[Trakit]

I've been doing a bunch of digging on different set ups lately. Everything from one touch options like AIM, Chase, etc. I think I've decided on the cheep route because of a couple reasons. 1. It lowers my investment initially and two, i got a great deal on a new GoPro. I could get a used system of the mainstream options for this but I think the video quality will be better with this setup

So here's my setup I am leaning towards. Prices are approx.
Phase 1- see how it all works and go from there, use gps on phone.
1. Racerender Trackaddict HD for iPhone $20 - already has the tracks I run at pre-programmed
2. GoPro3+ BE and extra batt - $300
3. Racerender software to compile vid and data into one stream $40

Phase 2 - Add granularity and data
1. OBD2 GoPoint BT1 - $120
2. GPS - Dual xgps150 - $150

Total after everything is bought will be $630 for HD Vid and data.

Is anyone doing something like this already? What iPhone app? I could use the vid on phone and save the gopro, but I already have the GoPro.

 

[marcinr]

I've got the exact same setup, using Harry's Lap Timer on the iPhone. I use the xgps150 and the GoPoint BT1A. The results look like this:
https://www.youtube.com/watch?v=wndB87oNPtw

This works ok, but not great - the OBD-II data streams are desynchronized from each other and from the GPS. They're not just offset by a given amount of time, but rather, there is jitter in the rate at which they arrive at the BT1, and OBD-II data isn't time stamped, so there's no easy way to adjust it.

I've started working on some code which looks at the first and second derivatives of the data and tries to figure out which should correlate, and corrects these drifts, but I don't have that working perfectly yet. For example, you know that speed and RPM are directly related when in gear, which you can deduce from just those two channels. Longitudinal acceleration is correlated with the first derivative of speed, and hence RPM. Throttle position is directly related to positive longitudinal acceleration, while you can figure out braking strength from longitudinal deceleration. Anyhow, it's tricky to get right, but I'll post the script here if I ever finish it

 

[jds62f]

thats a lot of dollars of your setup going into a disposable appliance (phone). I realize its a little more convenient but when I was looking I decided I preferred something standalone.

aim solo + go pro + racerender. unless you've got an 08+ car its going to be challenging collecting data from the ecu, probably cheaper in the long run to go trackmate and collect the data yourself. If you're new to data collection, staying on the cheaper side of things makes a lot of sense... video is cool but so far the graphs have proved more valuable for me.

 

[holmz]

I am not sure I would classify an iPhone, a "Harry's lap timer", and a GoPro as a "data acquisition" system... The same way that while tooth fairies and dentist work in teeth, they are not usually classified as as oral surgeons.

 

[Trakit]

Thanks a lot, that's great info. It's also very troubling that the obd2 data is that out of sync. Do you think that is just the nature of that module and a wifi one would be better?

I've got the exact same setup, using Harry's Lap Timer on the iPhone. I use the xgps150 and the GoPoint BT1A. The results look like this:
https://www.youtube.com/watch?v=wndB87oNPtw

This works ok, but not great - the OBD-II data streams are desynchronized from each other and from the GPS. They're not just offset by a given amount of time, but rather, there is jitter in the rate at which they arrive at the BT1, and OBD-II data isn't time stamped, so there's no easy way to adjust it.

I've started working on some code which looks at the first and second derivatives of the data and tries to figure out which should correlate, and corrects these drifts, but I don't have that working perfectly yet. For example, you know that speed and RPM are directly related when in gear, which you can deduce from just those two channels. Longitudinal acceleration is correlated with the first derivative of speed, and hence RPM. Throttle position is directly related to positive longitudinal acceleration, while you can figure out braking strength from longitudinal deceleration. Anyhow, it's tricky to get right, but I'll post the script here if I ever finish it

That's the whole thing. For my initial setup, I only am out 20 for the app if it doesn't work. If I go the full deal it's another 270 for the obd2 and the external gps. I guess if it's a total failure I lose what I would lose on resale of those items. Thanks a lot for the info. I have looked at the aim and have a buddy going that route so we can compare.

thats a lot of dollars of your setup going into a disposable appliance (phone). I realize its a little more convenient but when I was looking I decided I preferred something standalone.

The iPhone has a pretty good setup for capturing g's, I can get path and segment times, as well as some obd2 inputs(although marcinr is reporting there may be issues with that currently). I understand it's not an integratd solution, but my thoughts were that it would be pretty good for wetting the appetite and if demands dictated it seems as though second hand high end set ups are prevalent. It's acquiring data into a single view for analysis after the event, if not during the day of the event. To get multipleHD feeds on an integrated system runs upper 1k's. What specifically are your concerns? Perhaps the granularity of the data? What do you run?

I am not sure I would classify an iPhone, a "Harry's lap timer", and a GoPro as a "data acquisition" system... The same way that while tooth fairies and dentist work in teeth, they are not usually classified as as oral surgeons.

