka! I ordered an NK250 exhaust.:thanks:

Oh, and my crimpers were too small for 16ga wiring. Some slightly larger ones showed up today. Now I can complete the wiring.

deadwood83

11-26-2020, 12:10 AM

CAD is done. Just need to print my new tool and get an 8mm bolt and nut.
23510

JerryHawk250

11-26-2020, 08:37 AM

What is the HP rating on that NC250? GPX Moto's site say 32hp but specs. say 19kw which is just over 25hp

grumpyunk

11-27-2020, 11:14 AM

deadwood83

11-27-2020, 12:55 PM

Assuming you want to reduce the diameter A down to 42mm give or take... What difference (within reason) to you need to worry about being circular? IOW, the exhaust seals on the edge of the pipe, not the diameter. There can be irregularity in the diameter that will have no effect as long as the retainer can clamp tightly to hold the pipe against the sealing ring.
I do not know how trying to fit a 42mm ID tool onto a 45mm diameter pipe is going to work even if said tool is spinning at 1500 rpm. It will just bump into the 'end' of the 45mm pipe. I think I would have tried using a belt sander, vertical, with a table, such that I could hold the pipe against the moving belt, rotating as evenly as I could, to shear off some of the excess diameter. If you get a small table, you can work in 'arcs' of sanding, doing one arc, shifting the pipe, and then another 'arc'. Resting the pipe on the table and rotating should allow good control of the contact area without having the pipe moving around too much. You could get 'pretty close' and then use a flap disk or cutting wheel to shave off the resulting 'bump' areas. I think. Again, dunno how 42 is going to fit over 45 without some prodding???
tom

This is the beauty of 3d printing material selection. I am going to print in PET-G, and the 'leaves' of the tool will have some flex. For reference, it;s the same base material as 1gal bottles of water, just thicker and with an internal geometry to aid in rigidity (but no too much!
23530
I also designed a tapered bell crown into it. This section view might help explain more. The sandpaper is looped inside then extends back out between the leaves so that it is held in place but can still be removed/replaced easily. Yes, the pipe does seal against the face via gasket, but I am going to try and make it seal on a stepped face which is shown here in a rough approximation of the port using real dimensions.
23531

Electrical harness is more or less complete. Just deciding where I actually want to tap ignition voltage and return to VSense, and then I really need some sort of documentation to see if the display supports showing a 6 in the gear indicator field. Sadly, the manufacturer has not been very responsive. I am hesitant to re-wrap the loom until I know if I need to run another wire for the gear position sensor (would need to know the pin on the header)

grumpyunk

11-28-2020, 11:20 AM

deadwood83

11-28-2020, 10:04 PM

Is the engine used OEM in bikes that are available in US? That might be a source of a wiring diagram that would show the switch encoding. Unless it is a plain rotary, which it could be, needing that many conductors, it would be encoded such that in a specific gear, multiple conductors would be grounded(I think it is ground...), which would make the display show the proper gear.
I guess I'm not sure if it is the display on the cluster or the need for X number of conductors that is the exact question. But that's ok as you can look at it so given the display is NOT showing a gear position, you must be in 6th... no?
tom

Probably going to wind up with a GPS from a Suzuki Dr250. IT should be center ground and then the same sequential connection found in the Lifan box. THe freedom of doing my own wiring is that if it doesn't match, then I can just swap pins at the connector. With the alternator finished re-terminating I still need to throw that back in the crank cover. I'll worry about the position sensor when I am at the taking engine apart stage.

Speaking of taking the engine apart, I have NO IDEA how thick the castings are at any point in the head. A Spare head was $51 USD so I threw that in my alibaba agent order. The transmission is sitting here in DHL packaging waiting for when I am at the engine dismantling part. Trying to stage this out in a semi logical fashion.

Radiator, frame, prelim exhaust while the new engine is loosely placed in the frame (to check clearances) then remove engine again.
Rear suspension after removing engine so I can use my scissor jack at the frame point and minimize load while maximizing working space. Re-install rear suspension with new bits.
Front suspension where scissor jack can support at lower frame tube, and drilling for single rivnut at lowest point of radiator mount while front suspension is off and no engine present to minimize stress on front frame member.
Radiator bracketry after first (lower rad mount) rivnut installed while bike is on rear stand with nothing else mounted (except front wheel b/c hitting your knee on the jack is really painful. Re-cut petcock tube and re-thread at the same time.
Engine work
Re-mount engine
Plumb cooling and intake.
Fabricate/alter throttle cable and figure out clutch cable replacement model / new bracket.
Mount exhaust
Wait until it's bearable to ride (>45F)

And speaking of radiators, dare I say that the CB250 radiator is almost a perfect fit? It is within 1/4" of the lower width of the fairings, and has a tab (meant for the fan) which can get an enlarged hole dead center at the bottom for a grommet.

Ho-ly smokes. It's almost like obsessive measurements and everything paid off. But now I get to figure out how I want to make the upper bracketry. It is tempting to just do two L-brackets, but I would prefer to avoid excessive drilling/riveting on the front frame tube.

