CFR

who's using one?

more of a gadget than usefull??

but i do like a gadget

pictures of an install would be nice to see

CFR

NovakTwo

I checked our web site for install photos and didn't find any. But here is the primary information page with a link to the PDF of the instructions.

The quick start instruction file does show a hook up photo. Install photos would depend on how many sensors you plan to use:

Dawn of a New Sentry

I did find an install photo on page 7 of the instruction manual. The instructions' download is pretty cool. I had never looked at it before....

mattnin

iTrader: (44)

I like the Sentry, and I believe it is useful. It gives a lot of data about your vehicle. I am using a Sentry in my T4, but I plan on throwing it in my Jrxs-R once the touring season starts. There is a really fast guy at the track, and just yesterday I asked if I could log the G-forces on his truck just because I am curious. He seemed complimented by that too.

It all depends on how much you want to get out of it. Some people don't need it or care to use it, but I think it makes my racing experience richer. If you ever get a chance to pick up the textbook, Race Car Vehicle Dynamics, you'll see how you can analyze the Sentry data. I use the Sentry to do a G-G diagram, Total G graph, and I am even using it as a Dyno. You can do setup changes and if you are driving consistent, you can average out your G-Forces to see if the setup change actually helped what you wanted. You can have an actual number to see if you have more turning G's or forward bite G's, etc... The Sentry gives you all the data like an F1 engineer would need.

Aside from the software that comes with the Sentry, using a spreadsheet you can also do things like shown below with Sentry.

gameover

i found 0.1sec resolution is too slow to do anything useful with the gforce, rpm, steering position, throttle position or current measurements. if you're lucky you will get one point per corner and 3-5 points on a straight.

20x the sampling speed and increase the flash chip size to handle the extra data and i'll buy it again

mattnin

iTrader: (44)

.1s readings are standard for racing data loggers. Look at full scale racing data loggers, they take readings every .1s too.

Cammer

iTrader: (3)

I agree, .1 may be the standard in Formula One but they don't do a lap in 10.2 seconds. I find the data very choppy coming out of this.

mattnin

iTrader: (44)

I think it would be an incredible headache to have reads every .01s. Can you imagine all the data you would get after a 5 minute heat? That is 30,000 data sets that could consist of 5 temp sensors, 1 current, 1 voltage, 1 rpm, X and Y G forces, and throttle. This would be a very large file! Microsoft Excel has a limit of 32,768 rows or something around that. It would be quite difficult to sift through that large file and pinpoint a section of the track you would want to analyze. Can you imagine sifting through all that data to find turn 4 and lap 15? Impossible!

Instead, we can use statistical averaging and get meaningful results with a manageable file that isn't overly bloated taking reads every .1s.

Look at the G-G diagram. That graph is saturated. Can you imagine what it would look like with 10x the data? It would appear solid black!

Cammer

iTrader: (3)

I wanted to use mine so I could compare voltage, amps and rpms and use the info to setup my turbodyno. I could get some good info out of it but not really what I needed. Perhaps if I raced on more flowing or less technical tracks or offroad where you are not looking for the hard out of corner acceleration it would be different, but not for what I need it for.

mattnin

iTrader: (44)

Cammer, why do you think offroad sentry users aren't looking for hard out of corner acceleration? That is exactly what I am looking for. Look at the G-G diagram. If you would like, I can highlight the area that shows corner acceleration. Look at the bottom arrow shaped figure that appears on the graph. That is showing cornering G's on acceleration.

mattnin

iTrader: (44)

I would like to show everyone this. Below I have two graphs, Heat 2 and the Main. Heat 2 I was using Holeshots front and rear, and for the Main I was using Jconcepts barcodes rear and pink taper pins front.

I didn't label the graph, I should, but the X axis is the left and right hand turning in G's, Y axis up is braking, and Y axis down is acceleration.

Notice my diagram using Holeshots. Firstly, their isn't any real definition in acceleration corner G's and for that matter, hard braking in a straight line G's. The graph appears flat. Compare that to the Barcodes and Tapers, there is a distinct definition in hard straight braking and cornering on acceleration. And if you look at the graphs and view them as a slideshow, you can even see the angle at which cornering acceleration occurs is at a steeper slope, meaning I am accelerating harder. The upside-down Y is apparent.

