How important is data acquisition to racers?

[Honclfibr]

Question for all you veteran racers out there, have you ever considered/used data acquisition systems to help improve your lap times? And if not, is it something you would think might be useful? I'm talking about recording things like cycle speed, plotted course (gps lat/long), lean angle, suspension travel, when and where your knee hits the ground, steering travel, etc.

If you were able to view this data, correlated, after a lap, and compare to other laps, would it help you to improve your times? What about during a lap, in a sort of heads up display, would any of the data be useful then?

Mostly a curiosity, our company is somewhat involved with data acquisition systems and I've played with the idea of trying to outfit a motorcycle with one, just for fun, but I don't know enough about racing to know how much technology would help, and how much your lap times are improved just by learning and feeling out the course.

[Degsy]

I have thought about it. GMD has a setup you can rent. probably cost a fortune though.

It would be VERY VERY useful IF you had someone available who could read it and really analyse the results and make positive suggestions.

derek

[GMAN226]

It would be pretty neat. I think that the more important stuff would be suspension travel, RPM, Front and rear Wheel Speed, throttle position, brake application. Not so much the knee down or lean angle of the bike.

If you could do it affordably it would be worth it. I think the GMD setup runs about 12K or so to own...


G-man

[LM954RR]

I am sure that information would be useful, if one could find a way to utilize it. Another thing is, at speeds, a rider is typically completely focused on the task at hand, and leaves very little room for conscious decisions to be made based upon prior data study suggestions.

[gmdboston]

Data aquisition has been very helpful in chassis set up. I have a Pi Research system that is very flexible and expandable, it cost me just under $15,000. I can sell you a Drack system that is very good and also quite flexible and you have to struggle to send over $5000. Many of the systems out there can do some good data logging, but many of the cheap ones are a waste of money. Installation is a little tricky. there is no "bolt on" set up. you need to make your own wheel speed sensors and bracketry. I have a welder, lathe and mill that makes it easy. The last thing to consider are the rules. Many sactioning bidies prohibit dat aquisition, but I know at Loudon they will not DQ you for it unless some one protests. It is legal for Super Bike and GP.

[legalspeed]

I am interested. We have daq2000 boards here at work and have thought about it a lot. There are high impact data cubes for collection now too.

I would be interested in:

wheel spin at any given point on the track
RPM's, MPH, gear indicated, and of course, suspension travel.

All would have to be mapped to a given track position.

Tire temp, ambient weather, three axis attitude; (front to back lean, side to side, & elevation.

Fuel consumption and A/F mix would be nice too.

Can you have this by next Wed? We race again in two weeks.

[Honclfibr]

Hey thanks for the input everyone, that's some excellent feedback. I think it'll make a fun project, and if it turns out to be at all useful, I'd appreciate any test dummies who might volunteer

I'm going to start out by tinkering with the palm/gps combo that I have. Reason being that gps is the easiest to set up, there's software readily available to log the output, and it shows a good deal of data which is going to be important and difficult to get elsewhere. Mainly vehicle speed and coordinate position, speed could be nabbed from the bike but coordinate position is pretty essential to any DAQ system.

My concern with the gps is that it may be too slow and inaccurate for motorcycling. GPS absolute accuracy is stated at around 6m which isn't going to be sufficient, and at least the GPS's I've used usually only update about once a second, which is probably a bit slow for objects moving in triple digits. But my experience is that relative accuracy is much higher, which is what you're really concerned with, and there are ways to post-process the GPS data to make it a bit more accurate. I'm going to play around this weekend, see what motorcycle data looks like.

Beyond that, I think the next step would be some sort of attitude sensor. Just so happens that I've got a few (100?) prototype accelerometers hanging around that might work well for that application. They're meant to be high accuracy so they run in an 85 degree oven to keep temperature drift down, which might draw a bit too much power for motorcycle applications. But with a little work I think I could use the onboard temp sensor to do temperature compensation and get away with no oven for the accuracy needed here. The current application requires a differential accuracy (two sensors) of 0.2 degrees which is a bit overkill for this I think. Also you need a good deal of dsp filtering to get that, which adds lag and also complexity. I can hack the current sensors to output a serial stream, there's a small Micro onboard the sensor which does basic filtering and gets the accelerometer sensor noise down below half a degree anyway. I would figure that's enough to get some sort of useful data out of it.

Course, at 80 frames per second I dunno if my palm will keep up, plus the GPS is using it's only serial port. Might need to find another source of data recording, too bad because the palm is nice and small and cheap.

Anyway, thanks again for all your input, I'll keep you informed on how my little project progresses

[Honclfibr]

Originally posted by legalspeed
Fuel consumption and A/F mix would be nice too.

Can you have this by next Wed? We race again in two weeks.

No problem Dan, you got that P.O. for me yet?

[Honclfibr]

Originally posted by Degsy
It would be VERY VERY useful IF you had someone available who could read it and really analyse the results and make positive suggestions.

I'm pretty decent on the analysis and interpretation side, been doing that for a few years on tractor trailer road tests. What I lack is a strong racing background. So there's probably a limit to how useful I could be in the suggestion department. Guess if I wanna get serious about this, I'll have to learn to get fast

[Degsy]

Originally posted by Honclfibr
Hey thanks for the input everyone, that's some excellent feedback. I think it'll make a fun project, and if it turns out to be at all useful, I'd appreciate any test dummies who might volunteer

Test dummy number 105 reporting for duty, SIR!

Derek

[Degsy]

Just an FYI, suspension travel is the most important info in my opinion, that and wheel travel.
derek

[Kip]

Degsy, just get you cams fixed them start thinking of how you can beat Vella.