Frame in blue mounts in red. The T section looks and feels like the frame tubes aren't properly notched to each other and they just used a box reinforcement with tons of weld to glue it all together. Might be strong, but sure seems like it might also be a pain to drill/tap/rivnut there.

23545

deadwood83

01-11-2021, 04:22 PM

Alright, back at it.

I basically took the month of December more or less off because i just didn't feel like working on it. I hit a wall on top rad mount I would be happy with and I let it get under my skin and stall everything else.

The 2nd head came in, so I have a practice/final. The motor mount also came in so I can cut the Lifan and the Zong mounts and get them welded together.

SanSan also delivered the non-proprietary dash cluster. Glad I didn't put the wiring harness back on because it uses a completely different connector set.. and yes, it supports displaying 6 gears!

TO get myself off dead center on rad mounts, I created a rough cad so I can mentally visualize while considering options.
23746
I also started porting the head because it seemed more straightforward than sitting and beating myself up over the rad mount. The ports on the CBS head aren't tragically bad or anything, but it becomes evident why the power peaks at 9k looking at shape and overall size. Doubly so when your overall size is significantly reduced by the super nasty seat protrusion.

I'm keeping the porting really simple and minimal. Just flattening the floor a skoosh, removing casting flaws, port-matching, and matching seats to the port chambers.
In this photo, one side is about 50-60% done and the other is untouched (intake).

23743
There's not a ton of material on the walls in contact with the water jacket so material removal is rather limited outside of the center divider, which is solid. Valve guides really don't intrude much, so I'll be leaving those alone.

Once I tear down the engine for trans swap i'll finish valve work by lapping them to the head. Looking at the brand new sealing surfaces on the NIB head, I'm not sure if I trust them to be amazing. Probably just a bit of edge softening to fight knock as well.
23744

Also, Gary @ Pitster/GPX said he's just waiting on some wiring harnesses and then I'll have an NC250 EFI setup. Since this engine shares a whoooole lot of similarities where the fuel meats the air and spark, I think it will work rather well.

Last image, just an assortment of the tools I have designed so far to help with bushing thickness reducing, diameter reducing with a hand drill, locating tools for drilling rivnut holes on tubing, etc. I like these because it lets me use design abilities to compensate for lack of experience and training in precisely locating holes, etc by hand.

23745

deadwood83

01-16-2021, 04:25 PM

Porting and polishing is more or less done. Some small details to clean up manually, but I'm done with the power tools.

The chamber was a bit rough for my liking, and had some sharp ridges that just screamed pre-detonation, so it got polished. Too much porosity to fully eliminate, and it was crazy difficult to preserve the contours while polishing, but bristle disks and carefully cut soft sanding discs worked out okayish. I also did minor deshrouding on hard machined edges by the valve seats. From the Tianda engine manual, you can see from a used engine the air/fuel mix enters from a very narrow area (it removes carbon deposits to an extent)
23762
I focused mostly on the sides from centerline to the cleaner parts.

23759
To compensate for the polishing, I will have the head decked by ~6-10 thousandths. Since the piston is PVD coated I don't want to have to adjust the valve reliefs; hence the conservative number.

Exhaust bowls and runners got a polish after rough matching to valve seats. The shorter side is ever so slightly widened compared to the long side so that velocities will be more equalized.

23760

Intake was port matched. The included manifold has the same ID as the EFI manifold. I used a round-nose conical burr and gently ground until it witnessed on the manifold, then blended. I may go in with some 120 taped to my pinky just to clean things up. Intake was left alone after carbide, no sanding.
23761

Spark plug hole was deburred with a straight wire brush. It was dead slow, but worked well enough.

I feel my work is far from professional, but I am fairly satisfied with the outcome considering it;s the first time I've ported anything and my dremel has seen far better days. It 100% needs a new armature and new brushes.

Oh, and the valve seats are specced at 1.5mm in the manual, but they all measure 1.2mm. Neway pilot ant 45* cutter en route to fix those.

grumpyunk

01-17-2021, 10:12 AM

deadwood83

01-17-2021, 01:08 PM

Given how close the intake valve seats are to the edge of the chamber, cutting material off the head to bump the CR may run into problems. There is very little material left at the edge of the intake valve, and the seat may get cut by the milling machine.
If you look at the Tianda example there is some distance from the edge of the valve seats to the flat of the cylinder head where the head gasket(fire ring?) will be compressed. I just would be very careful as there is a lot less margin than in the image you showed as your example/guide.
I am NOT any sort of expert. But, I would take a look at the edge formed on the head where the bore for the valve seat was made. That slight corner, actually an arc of it, could be rounded, right where the Tianda picture seems to show intake flow cleaning off any carbon deposit. If the flow is from the center of the intake towards the exhaust valve, there is a lip that could have the corner take off with just a slight bit of work. I await some more expert comment... Were I doing the grinding, I'd run something along there to take off the corner, even just a portion of the diameter.
tom

Good observation! A sheet of Norton wet/dry automotive 1000 grit sandpaper is about 8.5 thou thick. The edge of the seat (not the edge of the sealing surface, which is deeper) is about 50 thou from the head face.

I'm not too worried because the piston is generously cut
23771

ANd it's hard to see, but there is actually a vertical of about 8 thou off of Zongshen's relief cuts.