Holeshots


Jconcepts Barcodes and Pink Tapers

gameover

you can't make a detailed analysis without all the data and at the speeds we are talking 10Hz just doesn't cut it. It is just plain luck that you "might" get the peak G's around any particular corner or see the signs of traction issues on corner exit etc. When i used it all i could see what basically left/right on the G's, i couldn't capture the steering turn-in profile on a corner entry and correlate it to G's , and on exit (after switching to monitor the throttle channel) i couldn't see the throttle ramp-in or the effects of looser/tighter diff settings, wheelspin, peak current due to gearing change etc.

I never said the system was without use, it is useful, just not what i was hoping for. It's excellent for monitoring temperatures and also gearing correctly on straights and other stuff that isn't very dynamic or if you want to do long term averaging as shown in your plots - which would be a lot clearer at a higher sampling rate Actually there isn't anything out there as the Eagle Tree loggers are also 10Hz. I know we are talking about RC level products and price points, but i think the MCU's and flash chips used in the product would be capable of more than 10Hz resolution (maybe i'm wrong) and i think many people would trade off options of storage space for higher resolution options. eg. 1 x 6mins @ 100Hz instead of 16 x 6mins @ 10Hz.

As an example if i do a hot lap, i can see that its a great lap from the timesheet (or a trackside timing device), then i want to go into as much detail as i can about what made that particular lap great. Then if i do a better lap i want to compare that lap to the other one in very fine detail. If i have two good laptimes but the data shows i was fast in different sections you want to combine the good stuff from both in hope of getting better etc.

most modern showroom cars can datalog in realtime at over 25hz, while also running the engine. Modern formula 1 dataloggers operate at sampling rates in the kHz range, even good quality karting loggers operate from 50-500hz.

If someone released an RC logger with 50-100Hz resolution and say 10mins recording time at full resolution and other lower resolution/time tradeoff options, everything the Sentry has, simultaneous throttle/steering logging, say 4 optional user defined 0-5v inputs instead of 6 temp sensors, with some decent graphing software attached to it below $500 i would buy it tomorrow.

Fred_B

iTrader: (1)

If you want high sample rates it's going to take a much larger commercially available logger (like a Pace Scientific). You'll have to adapt your own sensors to it but it will go to a fairly high sample rate. It's likely that most would get lost in the data.

Also, the data from an RC car isn't going to look anything like the data you would see on a full sized car due to the scale. 100 Hz data would look like a big rats nest when you dump it into Excel. You'd end up having to go corner by corner and it would take more time than it's worth.

Being able to compare changes on the track for gearing and battery changes is probably the most useful part of the system.

I do wish they would have included a ring type hall sensor for current (Amp-sense for instance) but the set-up they have isn't all that bad.

gameover

just different expectations i guess, with the amount of time spent bench racing, commenting on voodoo and snake oil on this forum i think spending the time to go corner by corner on some decent data would be time better spent

i'd have to respectfully disagree that the data will look a remarkably different on an RC car due to the scale, everything about RC car suspension design and setup looks identical to me... there is no reason you shouldn't be able to use hi-res Gforce data, steering angle, throttle % to diagnose setup, tires and cornering technique. Which in my experience is where races are won and lost. It would be cool if you could measure body roll or suspension position but the sensors for that would be a bit pricey, to some degree you can extrapolate that from G-force data.

I get tired of reading threads arguing the benefits of C rates on lipo's for motors that draw less current than a light bulb, or what brand of snake oil do you put on your widgets or why blue is better, lol. it would be cool to see some hi-res RC data and discuss why this and that is faster/slower based on something actually related to racing...

Fred_B

iTrader: (1)

The problem is that most people will get lost in data that's at a high rate.

I actually ran data on a sedan at 200Hz back in the day and it's barely fast enough to capture suspension movement. Yes, it "looks" like the full sized data that you would get from a normal race logger but we need a much higher frequency to capture meaningful data. Plus you're limited on the number of laps that will fit into excel and graph the result.

In the end the high frequency data acquisition was just an expensive way to prove out why a sedan double steers. It only confirmed our gut feel for what was happening.