 

[jds62f]

trying to datalog with odbii is junk. The system responds 'when it can', which doesn't give you any guarantee of consistency between anything you're logging. To get data more reliably you need to pull data through CAN, which is something actually possible with a 2010, or better yet, run your own sensors. If you are new to this all, you'll get pretty far with speed, lateral/longitudinal g's, and your position on track.

iphone gps resolution is ~5hz i think (although linked one says only '1+ hz'), not really ideal for data logging. With the standalone you get twice the resolution from the gps (10hz). Its great to have choices, but hard for me to understand why you'd chunk down $270 for what appears to be inferior GPS capability... you're not too far away from an equivalent standalone (aim solo) at that price. Maybe I'm missing something.

 

[Trakit]

Thanks for more info. I think that seals it then. I will just stick with the iPhone resolution to get my feet wet which is essentially $20. To your point I think that should be descent for initial feedback. Then if my desire grows, I'll get a full fledge setup, whatever brand that may be. Knowing me, that's a certainty so I'll start cheep, plus video and go from there. I had hoped for more from OBD2 but I guess it just isn't there. I'm not new to the track just been a while and data acq has come a ways since I was into it last.

trying to datalog with odbii is junk. The system responds 'when it can', which doesn't give you any guarantee of consistency between anything you're logging. To get data more reliably you need to pull data through CAN, which is something actually possible with a 2010, or better yet, run your own sensors. If you are new to this all, you'll get pretty far with speed, lateral/longitudinal g's, and your position on track.

iphone gps resolution is ~5hz i think (although linked one says only '1+ hz'), not really ideal for data logging. With the standalone you get twice the resolution from the gps (10hz). Its great to have choices, but hard for me to understand why you'd chunk down $270 for what appears to be inferior GPS capability... you're not too far away from an equivalent standalone (aim solo) at that price. Maybe I'm missing something.

 

[marcinr]

Thanks a lot, that's great info. It's also very troubling that the obd2 data is that out of sync. Do you think that is just the nature of that module and a wifi one would be better?

It's not a problem with the reader, it's a problem with the OBD-II protocol, it's not meant for any sort of realtime acquisition. It's a pull protocol where the reader has to ask for a specific set of data, and then the ECU pushes that data whenever it can. The GoPoint BT1 which I use has better latency than the wifi based PLX devices Kiwi which I also tried.

I'm investigating what it takes to read the CAN bus, which is a realtime firehose of data, but you have to have a magic decoder ring to understand the manufacturer specific binary format.

 

[holmz]

...
The iPhone has a pretty good setup for capturing g's, I can get path and segment times, as well as some obd2 inputs(although marcinr is reporting there may be issues with that currently). I understand it's not an integratd solution, but my thoughts were that it would be pretty good for wetting the appetite and if demands dictated it seems as though second hand high end set ups are prevalent. It's acquiring data into a single view for analysis after the event, if not during the day of the event. To get multipleHD feeds on an integrated system runs upper 1k's. What specifically are your concerns? Perhaps the granularity of the data? What do you run?

The only concern was semantics.
An iPhone app and a goPro may be a perfect way to start.

I suppose what is the goal?
If it is chassis set up then you have a specific set of data that is important.
For the driver then it is other data, and the camera(s)

I have a dedicated logger which takes in 4 CAN channels and an RS232.
A 10 Hz GPS, and a 100 HZ IMU.
When it is in the car, I have some other sensors to put in (damper/shock positions and steering position), and plan to put in brake pressure sensors.

When it is not in the car, then I may be doing something like this:

 

[Trakit]

The one time I try to save money you guys bust my bubble. rotfl

I still will go simple for a while with just the trackaddict app and then keep my eyes open for an integrated solution. So far I like Trackmate, but I'll do more research before I plunge and hopefully get the use all of the systems before choosing.

 

[holmz]

It is mostly semantics as a go pro (just spell checked as hopeless) and an iPhone is NOT a DAQ system.
But it is perfect for what you want and the camera is useful in any system. And the phone one can use for calls.

Knowing the goal(s) then helps to define the data that is needed and the analysis that is needed. This is seldom done first in a systems engineering fashion.

 

[bri3d]

If you want speed and tach data off CAN the legendary "Retrofit 08 black dash" thread will teach you what you need to know provided you already have a device that can capture arbitrary CAN data.

http://www.lotustalk.com/forums/f129/retrofit-08-instrument-cluster-61076/index3.html

 

[Trakit]

holmz, the goal of this will evolve as most things do. Initially it will be to improve me so that I understand when I maximize a turn and line and when I don't. It will also be a way to compare set ups, but I will keep that very simple(I usually change one thing per day/weekend). With video I can also see problems and correct or relish in my brilliance(unlikely). I "believe" I can do this simply(read simple setup) initially with sector times and lap times plus vid. That way I can objectively see G's under braking and when I start braking. It won't show me trail braking and corner velocities at high hz, but should be a start.

Over time I think my appetite will turn into true analysis. I'll probably have a predictive model that will show tire degradation based on driving style, speed and environmental factors. I'm a data dork, but I have to resist that initially because I know I have a lot to improve on behind the wheel. I have had many other cars with data that I gather from standalone ecu's, just never for road racing purposes.

bri3d, thanks I will start reading that. I ran across it a while ago, but didn't know why I wanted to read that. Now I will have a reason to learn it.