23770

Also, look at how terrible their valve surfaces are. Some of it is dirt (I have only cleaned the head with a vacuum) but their cutters are... ehhhhhhhhh.

I do not have a mill or anything of the sort, so any decking would be done my an engine shop. I will of course ask them their professional opinion at the time.

Also, I ordered this head specifically so I still have the head the engine same with in case I messed anything up too bad hahahah.

By the edges that could maybe be taken down a tad, were you talking about these guys?
23772
They got softened a bit during the polish while I tried to keep it more defined between the valves. My thinking is that the Tianda manual

(same engine BTW, so margins are the same, just the carbon ring shows what the actual bore/piston area is compared to this clean head)
shows where the flow is going as stock. The geometry (as stock) is biasing it towards the center meaning there is greater resistance to flow moving away from center. I softened the edges to each side of where the Tianda image shows the flow with the intended goal of promoting charge to enter across a greater section of the valve. (this is just my seat of the pants thinking though, and I could be completely backwards!. ... which I think is part of the fun.)

More than that though, I think the greatest benefit will be from matching the seats to the bowls. The fit from factory was.... uninspiring.

23773

And if I messed it all up.. I still have another head.... Though I should probably fix my dremel before attacking that one.

grumpyunk

01-18-2021, 10:45 AM

The circled (in red) areas are exactly what I was considering. The 'clean' area is a bit more towards the center of the chamber, but... removing a bit of that ridge would increase the flow along a wider swath, I think. Even just taking the sharp edge off a bit more would allow better flow. I do not think you want turbulence there, inside, a tumble, yes, but more flow is a goal in that area. I think. Again, not a 'port person' by any stretch. Though, am considering getting a second head for a CG250 for 'experimental purposes'. Right now, too many other things taking up my time...
I agree that cleaning up the fit of the pressed in valve seats should make things flow better. That alone might be the best investment of time/material. Were I doing a cleanup, I would tackle that first as it to me is 'best bang for the buck', and I would be a lot less worried about cutting away too much material and weakening 'things' in that area. It should stand a good grind with no problems, and give good results. I have watched a vid about work on some 302 heads.. and I think the guy spent a lot of time on some stuff, cutting away around the valve seat, sort of near the cylinder wall when the head would be installed, and to my mind, little gain. Best work was port matching and cleaning up the seat 'fit' to the port wall. (removing the 'ring' right behind the seat)
The second pic in post #40, intakes on the right, you could round off the 'ridge' on both intake & exhaust. Not remove much, just make the ridge be a 'not ridge'. Actually, it is not a ridge, just a more sharply defined corner.
Again with the youtube, there's a guy name Paul who 'fixes' messed up heads. He does machining. Went .. and looked. Paul X

See here:
https://www.youtube.com/watch?v=dF2j3rKT8wo

see how he cleans the chamber and around the valves. I again repeat, I know NOTHING, but what I saw demonstrated. He seems to know his 'stuff'. If you watch his other work, about cams and cam bearings, I go WOW, he's a true artist. IMO
Another is myvintageiron7512, the guy who did the head cutting to relieve around the valves and allow more area to flow.
Here's the 1st of a 5-part:
https://www.youtube.com/watch?v=ckQJHVABRy0

I watched them all. Made me want to 'do' a 302 with some of those heads.
tom

deadwood83

01-23-2021, 11:47 PM

Neway cutter came in.

WOW.

Try to guess which seat found a Neway.
23830

I also found a bunch of rear suspension and frame issues that lifan willfully ignored or tried to cheap around. I don't feel like documenting those today, I'll stick with my seats.

deadwood83

01-24-2021, 12:35 AM

Impeccable. (top left is a piece of swarf from cutting I hadn't blown away yet).
23831

Now just deck and swap heads while I rebuild the trans. Oh, and camshaft. But I can't compare angles and lift until I have both in hand.

Falkon45

01-25-2021, 07:29 AM

I'm loving the outcome of this!!!

deadwood83

01-25-2021, 08:50 AM

This is the question we all want to know.

Until now, nobody was sure. All we knew was:

CRF150/230-based engine.
Kinda looks like a Z125 if you squint really hard and take off the tailpiece
Embraces the recent 'My motorcycle is an insect' trend.

But what is that frame? What is it? Why do there seem to be no drop-in shocks? What do I do if I need stem bearings? Does every mod to this bike need to be custom made? These are the important questions.

I started by looking at the suspension. I think the tail end is too low. I would say I range from a bear of a person to a whale of a person depending on how recently I have had a haircut. So a rear lift is in order to fight my natural booty sag.

A gentleman by the name of Mike Davis posted some months back in the KP Mini group about drilling out the rear linkages to raise the rear end, but the more I looked at it, the more it seemed that shortening the lower linkage just broke the geometry of the bike. Sure, it effectively lifted the rear and increased the preload, but it seemed like a band-aid for a poor design.
23837
THis was reinforced when I took some snapshots of the bottom of the suspension geometry, the did some CAD after scanning and measuring the cush linkage.

