About a year ago, I started work on my new project - home made G-meter. The equipment is almost done and here are the first results. I was on the track yesterday, so it was a perfect opportunity to test it.

Here is our track with numbered turns, located in Czech Republic, Europe:


And here are the data from my G-meter:


It is a record of my second lap. As you can see, the G-meter measures both acceleration/braking and cornering G, engine speed and time.

The main thing that you can see on the graph, is that I am a beginer on a track .

I also didn't stay in VTEC while leaving the turns, probably a big mistake.

Finnaly, I could not race until the end of the day, because of some overheating problems - see this thread if you are interested.

 

That's really cool! What tires were you using? Looks like you were pulling a little over a g in a few places. I bet you could use your G-meter as a dyno too. Even if it doesn't give you hp numbers, if it's repeatable you'll be able see differences from engine mods. Sorry your track time was cut short; that looks like a fun course. How fast are the straights?

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by flyrod &raquo;</TD></TR><TR><TD CLASS="quote">That's really cool! What tires were you using? Looks like you were pulling a little over a g in a few places. I bet you could use your G-meter as a dyno too. Even if it doesn't give you hp numbers, if it's repeatable you'll be able see differences from engine mods. Sorry your track time was cut short; that looks like a fun course. How fast are the straights?</TD></TR></TABLE>

I'm using some cheap Yokohamas and my car is 95 Prelude VTEC, completely stock.

Yes, it can be a dyno too, but I have some problems with vibrations at the moment.

The top speed here is from 180kmh (120mph) to 200kmh (130mph) in a Prelude 4G VTEC, depending on a driver. I reach only 180, my friends 200, I'm not a very good driver

 

Color me jealous. I wish I had a real track in my neighborhood. That is soo cool. Get any footage? looks like it would be a sweet lap. Any pics of the Accelerometer?
the Pirate

 

It is a very good track, comporable to F1 tracks, even if F1 cars are only testing on this track, not racing. Motorcycles are racing here with MotoGP races.

 

Petr,
this is amazing
Hope you car will be cured ASAP.

Tom

 

that track looks beautiful, you got very nice project going

 

I was on the track again today. I added some more inputs to the G-meter, I was measuring, both G's, RPM, vehicle speed, water and air temperature, throttle angle, MAP sensor, and VTEC.

Here are some results:

 

may i ask what kind of software are you using to log the rpm and all the other sensor inputs, i am very curious

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by petrv &raquo;</TD></TR><TR><TD CLASS="quote">I was on the track again today. I added some more inputs to the G-meter, I was measuring, both G's, RPM, vehicle speed, water and air temperature, throttle angle, MAP sensor, and VTEC.

Here are some results:
</TD></TR></TABLE>

man thats wicked tight, nice clear graph too, currious what the hardware softwae setup is too? let us know

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Eddiebx &raquo;</TD></TR><TR><TD CLASS="quote">may i ask what kind of software are you using to log the rpm and all the other sensor inputs, i am very curious</TD></TR></TABLE>
This all (SW+HW) is my custom work. .

It looks like this:

 

NICE!!!
maybe you can help me with this infrared morse code detector circuitry. my problem with it is that it's boring as hell to test! yours on the other hand... looks mighty fun.
I especially like how you've mixed all those new SMT devices and the PLCC controller, then you have those old-school resistors all crammed up. that's sweet.
Are the four cans with the markings (with a real professional blue marker, I might add) gyros or accels?

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Jagan'92 &raquo;</TD></TR><TR><TD CLASS="quote">
Are the four cans with the markings (with a real professional blue marker, I might add) gyros or accels?
</TD></TR></TABLE>
I use accelerometers, not gyroscope.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by petrv &raquo;</TD></TR><TR><TD CLASS="quote">
This all (SW+HW) is my custom work. .
</TD></TR></TABLE>
I like it.
I assume you have three accels then for the g-metrics, but then what about the fourth? Is that the rotational measurement? Or are you compensating for gravity when your car is on an incline by using it to average out the other three? Where did you come up with all of your parts and the design?
Keep up the good work and R&D!

 

The accelerometer is only 2-axis, acceleration and cornering.

 

Ok got it: a single 2-axis accel, mounted horizontal in the car so it won't measure any gravity, and lined up so the two axes are along the same axes as the car's. If the car tilts, won't the readings add some gravity?

We have a UAV I developed sensor circuitry for, and though it performs much different than a car, we used 4 accels (one redundant axis), 3 gyros (each axis), an absolute pressure and differential pressure. With all of these, we could determine all of the forces on the plane, roll rates, plus it's altitude and speed. However, we never did have a spiffy LCD screen, just the molex connectors with analog values out.

 

wow very sweet, how do you log the data? is there an onboard rom, or do you have some kind of serial connection on there.

And what language did you write the software in? I wish I would get off my fat *** and do something like that,

 

Beautiful surface mount! Still really cool but how hard would it be to add another axis for accuracy? What are the accerometer's tolerances? Like deviation. I guess I am wondering how accurate they are... Damn that is soo cool

the Pirate

 

Very nice. I'm doing the same thing, but I'm just in the prototype stage.



Components (top right clockwise): SD card for datalogging, 2 axis accel ADXL202AE, rs-232 (laptop interface), bottom left is the RPM VSS analog input board.

The SD card is a nice feature and I can get 64MB+ of datalog data into it before I need to move it to the laptop.

This will also be running my dash and I have RC servos for analog display of Tach and speed, with 7 segment and bar displays for other data.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by DirtyLude &raquo;</TD></TR><TR><TD CLASS="quote">Very nice. I'm doing the same thing, but I'm just in the prototype stage.</TD></TR></TABLE>
I wish you good luck!

 

I wanna try it in my car!!!!

 

Actually I think we talked about this about 6 months ago or so when I first just free wired the accelerometer onto a board on the Toronto Club forum.

What I would really like is to get some PC software for analysis of the G data and playback of the datalog. I started doing some Visual Basic programming, but I'm not very good at it, and it's going slow.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by DirtyLude &raquo;</TD></TR><TR><TD CLASS="quote">Actually I think we talked about this about 6 months ago or so when I first just free wired the accelerometer onto a board on the Toronto Club forum.</TD></TR></TABLE>
It's possible, I have spoken to many people about this

 

I wired some more signals and tested it on the track yesterday. Here are the results:

 

Petr,
I'm analyzing your charts just for pleasure at work and noticed that you achieve over 1G lateral acceleration on corners... Do you drift then? Lude can take near 0.8x G in normal conditions (but can't describe what is "normal" though).
I have similiar project to yours, but I think about logging things direcly from ECU...