Go Back  R/C Tech Forums > General Forums > Electric On-Road
Testing - What's better?  Lets do some science, and share the results. >

Testing - What's better? Lets do some science, and share the results.

Testing - What's better? Lets do some science, and share the results.

Old 12-16-2016, 05:41 PM
  #61  
Tech Elite
Thread Starter
iTrader: (2)
 
Join Date: Sep 2010
Location: Chicagoland
Posts: 2,196
Trader Rating: 2 (100%+)
Default

Originally Posted by roosterreagan View Post
What if we chip in for a FPV setup? Then use gradients and locators and the like visually for shock/arms/link movement. Plus it'd just be cool to watch the action in real time.
Chip in? I've got a spare Tiny Whoop camera and mount, and a DVR equipped monitor. "it's already bought".

If you have something you wanna do, PM me, I'll ship it to you.

Edit: Seriously, if anyone wants to contribute, but doesn't have ~exactly what they need~ say something. EG: Voltmeter, test tube and ball bearing for oil testing, etc... We (I?) will try to make it happen for you.

Last edited by Nerobro; 12-16-2016 at 05:54 PM.
Nerobro is offline  
Old 12-16-2016, 07:34 PM
  #62  
Tech Regular
iTrader: (16)
 
roosterreagan's Avatar
 
Join Date: Apr 2016
Location: music city
Posts: 293
Trader Rating: 16 (100%+)
Default

Originally Posted by Nerobro View Post
Chip in? I've got a spare Tiny Whoop camera and mount, and a DVR equipped monitor. "it's already bought".

If you have something you wanna do, PM me, I'll ship it to you.

Edit: Seriously, if anyone wants to contribute, but doesn't have ~exactly what they need~ say something. EG: Voltmeter, test tube and ball bearing for oil testing, etc... We (I?) will try to make it happen for you.
Great! I was suggesting to 30tooth the fpv setup for his dynamic chassis setup, I think being able to watch these components 'real world' and use visual measuring points would be idea, not to mention just plain neato. I too will offer any help or resources at my disposal.
roosterreagan is offline  
Old 12-16-2016, 10:53 PM
  #63  
Tech Apprentice
 
icecyc1's Avatar
 
Join Date: Sep 2014
Posts: 99
Default

Originally Posted by 30Tooth View Post
Oh I see, static and dynamic similarity in tire load. Easily doable only have to buy another set of linear potentiometers. Good thing I've got this Pro4 I can say if and how torque steer happens.

My TODO list:
-torque steer and tweak deviation;
-simple shock dyno;
-tire grip comparison (hopefully tire slip angles can be known);
-roll vs traction;
-flex as adjustment;
-aerodynamics and effects;
-blue wire on the motor sensor harness is temp so I can piggyback and record motor temp;
-acceleration vs time graphs.
30Tooth.... I know you've followed my thread and results for my shock dyno, and thanks for posting it to this thread.

I've kind of slacked on my shock dyno testing, or at least reporting lately. If you are interested in building a shock dyno, I don't mind providing you with info about how to build one, and the little pitfalls I discovered along the way. My first attempts didn't go so well, and it took having access to very expensive and high precision data acquisition and instrumentation to discover what was needed. But after that, I do feel pretty confident in the "low speed" dyno results.

I have built an impact tester and I had started gaining confidence in the data, but it still is not clear to me about how to properly "normalize" the pack results. It might not be a critical feature on this on-road thread, but it's one of the most critical to off road, and I think I've started to figure it out, but I don't get enough track time to really understand and prove it out.

I love this thread and believe the OP is doing the right thing with proving claims with testing. That was my goal with building the shock dyno, and the results were very clear and obvious that most claims with shocks and pistons were pure marketing hype and did nothing that could not easily be replicated with a simple flat piston with the right oil, especially for low speed damping.

As for shock oil, what I have learned in working with a guy who actually makes and sells shock oil is that the inconsistency is in the mixing. Sloppy mixing makes for inconsistent oils from batch to batch. I believe given an equivalent actual viscosity, regardless of the brand, it will be affected by temperature only, so thus temperature is a significant qualifier needed. Predicting that variability for each of the brands is fairly impossible and inconsistent. The amount of testing will be overwhelming and is likely not worth it. The best bet is the advice given to stick with one brand and hope they have tight quality control. But knowing someone who has tested many batches of many brands, the consistency control is not as good as you'd like to believe. (his testing is in a highly controlled temperature and humidity environment, so I tend to believe him. I also did all my 1/8 scale testing with his oils that he made sure I got what I wanted)
icecyc1 is offline  
Old 12-17-2016, 07:32 AM
  #64  
Tech Elite
iTrader: (1)
 
Join Date: Sep 2010
Location: My house.
Posts: 3,567
Trader Rating: 1 (100%+)
Default

Originally Posted by Nerobro View Post
Hall effect sensors can be analog too. Gotta find the right ones. Like these: http://www.allegromicro.com/en/Produ...s/A1318-9.aspx

You're right, going to the ground is a fools errand. You could put a reflective sticker on an arm, and measure to that... The gradiant sticker on a sheet of plastic and phototransisor/photoresistor/photodiode/photocell method might be cheaper... but it will take more work to make work consistently. As ambiant light could be an issue.
Amazing. I'll be on it! The angular one seems the best solution.