Why is the suspension alignment in the rear so wonky? Why is lifan taking a simple triangular part and translating a simple force vector in the ZY plane cross multiple ZY planes in the XYZ coordinate space? All the Japanese and European monoshock bikes with symmetrical swingarms are keeping their linkages symmetrical, and even the 150cc bikes from the big four have a full motor cage/cradle. Why does the rear suspension linkage look so half-baked? The swingarm pivot is centered, the frame mount is centered, so why is everything else looking so skewed and why did they cast THREE different offsets into the cush pivot? What is this?
23838

Enter; The Googles.
Oh I see. I see now. It's an XR100/CRF100 hybrid with a chopped XR200 front frame tube. We must go back to basics in order to understand why they did this.

23839

Castings/molds are expensive
R&D is expensive
People are expensive.

So if I were Lifan, how would I cut costs and maximize profit to keep the Chinese government happy and push the rhetoric of 5 year economic plans?

Build to a standard that is cheap, and is strong enough to outcompete other Chinese minis while being weak enough to keep customers replacing bikes.

Copy a frame that is known to bend long-term with abusive riding, but put in a small enough powerplant to prevent this from being obvious
Remove the lower frame cradle from the XR200 front setup copied off of older bikes to keep assembly cheap. If a machine/lift can place the engine, then workers need to be less fit, less skilled, and less well-insured while minimizing workplace injury risks.
Old Japanese tooling can be copied and design examples are plentiful. R&D is cheaper if we don;t have to do much of it at all.

Minimize costs by reusing parts that exist. This was reinforced in my first engineering design class. "If somebody already makes it, do not reinvent it, that makes you less valuable since your efforts hurt the bottom line."
Cut corners where they won't impact the overall perception of quality, and won't be noticed as a major failing.

This is the rear linkage in a nutshell. Look at the photo above of the Honda XR80/XR100/CRF100 linkage. It is a cast piece (expensive). It was cast so that Honda could mold in the offset of one of the legs in a repeatable fashion rather than build out a whole series of tooling to press/forge a single bend. This means that for an initial expense (molds and process) they can reduce the number of rejected parts due to metal fatigue and machine wear on the bending/forging of the part.

Lifan decided that (from what I can tell):

We will already be casting one piece (the cush linkage)
We can add offsets to the linkage to compensate for offsets in the previously cast piece.
Assembly costs and failure rates can be cut down by having two loose hoop bars over a steel tube which is welded together in a cheap jig.
By utilizing sliding through-bearings in the cush linkage, assembly precision will be less critical because slop will be designed in, and opposite tolerance stacking will take care of the rest.
Cut steel tubing/plate/cast parts oversize so as to avoid using custom length bearings and eliminate the cost and time of assembling bearings with wipers and endcaps.

Only body parts, seat, rear brake lever, swingarm (sort of? just mount the grom pivot on the bottom) are unique to the KP mini. Tail (subframe) is altered from the XR series but I am fully confident if I were to dig deep enough into the XR lineup of the mid eighties, I could find a matching subframe that Lifan copied.

Lifan is not stupid, and their engineers are not stupid. They achieved their goals, but the end product is not quite as premium as the initial inspection or curbside-analysis would have you believe. For most people, they would probably not notice or care about this. I wanted to tweak the suspension, and I wanted to do it right. I wanted to prevent my frame from bending at the headstock.

Implications of the frame basis discovery:

The XR200 frame accepts the same motor layout and mounting as employed by Lifan. Since the frame is mostly built around the XR100/XR200, almost no engineering work is required. People have even swapped lifan 200CC engines into the XR200. https://www.youtube.com/watch?v=zP4NTfQUcFc&fbclid=IwAR3rXJnrwL-fsF0pEHcOpef69cVIytC2bHEc8H-moGaJrk129jbo1IcrxP0
The suspension setup is nearly 40 years old.
Aftermarket shock options without wonky workarounds! https://tboltusa.com/.../tbparts-dnm-rear-shock-for-honda... (https://tboltusa.com/store/tbparts-dnm-rear-shock-for-honda-crf100-xr100-650lbs-p-10973.html?fbclid=IwAR39MXbcHi9wBSM9YEScbj-_d5ccMBFqlTmqNX65E2n6-3Z1UDYvs4KGXZg) // https://www.vonkat.us/shop/100x-shock (https://www.vonkat.us/shop/100x-shock?fbclid=IwAR0t7RTaVo-0zeywFPN9I4CQlFvb9AVgd51ikeAkaADWk53F6nkW457VzqQ)
Lifting links! https://rhmotoworx.com/product/rear-links/ (https://rhmotoworx.com/product/rear-links/?fbclid=IwAR0KqUQ9iBSLb4gbLb_0fhUl9yLNMTyy_vbKMhPL S_kW5k7eRSYRmRQCL9Y) (will also require the CRF100/XR80 wishbone/lower linkage, and potentially gentle bending if your lower linkage is as f**cked as mine came from factory). I have a full CRF150 wishbone+"cush link" en route.
If you stunt a lot, expect the frame to either flex or the rear top engine case mount to break.
The geometry of the default KP mini rear suspension is GARBAGE.
Lifan compromised the frame by removing the lower engine cradle. With cradle, old XR200s are still kicking even after 30+ years of jumping, trail riding, tree-crashing.

