data acquisition systems

[sean996]

Hey all,

Anyone have any experience or opinions about available data acquistion systems?

I have an Innovate Motorsports systems but it requires you to use the wideband O2 sensor to internally capture data . . . also no GPS input for correction when track mapping . . . curious how accurate this stuff can be with all the bells and whistles . . . hoping I can get something like my F1 game's data acquistion and analysis.

Thanks,

Sean

[Mike]

I have heard some debate over GPS and beacon systems.
Is it true that GPS though convenient but may not be consistent from weekend to weekend?

I'd be interested in experiences or comments regarding this system.
http://www.aimsports.com/auto/index.html

[kary]

I am in the process of looking at the G2X from Racepak. Nice system, probably plenty for us weekend warriors but then there are those that think they need everything like a professional. With that comes high dollars to purchase and to install. I have been talking with racepak about distributing these units. They are affordable and easy to install.

The web site is a bit weak at the moment but I have used the software with their demo data from Infineon and it is quite nice. Provides G's in all directions, RPM's, Speed, and GPS coordinates, tracklayouts and various other analysis features. The software allows you to do many things but of most interest is the ability to segment the track as you wish and obtain segment entry and exit speeds and split time. It also provides great comparison of other sessions against a session by lap. Really too many features to really write about here, but I believe that many folks, when talking about GPS, talk about years past in accuracy. That is simply not the case today. The other thing to realize is this is not meant to be an exact timing system though they say through data testing it is within 0.10 seconds of a regular timing system. Good enough for analysis for sure. Racepak does alot with NASCAR and drag racing as well. Have been in the business for many years.

Anyway, your mileage might vary but I thought it useful to post this information to your question. Here are some links from other boards that discuss this product as well as a comparison of other products. I should be be carrying the G2X within a few weeks.

They just updated the website and it provides some great info and pictures:

http://www.g2extreme.com/
http://www.s2ki.com/forums/index.php?showtopic=307682
http://forums.rennlist.com/rennforums/s ... p?t=231661
http://www.race-technology.com/HomePage.html
http://www.traqmate.com/index.php

[Chris Benbow]

I had the AIM Mychron Gold in the Spec car and really liked it.
The display is really cool. You can put several different numbers on the screen simultaneously, set several warning lights to whatever values you want. Then you get the F1 style shift lights as well.

The data acquisition is wonderful for telling you what's going on in the car. You can compare data from different days, different setups, different drivers. You can draw a track map and divide it into sectors. Monitor cornering force, braking force, throttle position, oil pressure, exhaust gasses, shock movement...The possibilities are endless.

GPS is cool and convenient but not necessary. You can get a very good idea of your position on the track from the lateral accelerometer + wheel speed. GPS is a separate data point to compare to but still isn't accurate enough to pinpoint your location well enough to compare lines through a turn or anything.

The Mychron works very well once it is all set up. The software (same thing for the MXL) is pretty user friendly and has lots of great features. It's fairly easy to get familiar with.

The setup can be a challenge. Sometimes you can use existing sending units already in the car, sometimes you have to use theirs and adapt them to your car. It's a case by case situation. Then you have the hardware/software interface. The mechanics know the car but not the computer side. We spent a lot of time calling back to AIM for instruction on how to make things work. Their office is up in Lake Elsinore. If I were to do it again I would seriously consider hiring them to do the work (or to come down and work with the shop). Try to get a fixed price for install. If it's the first time they've done your model of car it could take a lot of time to figure out the best setup.

It's an awesome tool but you need to have a good head for graphs and reading data displays to make good use of it. Unfortunately there are few books on the topic either.
Bottom line: The AIM is more for the racer who wants to monitor lots of different parameters. Kary's G2X will be very functional, convenient, and less expensive to install.

[Tim Comeau]

I was just reading about the xtreme3x.com product. Cool that it integrates up to 4 cameras at once with the data. I've put in a call to them regarding recording time capability. Retail is 3K!

[kary]

GPS is cool and convenient but not necessary. You can get a very good idea of your position on the track from the lateral accelerometer + wheel speed. GPS is a separate data point to compare to but still isn't accurate enough to pinpoint your location well enough to compare lines through a turn or anything.

What are the facts behind this statemement about inaccurate lines with GPS?

[Chris Benbow]

Hi Kary,
Don't get me wrong here I've decided after recent research that GPS driven systems are much better that they used to be and plan to buy one soon. Nevertheless they still have limitations.
I guess the biggest "fact" is that I don't see any of the major systems boasting that they enable you to compare lines.
We can also analyze it mathematically:
Tim at Racepak quotes a laptime variance of .05 sec (which is pretty impressive by the way)
Take a car going 60 MPH. That's one mile per minute, 5280 feet per minute, 88 feet per second, and 4.4 feet in the quoted variance period of .05 sec. I'm not a math major so correct me if I'm wrong but I'm interpreting that the GPS can only locate the car within a circle that is 4.4 feet in circumference. That doesn't seem accurate enough to compare lines from one lap to the next. Unless perhaps we're talking about the banked turns at CFS.
If you look at online sources to determine GPS accuracy they describe much larger numbers (ie. 2-4 meters) http://users.erols.com/dlwilson/gpsacc.htm

[Chris Benbow]

I'm realizing that perhaps I wasn't clear.
By comparing lines I meant with a track map drawing. Comparing one drawing to the next.
Clearly with sector timing you can drive one line through a turn on one lap and another line through that turn on the second lap and compare times.
Assuming you're really driving the line that you think you are.