 

[holmz]

...
...I'm a data dork, but I have to resist that initially because I know I have a lot to improve on behind the wheel. I have had many other cars with data that I gather from standalone ecu's, just never for road racing purposes.
...

Well if you are a "data dork", then :
1) maybe a dedicated logger that can move from car to car can allow the investment to not be lost over time?
2) It might be the fastest way to improve behind the wheel?

And the screen shot I provided should be proof that you may not be alone...:sheep::sheep:
This is not from the same event as the data, but same idea.

 

[pwildfire]

If you want it to be more than a little useful for driver training, you really need to consider an option that has configurable inputs, as it is pretty hard to tell what the difference is from lap to lap unless you can see driver inputs. If you can measure RPM, steering angle, throttle position, and brake pressure that will be a huge improvement over measuring just outputs. You also might want to try to find an option with a yaw measurement.

If you become a good enough driver that you can utilize that data, you are also going to find that you want vehicle data as well, so shock positions and tire temps and etc.

The only real limit to how much data is useful is whether you can figure out how to properly analyze it, and whether you can afford the equipment.

If you are just starting out at road racing, I would not worry too much about the future, and go with the cheapest, simplest option, which is a phone app and a 10Hz gps unit. Once you run a few lapping sessions with that, then you can evaluate whether it is really worth the money to go to a real datalogger. IMO it is sort of all or nothing, as the in-between options don't offer much more useful data than a phone app, but cost is a significant fraction of what you would pay for a "real" data system such as Motec.

You might look into 'RaceCapturePro' as a very low cost (comparatively) data system as well.

 

[sturgeongeneral]

I must put my vote in for the Traqmate system. I have the Traqmate complete system and the data analysis you can do with it is incredible. Tie in the Gopro and you will have a comprehensive system that is unmatched. Well worth the investment!

 

[Exiged]

I'm putting this Motec C125 into a lotus 211 right now

http://www.motec.com/image_display.php?path=676306784_c125_ANIM-combined4.gif&w=570&h=306


Sent from AutoGuide.com App

 

[holmz]

If you want it to be more than a little useful for driver training, you really need to consider an option that has configurable inputs, as it is pretty hard to tell what the difference is from lap to lap unless you can see driver inputs. If you can measure RPM, steering angle, throttle position, and brake pressure that will be a huge improvement over measuring just outputs. You also might want to try to find an option with a yaw measurement.

If you become a good enough driver that you can utilize that data, you are also going to find that you want vehicle data as well, so shock positions and tire temps and etc.

The only real limit to how much data is useful is whether you can figure out how to properly analyze it, and whether you can afford the equipment.

If you are just starting out at road racing, I would not worry too much about the future, and go with the cheapest, simplest option, which is a phone app and a 10Hz gps unit. Once you run a few lapping sessions with that, then you can evaluate whether it is really worth the money to go to a real datalogger. IMO it is sort of all or nothing, as the in-between options don't offer much more useful data than a phone app, but cost is a significant fraction of what you would pay for a "real" data system such as Motec.

You might look into 'RaceCapturePro' as a very low cost (comparatively) data system as well.

^This^

An iPhone and GoPro are not a really a system unless you have some other "systems" way to tie them together. You will also not be able tell you much from sector times unless you are very consistent. But it's a great place to start.

If you want to assess the driver that was also well outlined above.

 

[Trakit]

I realize anything is possible, with enough time and money, but pwildfire's comments started me thinking(never a good thing). Do people utilize slip angle calculations for car performance analysis? I would think that that would be very valuable. For the rear it would be an easy measurement, but for the front it would be a little more difficult as you would have to have steering angle measured. This would really put significant value into sector and corner times.

Also, I've always shy'd away from Motec in the past as the "tuners" for them had the devices locked down, but that is for Engine management not data acq. I will have to see if the other devices like AEM/Haltech/etc. have data acq. just for curiosities sake. And another layer of the Onion peals off.

Exiged, let us know how the install and use of the Motec goes. I'm very interested in the setup. I shudder to ask, but what was the cost of entry?

 

[Exiged]

I realize anything is possible, with enough time and money, but pwildfire's comments started me thinking(never a good thing). Do people utilize slip angle calculations for car performance analysis? I would think that that would be very valuable. For the rear it would be an easy measurement, but for the front it would be a little more difficult as you would have to have steering angle measured. This would really put significant value into sector and corner times.



Also, I've always shy'd away from Motec in the past as the "tuners" for them had the devices locked down, but that is for Engine management not data acq. I will have to see if the other devices like AEM/Haltech/etc. have data acq. just for curiosities sake. And another layer of the Onion peals off.



Exiged, let us know how the install and use of the Motec goes. I'm very interested in the setup. I shudder to ask, but what was the cost of entry?

The C125 with the race dash logging kit, expanded I/O loom. I/O upgrade charges, and pro analysis software is roughly $5k


Sent from AutoGuide.com App