Here is an image of bottom-up on the KP mini compared to top-down on the XR100. Note the overall suspension geometry is nearly identical, with the Lifan having a wider lower linkage and malformed cush linkage to avoid paying to cast more parts. I pulled all bolts from the rear suspension and compared to parts on CMSNL, and the fastener specs are identical between the CRF100 and the KP-Mini.
23841
https://www.cmsnl.com/honda-crf100f-2004-4-usa_model32126/partslist/F++19.html?fbclid=IwAR3G6R4TguwZKY5029xYWXE-qUYdNrvFu2pNXDz3nWf-VseDocB74HpZMXw#.YAzduhZlDUQ

The fix:

I have ordered a CRF100 rear linkage set. If this drops in (which it should) then I can use one of the pre-made rear linkage lifts to preserve proper suspension geometry (keep some linearity of the rear shock travel in motion).
I have ordered an XR200 frame without paperwork. The price was right, and I can cut out the bottom and have it welded to the KP frame. This will be cheaper than paying somebody to fabricate and weld a while new cradle system. I have neither tube bender, notcher, nor welder.

23840

franque

01-25-2021, 10:35 AM

Are you planning on bracing/gusseting the headstock?

Great info, love the research, it reminds me of the thread I made on here about the clutch covers, though yours is a bit more pertinent ;)

deadwood83

01-29-2021, 12:12 PM

Are you planning on bracing/gusseting the headstock?

Great info, love the research, it reminds me of the thread I made on here about the clutch covers, though yours is a bit more pertinent ;)

Nah, headstock is fine after the full cradle is welded in. YOu then have three stress members connecting the headstock to the 'core' of the frame structure and have better used the engine as another stress member since it is now connected to the complete cradle. If I were putting MX suspension and planning to jump it super high then it would be an issue, but it's a street bike.

Also, RAD MOUNTS ARE GETTING MADE.

Huge thanks to SendCutSend.com. They offer laser cutting AND bending from your DXF files. Their bend calculators also supply K-Factor for your selected material .

Going full cradle was what really made this type of mount easy and possible. With the factory setup, I wanted to avoid drilling at the brace area at all costs. With full cradle, footpeg weight is taken off the engine (moved to cradle) and the cradle translates most of the forces at the crossmember-to-downtube point onto compression forces rather than tension forces. Adding steel rivnuts there now becomes no big deal at all.

Huzzah!
23868
23867

In other news, I realized I am missing a MASSIVELY important bearing!
The 5 speed transmission (which has a kickstart and is currently preinstalled into the engine) uses the standard clotch boss/bearing.gear through a 6006 bearing. (55mm OD, 30mm ID, 13mm thick). The 6 speed transmission omits this boss/gear combo because the extra space from the kickstart is necessary to ensure adequate gear thickness.

23869

There is NO STANDARD BEARING that has a 55mm OD and 20mm ID (the shaft thickness). It also seems like having a mainshaft flopping around all willy-nilly might be bad for engine longevity. SO I put on my Sherlock Holmes hat and cloak.

Tianda uses the 6 speed transmission, so something must exist, whether that is a different engine case, bearing spacer, etc.
The Tianda parts fiche is just a ripoff of the Zongshen 5 speed parts fiche, so it has nothing relevant.
SV Racing was way less than helpful once I revealed that I didn't own a Tianda, and just wanted to buy parts. The owner wholesale ignored me after that revelation.
Attempts to reach Tianda directly in the past also resulted in being ignored.
Zongshen told me in no uncertain terms to "please not contact us for individual requests."

Hmmmm........

Ah! But the CRF230L 6-speed shift star fit. I know their transmission is 100% not compatible, but I also know they do not use a kick starter. Let's see what the engine case looks like.
23870

Um. Okay. That is interesting. That hole looks way less than 30mm. I can't use the retainer because the gear against the bearing on the zongshen is rather large, and would potentially hit the retainer's raised edges. But what size is this? How close is it really?
23870

Oh. And that bearing? I need a part number.
BEARING, RADIAL BALL (20X55X11)
91005-KCN-003 (https://www.hondapartshouse.com/oemparts/p/honda/91005-kcn-003/bearing-radial-ball-20x55x11)

The zongshen bearing sits proud of the cases. The honda bearing may not. I am hoping the tolerances will work themselves out, but if they do not I can always steal 1-2 of the flat washers from the 5 speed gearbox which no longer has a home.

ANOTHER VICTORY! :thanks:

OH SNAP! A surprise and unplanned update! While typing this up, the XR200 frame arrived. Cloneluminati, confirmed. That cradle addition should be cake.

23871

franque

01-30-2021, 06:40 AM

grumpyunk

01-30-2021, 09:06 AM

Previously on Self-Destruction, a six-speed gearset was obtained from a undisclosed location in the Far East. Question is: Was a six-speed shift drum also obtained?
Looking at the gear pairs, side by side, it seems the position of the pairs is somewhat different, necessitating a substantially different drum arrangement.
As I understand the previous, the kick start end of the counter shaft uses that space and results in a 20mm shaft, the 5, a 30mm. You found a gear(Honda?) with the proper ID and OD that just might make your shaft work in the case you have. Does that summarize properly?
OT {The pictures are so large, reading the text is difficult as a L-R scroll is needed. How in heck did you intersperse words with images. All I can do is 'attach image' and it puts them all at the end of the post.}

tom

deadwood83

01-30-2021, 12:14 PM

Great stuff, I'm excited to see how this build is coming together!