[kary]

Hi Kary,
Don't get me wrong here I've decided after recent research that GPS driven systems are much better that they used to be and plan to buy one soon. Nevertheless they still have limitations.
I guess the biggest "fact" is that I don't see any of the major systems boasting that they enable you to compare lines.
We can also analyze it mathematically:
Tim at Racepak quotes a laptime variance of .05 sec (which is pretty impressive by the way)
Take a car going 60 MPH. That's one mile per minute, 5280 feet per minute, 88 feet per second, and 4.4 feet in the quoted variance period of .05 sec. I'm not a math major so correct me if I'm wrong but I'm interpreting that the GPS can only locate the car within a circle that is 4.4 feet in circumference. That doesn't seem accurate enough to compare lines from one lap to the next. Unless perhaps we're talking about the banked turns at CFS.
If you look at online sources to determine GPS accuracy they describe much larger numbers (ie. 2-4 meters) http://users.erols.com/dlwilson/gpsacc.htm

Chris, I believe what you have quoted there is correct math but in absolute terms. That is a summation of error over the distance not the error at any one point. Also, Tim at Raepak has posted (on the thread links I provided) that the accuracy varies little on a given day but will vary from month to month. That said, I believe the accuracy on a given day is good and comparing lap lines on that weekend for instance will be quite accurate. Remember it is all relative not absolute. When comparing lap data from one year to the next I believe would be a bit misleading, but then this tool is used to tune your line for the conditions that weekend not from year to year because conditions change that force us to change setups and lines anyway.

Your points are well taken but I think with the segmented corner entry and exit speeds it will become obvious which line is better. Given the demo data provided from Tim @Racepak, there was clear change in line with the data graphically. Maybe they are fooling us, but it looked pretty good.

[sean996]

Thanks for all the input everyone.

The simultaneous video capture is also a nice feature for comparing lines when you play it back with all your data.

I'll post what I end up buying.

Sean

[sean996]

race-technology.com is another system with the video capture capability . . . I'd like to combine this with 4 to 6 flat screens and you'd have you're own little "make believe" F1 pit!!!

The installation of sensors, computer integration, and massive data analysis are all things I've done for a living building robotics and automation machinery . . . can't wait to apply it to something really fun!!! I'm also putting a haltech e6x on one of my cars "just for fun".

Sean

[kary]

Thanks for all the input everyone.

The simultaneous video capture is also a nice feature for comparing lines when you play it back with all your data.

I'll post what I end up buying.

Sean

Sean,

I used to think that video was useful when doing analysis but after talking with many DA companies and talking with professional car and cart racing folks, video really provides very little in the way of useful analysis. It is cool and fun but the reality is drivers need to analyze their laps right after the session with hard numbers not pictures. The analysis software provides great graphs and quantitative data that shows you the fast sessions and the slow sections. It also shows you why when looking at the various inputs.

So I would not focus on video if you want real DA. If you just want cool video with overlays of speed then cool, but that is not going to quickly help at the track when you need it most.

A number of fellow PCA'ers are buying the G2X unit so maybe we can share some data when we get them installed and in use to demonstrate what I am saying here.

[Mike]

Kary,
Having many club members using and familiar with the same system (G2X) will create a meaningful resource.
Specifically system operation and shared data interpretation.

A race car engineer friend mentioned that amateur drivers can gain insights with throttle, steering angle and brake data.
It appears this is available as well.
Can that be added to the G2X at a later time?

I have heard too that video overlays are best for just making great videos.

Thanks,
Mike

[kary]

Kary,
Having many club members using and familiar with the same system (G2X) will create a meaningful resource.
Specifically system operation and shared data interpretation.

A race car engineer friend mentioned that amateur drivers can gain insights with throttle, steering angle and brake data.
It appears this is available as well.
Can that be added to the G2X at a later time?

I have heard too that video overlays are best for just making great videos.

Thanks,
Mike

Mike, yes the G2X can add any number of additional sensors to the unit using teh V-net technology they have. This technology is essentially piggybacking as many sensors you want to the cable connected to the G2X box. The additional sensors are available just like the basic data of the G2X in the analysis software.

There is a a package for brake, throttle, and steering as well as oil pressure, exhuast, wheel speed, etc.....quite extensive really.

I think that in the end, if you want all or most of the sensors available, customers will pay the same for any product of its class. The difference in my opinion is the entry level cost of the G2X is far less allowing you to add on when you are ready. Other products are simply $2000 to $4000 straight away because you are buying into the whole DA up front. I like the lower entry cost and "add on when you are ready" model much better, as do most folks.

[sean996]

Thanks for all the input everyone.

The simultaneous video capture is also a nice feature for comparing lines when you play it back with all your data.

I'll post what I end up buying.

Sean

Sean,

I used to think that video was useful when doing analysis but after talking with many DA companies and talking with professional car and cart racing folks, video really provides very little in the way of useful analysis. It is cool and fun but the reality is drivers need to analyze their laps right after the session with hard numbers not pictures. The analysis software provides great graphs and quantitative data that shows you the fast sessions and the slow sections. It also shows you why when looking at the various inputs.

So I would not focus on video if you want real DA. If you just want cool video with overlays of speed then cool, but that is not going to quickly help at the track when you need it most.

A number of fellow PCA'ers are buying the G2X unit so maybe we can share some data when we get them installed and in use to demonstrate what I am saying here.

A thing an instructor mentioned to me about the video was you could compare lines/apex points more accurately than the mapping systems that are using gps and the accelerometer interpolation . . . specifically he mentioned brake pressure changes during braking . . .

I am definitely a person who likes to experiment with things myself since everyone's level of "usefulness" can be different. The video thing made sense, but in the end it may only be a toy since I'm use to staring at graphs anyway.


Sean