Roll your own dyno for performance tuning!

[Buccaneer]

too tedious. sorry.

[Weldangrind]

Perhaps you would reconsider posting your performance info. Plenty of very detail-oriented people here who would appreciate it.

[JerryHawk250]

Same here. Was kind of disappointed you had deleted it before I could read it. That kind of stuff doesn't bore me.

[Buccaneer]

I'll revise and resubmit.

[Buccaneer]

I thought I would share my recent efforts to establish a realistic top speed for my SSR/Italjet Buccaneer 250i, since I see a lot of interest in top speeds here. I've worked out some general ideas on how to do repeatable, valid top speed tests on the street. Following those ideas, I now have a sort of "roll-your-own" dyno I can use whenever I want.

My goal was a two-way average top speed measurement procedure, which can cancel the effects of wind if done properly. This is how top speeds are tested at Bonneville under FIM rules. Anything else invites inflated reports that are partly due to wind conditions, downhill slopes, and so on.

My final result for my Buccaneer was a two-way average of 72 mph, with me aboard in a tucked position, as I will explain. This is in contrast with some reports in the press of over 80 mph, and with this, from Revzilla.com: "With a top speed of about 90 mph, the Buccaneer will get up to speed and keep up with traffic on the freeway...." My 72 mph is not too different from the 75 mph claimed in the owner's manual.

Several considerations are necessary to a good job of measuring a two-way top speed.

Site selection is absolutely crucial. I found this out when I tried a two-way test on a stretch of divided highway that looked about right, on impulse. I got 78 mph one way, and 77 the other way, which I recognized as far too high compared to earlier runs. My mistakes are a good object lesson in what not to do.

Returning to the same stretch of road later I recognized several problems. First, the apparently straight and level stretch actually had a shallow, barely-noticeable valley profile, so that I was going downhill when I arrived at the middle of it from both directions! This allowed gravity to increase my speed both ways. I also noticed a lot of small hills of dirt due to construction, and other wind deflectors. Since the road was a divided highway with a large median, there was plenty of opportunity for such features to affect one direction more than the other.

Finally, I realized I was making a really fundamental error in simply recording my highest speed wherever it occurred in each direction along a mile or so of highway. This procedure meant I was actually measuring top speeds at different points on the roadway in the two directions, and that those points could be differentially advantaged by wind and gravity due to the differences in local topography.

Since I had already gotten a two-way average of about 73 mph in more controlled conditions, I recognized I needed to be more disciplined.

I arrived at these criteria for good measurement:

1. a straight stretch of road about 1 1/2 miles long, with no overall valley or hill profile, and especially with a dead-level central segment of about 1/4 mile. The purpose is to avoid the "valley speed effect" described earlier.

2. a low-traffic, two-lane road with clear lines of sight, removed from wind-blocking topographic or constructed features. The reason for the two-lane requirement is to keep the wind conditions nearly the same in both directions, since the two lanes would be adjacent to each other. Clear lines of sight are needed for detecting deer, traffic, police, and other dangers. Avoidance of wind-blocking features should help keep the wind steady at the point of measurement.

3. a pre-selected, definite location on the stretch of road: an exact point where the speed will be recorded in both directions. I call this the trap, and mine is a particular sign post near the middle of the 1 1/2 mile stretch, and confirmed to be clear of even slight dips or rises for about 1/8 mile in each direction. Using the same trap for both directions should allow wind and topography effects to remain essentially unchanged on the passes in the two directions, so that any effects will be cancelled by averaging the speeds.

Before getting to my test results, I'll say more about how I went about satisfying the first two criteria. I first used Google Maps on my phone to look for straight roads of 1 1/2 miles or more, with few or no crossroads. I had some ideas before I started, but the map let me be more objective. I then followed up with US Topo Maps, a free app that has all the different-scale U.S. Geodetic Survey topographic map series combined into a zoomable format(!). With this I could examine hills and slopes, to the resolution of 20-foot contour lines. I settled on two likely stretches about an hour west of me that looked good on those maps. One of them had another advantage that turned out to be extremely useful: It was built right beside a railroad track that was following the bank of the lower Wisconsin River.