While you've got both transmissions out (I'm assuming, perhaps incorrectly, that they're close, if not identical to the CB/CG transmissions?), do you think you could measure the OAL and diameter of both shafts, OAL of the assembled gear sets, and if there's any difference of protrusion for either the main shaft or countershaft on the assembled units?

I'm just curious, as I've got access to some 6 speed Honda transmissions that were for E-start only bikes, and I'm curious as to the differences/similarities. I think that the 6-speed with kick start, à la XR200, is different, as they used wider cases, and post '86 had a 5mm longer countershaft (or else the sprockets were spaced 5mm outward) but as far as I can tell, the cases for the 6-speed E-start models are the same as 5-speed kick only or kick/E-start. I can provide the same info on my end once I have the cases split.

Sure thing. From my understanding of all the reading I did, there were briefly 6-speed kickers which tended to object to the supermoto lifestyle because each individual gear was made thinner to compensate.

Subsequently, we saw the return of the 5 speed units with kick. button bikes later came back with that 6-speed lifestyle after Honda began to produce them without the kicker mechanism (internal to the cases) which returned a couple mm to each gear and dog. This is anecdotally supported in several places without direct measurements such as https://thumpertalk.com/forums/topic/220366-200x-5-speed-vs-xr200-6-speed/

This is further supported by the presence of all the kickstart pockets/holes/molding features in the 6-speed 230 cases. The only telling feature is the right side cover having a blanked kickstart boss. (https://www.hondaprokevin.com/pictures/crf230f-oem/honda-crf230f-engine-review-specs-dirt-trail-bike-motorcycle-crf-230-230f-crf230-3.jpg)

Shaft diameters are identical, with the kicker gear living on a busing/boss that slides over the end of the mainshaft. This bushing/gear piece acts as a bush to increase the effective diameter at bearing seat to 30mm from the 20mm of the shaft. Translations of the part's function description from Chinese are hilarious.
23875
The sole purpose of this bushing/gear/boss interface into the outer clutch basket was to transfer motion from the internal kickstart mechanism into the outer clutch basket which then translated the same motion back through to the crank.

Both the CBS300 5-speed and the CBS300 6-speed use an inner clutch basket spline pattern similar to the output shaft (countershaft) with the retainer being a circlip as opposed to Honda's nut and bolt design. This is probably to cut costs, but should be fine for a sub-40hp street bike.

With Zong's MX-specific motors (NC250, NC450) they switched to Honda's mainshaft fastening method of nut and bolt. The NC series also follows the vein of the CRF250R/CRF450R/KXF250/RMZ250 of having the primary-type kick start outside the cases.

Also, I haven't see anybody else say it... but the DOHC configuration of the NC250 looks identical to an a KXF250/RMZ250, as does the water pump, boreXstroke (identical), cylinder, stud spacing, etc. (for the DOHC). Seems like a fairly inexpensive risk for potentially Japanese performance.

Previously on Self-Destruction, a six-speed gearset was obtained from a undisclosed location in the Far East. Question is: Was a six-speed shift drum also obtained?
Looking at the gear pairs, side by side, it seems the position of the pairs is somewhat different, necessitating a substantially different drum arrangement.
As I understand the previous, the kick start end of the counter shaft uses that space and results in a 20mm shaft, the 5, a 30mm. You found a gear(Honda?) with the proper ID and OD that just might make your shaft work in the case you have. Does that summarize properly?
OT {The pictures are so large, reading the text is difficult as a L-R scroll is needed. How in heck did you intersperse words with images. All I can do is 'attach image' and it puts them all at the end of the post.}

tom

6-speed drum, 6-speed forks, 6-speed pawl (smaller diameter roller) and 6-speed pawl spring (stiffer since shorter range of travel) were all procured at the same time as the 6-speed gearset.

I am also quite confident about the bearing because Tianda's service manual shows a picture of the mainshaft with a very non-standard bearing (looks suspiciously identical to the Honda) even though their parts diagrams only show the 5-speed engine (all production Tianda bikes are 6 speed).
23874

In other news, the CRF100 suspension linkages fit and are in. THe seller described the part as "good working order" but I suspect they didn't actually look at the bearings. THe bearings are kept from excessive play by copious amounts of rust. Win some, lose some. This means the CRF100 lift linkage will work perfectly, and manual manipulation of the swingarm shows a MUCH more linear and constrained motion.

Lifan clearly welded the linkage tabs on from the left side first, as both tabs have a slight inclination that direction. I can nudge them both about half a mm to the right, which will re-center the linkage, and the excess slop can be consumed by a slice of the old Lifan sleeve. Honda's design is nice because the bearing caps and sleeves only serve to locate the moving element whereas Lifan tries to compensate for tolerance stacking and fitment by constraining the moving element with fastening force. The Lifan method is not very healthy for welds in the long term. It also explains why on my brother's Kp Mini (stock rear suspension) the back end feels really dead and lifeless compared to even my previous Hellcat.
23876

Billet lifting link ordered.

franque

01-30-2021, 01:44 PM

deadwood83

01-31-2021, 10:40 AM

I'm not sure what you're speaking of when you say Honda's nut and bolt design, but my clutch basket on my NX125, which I'm fairly certain is pretty standard, is retained by a circlip on the mainshaft.

Hahah that is my bad. I should have specified nut and bolt as seen on the CRF150+ motors. The mainshaft basket end is threaded.

23887

Last night I tore everything off the lifan frame except the headstock so I could leave it on the lift for fitment assessment after cutting the cradle from the XR.

I also puzzled over the wiring. I cut the OEM head unit plug off and re-terminated for a bunch of 2.8mm sumitomo spades. I cannot figure out how Lifan is getting RPM signal on the 2021 models.

23886
On their own wiring diagram it says Bl/W (blue/white, coil output) should go to the dash... but it doesn't. The only wire which goes to the dash from the CDI is the Br/Bl (brown blue) which connects (per the diagram) to a switched positive (key), and goes where the CDI kill switch (ground = kill) normally goes. There's yet another switched positive on the engine stop right below it which connects to the engine stop switch.

On the EFI front, I found the injector which reviewers say cross-references with the Rojo setup. WIth those specs, injector swaps should be possible. Since Rojo is hesitant to release new maps, or give concrete information about how to pull and alter them, injector swaps should provide a way of tuning.

The CBS300 also uses a different manifold mount angle compared to NC250 which means that I will need to attach my own injector bung if I want to get the spray pattern to evenly aim at both intake runners. Somebody makes a weld-on affair for just that. https://billet-speed.com/products/fuel-injector-holder-small

THe manifold, however, is pot metal. Welding pot metal is.... ehhhhhhh. I may have to settle for JB weld and a freezer/oven fit. If I used the normal NC250 manifold, it would point the intake tract directly at the central frame member.

23888
TIme to email the mfg of that adapter and

grumpyunk

01-31-2021, 01:24 PM

FWIW, the bk/y on a lot of china bikes is the tach/coil signal. I expect that the diagram is incorrect to some degree. The injector and associated brain need a tach signal, to know when to apply ground to the injector winding. It may be there is another sensor that amounts to a crankshaft position sensor... or a camshaft position sensor. The latter would make more sense if you injected only near the intake stroke... the former more likely to be used in a 'waste spark' system, firing the sparky plug at each occurrence of TDC(most do).
I think there is a diagram/schematic not being shown here... I know that if I wanted to add a tach, the bk/y wire was the one giving the proper signal. One other possibility is an inductive pickup on the output (secondary) of the coil, much as the 'wrap around' operated tachs use. I guess you need a pinout of the tach, and then follow the wire back to its other end... hard to do from here.
Thinking a bit about the injector and the adapters from billet-speed, the 45 or 90 end of their product could be cut to any desired angle for clearance and pointing purposes. Could the manifold be installed 'upside down' to more properly align? The manifold and the injector should not give a fig as far as operability. The squirt from the injector should be enough to get to the center of flow, and if gravity would 'drop' it too much, change the install angle to compensate by aiming a bit higher than center-of-flow. Maybe?
The billet-speed adapter could be clamped to the OD of the manifold, with a smaller diameter bored hole for spray to pass through. Fillet or curve the adapter to match the manifold curvature, and clamp the injector adapter using the 'clamp points' where the actual injector is captured in the adapter, just pulling the other way.
If desired, a undercut(terminology not my thing here) such as a protrusion, material NOT cut from the adapter, could be left to match the bung hole in the manifold. That would help locate the adapter along the length and side to side on the manifold(centering...). Clamp using modified screw-type hose clamps, and fillet with JB Weld or using rubber/silicone washers to seal the junction.
If you were really into design/construction/etc, you could use some direct-injection style, and port them directly into the cylinder head. sounds like fun! not...
tom

deadwood83

02-09-2021, 11:57 PM

THe Xr100 swingarm needle bearing kit came in. The supplied needle bearings are too small. Turns out Lifan saved money by using grom-style (25mm OD, 44mm long) swingarm bushings in the swingarm tube with a rolled sheet metal tube between the bushings. The swingarm is HEAVY, but aside from a GCraft alu swingarm for the XR100M (Motard, JDM) there doesn't seem to be anything else.

After half a day's research, I decided on NK17/20 bearigs (2 per side) and 1 seal per side on the outer edge. Grease zerk in the middle of the arm and one AS1226 thrust washer per side to handle axial loads from the swingarm. The central tube from the XR100 Pivot Works kit will be reused since it is the exact length of the swingarm tube. It's also machined, all oily, and very shiny. It;s probably the bulk of the swingarm rebuild kit cost.

I am still struggling to get the original bushings out. I do not dare put the swingarm in the press because the tube edge is too thin. I think I may just need to get a long flat head screwdriver and hammer on it until the handle breaks, then hammer on the steel end. The press can be used rather easily for the reinstall since the NK17/20 is a machined bearing as opposed to a drawn cup. It can take a weeeee bit of abuse.