The lower Wisconsin River has a "uniform, low gradient" of only 1.5 feet per mile (per U.S. Geol. Survey Open-File Report 75-582, online). This is a (1.5/5280)*100 = 0.03% gradient, well below the 1% maximum allowed by the FIM, which sets the rules for speed tests at Bonneville and around the world (per the Wikipedia article on land speed records). In short, the river is almost dead level along this stretch.

The train track built beside the river shares its gradient, as can be seen by its constant relative elevation above the river over a long distance. Train tracks are essentially free of minor dips and rises by design. Since the track is only 20 yards or so from the highway, it is easy to tell when the highway has small rises or dips, by sighting over to the track.

The train track provides much more information about the road's topography than the 20-foot contour topographic map, and was extremely helpful in my choice of the central segment of the overall test stretch where the trap would be located, helping me detect a slight rise over a culvert and another slight dip.

There are some small hills near one end of my track, and a road tees into it a hundred yards before my trap, but otherwise it would be hard to improve on. It seems nearly perfect for speed testing. Now that I have found it, I can use it going forward, knowing that I have established a reliable standard.

Some other considerations for the testing: Winds should be calm or light, and not variable. A strong, steady wind would throw off results even in a two-way test since a head wind slows a vehicle more than the same velocity tail wind speeds it up. (There is a mathematical reason for this.) Of course variable or gusty winds could defeat the two-way measurement entirely if their direction and/or speed differed at the trap on the second pass coming back in the opposite direction.

You should control as many variables as possible. Set your tires to the same pressure before all tests. Wear similar clothes, zipping up your jacket. Use the same or similar helmet in all test sessions: an open-face helmet is not as streamlined as a full-face helmet with the visor down. Choose and practice the position you will take on the motorcycle: probably either a close tuck for highest speed or a normal riding position for more practical results. Tuck your legs in to the sides, maybe put your left arm on top of the gas tank. But whatever you do, do the same thing each time unless you decide to change the protocol.

Each test should be done in three passes through the course. The first, done at legal speed, is a sighting pass to look for cops or other problems. The second and third are done flat out for top speed, coming back and then turning back around. Don't forget you must note your speed only at the pre-selected trap, in both directions. At the end of the third pass you will be heading in your original direction and ready to continue on your way before "the heat" arrives.

One more thing: Make sure to record or remember whether you used a GPS for testing. If not, some kind of calibration of your speedometer would help interpret results.

My test results:

I tied a short pencil to my handlebar, and taped a note card on my tank, to make it possible to record results right away, while still riding. Here is the transcribed data from the card, with the two-way average GPS speeds and other notes added later:

    trial   low   high   average   location         note      tires   trap  
        1    70     72        71      Cty C   20W/50 oil   28/30psi     no  
        2    70     76        73      Cty C   10W/40 oil   36/36psi     no  
        3    69     77        73      Cty C   10W/40 oil   28/30psi     no  
        4    67     74      70.5     WI 133   10W/40 oil   28/30psi    yes  
        5    68     77      72.5     WI 133   10W/40 oil   28/30psi    yes  
        6    67     77        72     WI 133   10W/40 oil   28/30psi    yes

I don't consider all six tests to be strictly comparable, but for now their average top speed of 72 mph is my claim for the performance of a well-broken-in Buccaneer, with me aboard, in a tuck.

Concluding comments:

Carefully-controlled two-way average top speeds reflect the horsepower the motorcycle is producing at top speed. This is not necessarily the peak horsepower available from the engine, since top speed may not occur at the power peak. Top speed observations can be the basis for tuning decisions, but I'll leave that topic for later if at all.

Of course not everybody will go to the trouble of setting up their own natural dynamometer along a stretch of highway as I did. One reason might be that careful measurements will almost always be disappointing compared to careless ones (and the results won't be much for bragging about). But I've noticed a good deal of serious discussion of top speeds here, and hope some of these ideas will be worth others' consideration.

How lucky we are not to be dealing with 150 mph+ sport bikes, for which top speeds are almost always only a speculative topic. We get the fun of testing and tuning for them on the street, with the chance to do careful measurements to evaluate our work. And all without paying some knucklehead $300 only to have him make fun of our puny engines while testing them on a "real dynamometer."

[JerryHawk250]

Thanks for posting.