In other news, the lifting linkage is awesome. It is built with a brand new Honda bushing setup. If I were a smarter man, I would have bought Pivot Works since they offer needle bearings there as well. When it eventually fails, I can just swap at that time.

HUUUUUGE plus to the linkage upgrade:
I was able to take the 10mm bearing sleeve from the bottom of the old (Honda) linkage and it fit perfectly in the bronze pivot bushing. My cheap ebay shock with 10mm hole is now useful.
23954

Rad mounts came back from the cutter/bender. They are beautifully made. The design is really simple and hard to screw up, but I managed to do just that. Each side is about 7.75mm offset from the frame. No worries though. Considering there will be a steel bolt through the steel rivnut in the steel frame, I think in this instance it is okay to use nylon spacers around the screws to get the offset correct.

23955

I cut up the front mount and made it top hole only. It came out fine. Deburred everything and smoothed the edges with hand filing.

Now... the bad news:
The donor XR200 cradle is about 3" short. I did not properly analyze downtube angle differences.

This means..... it is time to learn welding. I am working on TIG because it can be done for short bursts indoors. It also made the most sense to me, since I have been surface/PCB/SMT soldering for years. Heat area to be joined, dab the filler (solder) on the hot area and not the arc (iron) and move forward. I'm sure I will screw it up a lot but that's fine. If necessary, that can be the last thing I do.

I have ordered all the supplies, just waiting on an argon retailer to get back to me. If I don't hear anything by Friday, I'm driving to Argas to speak with Jake Mecham.

grumpyunk

02-10-2021, 09:39 AM

If you have access to a weird chisel, cutting the old bushings, splitting them, will allow them to collapse and be driven out more easily.
I think for welding, TIG is maybe more artful to learn than the flux-cored style. The sheet metal and in general, thin metal was welded by doing a lot of 'spots' and moving on a bit, leaving space between spots, and allowing it to cool. Then come back and weld next to the previous spot. Rinse repeat. A whole lot of connected 'dots' of weld that are then finish ground. Avoids distortion. I don't think you can TIG that way as you have to build a puddle to add material to. I very possibly could be wrong. If you don't like the flux-core, then get the next one up the line, with the inert gas that prevents oxidation.
Just re-read. I thought doing sheet metal, it would warp and hole before you could add material. Guess you will find out. My welding skills are little to non-existent, having only repaired a few things that split. I used flux as it was low investment compared to a TIG/MIG or other. The 'stack of dimes' that seems to be a goal of pro and amateurs made me think that TIG would require more 'hovering' over a spot leading to distortion of thin metal.
There are special chisels that are used to split bushings, such as in transmission repair/rebuild. Once spit, they drive out almost effortlessly.
tom

deadwood83

03-01-2021, 11:55 PM

Rear suspension is in. Machined needle rollers on machined through-length bushing, with thrust washers and dust caps.

New linkage is in. Rebuilt using Honda OEM. Sort of wish I went Pivot Works on the linkage.
24027

Radiator swap... again. This time SV650. Easier plumbing. I'll couple with 4" SPAL fan and thermo-bob for true bypass thermostat integration. Also supports inlet and outlet design to support thermo-siphon while bike cools (doubtful much will happen, but for peace of mind)

I can now do this because... I have been learning to TIG. I'm only about 2 hours in, but after 1 hour of playing around, I made a bead I was pretty happy with considering my lack of experience. Made it on the metal frame the KP Mini shipped in. I still have a ways to go, but I am pretty happy with my progress for never welding before.
24028

Transmission: It is in. Had some trouble along the way with a bolt snapping, but it fits with the Honda bearing and one shim washer from the old mainshaft. All new gaskets along the way. I also needed to cut down the kick starter gear to act as a bushing (it has a bearing insert) for the base of the clutch basket. The shift shaft also didn't fit the 6-speed shift star. CRF230 shift shaft en route to replace that. Kick starter removed. I'll measure and order a plug for the right case cover.
24029

Got the cylinder decked by 7 thou. Machine shop locally was very competitive, had out in and out same day, and for a whopping $43 after tax.
24030

To round out almost the remainder of the engine work, I finished a 3-angle valve job with 60, 45, and 31 degree cuts. Photographing the angles with my phone is crazy hard. You can see my lazy-boy dychem job (sharpie).
24031

To Do:

Investigate the possibility of swapping for the NC250 cam.
Finish engine assembly. Swap valves, motoseal on base gasket where case seam is, set valves slightly loose
Weld radiator bracketry
Extend cradle and weld to frame
Plumbing
Fan and temp-controlled relay
Fueling (still no word from Gary, and the Chinese wanted to sell me the ECU without any wiring or supporting bits for the same price as the whole kit). Wiring harness will be addressed once I do fueling (last) with this.
Exhaust. NK250 is... close. THe header flange is too wide and im looking for the name of the cup they weld to the piping at the head so I can order one of the proper size.

This is all going a lot faster now that I have a welder, I don;t have to spend so long researching ways to properly bolt/press-fit/seal things together.