Originally Posted by roosterreagan View Post
What if we chip in for a FPV setup? Then use gradients and locators and the like visually for shock/arms/link movement. Plus it'd just be cool to watch the action in real time.
I haven't tried FPV but I have one gopro clone I could put inside the bodyshell and see how it looks!

Originally Posted by roosterreagan View Post
Great! I was suggesting to 30tooth the fpv setup for his dynamic chassis setup, I think being able to watch these components 'real world' and use visual measuring points would be idea, not to mention just plain neato. I too will offer any help or resources at my disposal.
Will do, thinking about painting the lower or upper arm with a hi-vis colour to be easily seen in the vid.

Originally Posted by icecyc1 View Post
30Tooth.... I know you've followed my thread and results for my shock dyno, and thanks for posting it to this thread.

I've kind of slacked on my shock dyno testing, or at least reporting lately. If you are interested in building a shock dyno, I don't mind providing you with info about how to build one, and the little pitfalls I discovered along the way. My first attempts didn't go so well, and it took having access to very expensive and high precision data acquisition and instrumentation to discover what was needed. But after that, I do feel pretty confident in the "low speed" dyno results.

I have built an impact tester and I had started gaining confidence in the data, but it still is not clear to me about how to properly "normalize" the pack results. It might not be a critical feature on this on-road thread, but it's one of the most critical to off road, and I think I've started to figure it out, but I don't get enough track time to really understand and prove it out.

I love this thread and believe the OP is doing the right thing with proving claims with testing. That was my goal with building the shock dyno, and the results were very clear and obvious that most claims with shocks and pistons were pure marketing hype and did nothing that could not easily be replicated with a simple flat piston with the right oil, especially for low speed damping.

As for shock oil, what I have learned in working with a guy who actually makes and sells shock oil is that the inconsistency is in the mixing. Sloppy mixing makes for inconsistent oils from batch to batch. I believe given an equivalent actual viscosity, regardless of the brand, it will be affected by temperature only, so thus temperature is a significant qualifier needed. Predicting that variability for each of the brands is fairly impossible and inconsistent. The amount of testing will be overwhelming and is likely not worth it. The best bet is the advice given to stick with one brand and hope they have tight quality control. But knowing someone who has tested many batches of many brands, the consistency control is not as good as you'd like to believe. (his testing is in a highly controlled temperature and humidity environment, so I tend to believe him. I also did all my 1/8 scale testing with his oils that he made sure I got what I wanted)
Don't thank me, I just did what I had to do which was sharing the best work I've seen in RCT, if we have to thank someone is you. Shock dynos are worth gold here.

I am interested, will send my email address by PM ASAP. I did test my arduino with 50Hz data acquisition and was OK, 500hz is untested but will shoot for that number. Should be enough for TC applications.

Thank you all.
30Tooth is offline  
Old 12-17-2016, 08:00 AM
  #65  
Tech Champion
iTrader: (22)
 
robk's Avatar
 
Join Date: Aug 2001
Location: Macho Business Donkey Wrestler
Posts: 8,069
Trader Rating: 22 (100%+)
Default

Originally Posted by 30Tooth View Post
Exactly, flex is a something that adds a lot of margin to account for driver/mechanic errors. We could run stiff chassis with stiff arms (they aren't needed though) if we knew everything about our car handling performance. The more you know the less flex in components is a help. 1:1 setup theory applies 100%.
No it doesn't. Not in the way you're thinking of anyway, and you've pointed it out in your response. 99.9% of drivers will not have a data acquisition system available to them, or the data scraping skills to use that data to make decisions.

Further, I have actually raced with some guys who have full scale experience. One guy in particular I have seen running a super stiff chassis, and his car was really good, but his base of knowledge was vast in both rc and full size. I'm not sure what he was doing would be the best solution for most rc racers who didn't have his frame of reference. That's why development has gone the direction it has for cars available for retail sale.
robk is offline  
Old 12-17-2016, 10:01 AM
  #66  
Tech Elite
 
niznai's Avatar
 
Join Date: Feb 2009
Location: All over the place
Posts: 2,974
Default

Originally Posted by roosterreagan View Post
I totally could see tuning flex as easier, I guess that makes sense. I often forget that not everyone is as much as a nerd as I am. I understand you can never get ALL flex out, I was speaking to adding flex as a means to gain traction, it was late last night, I'm fairly certain I just rambled through that post, the key is: the longer the post, the greater chance I kind of forgot why I was posting in the first place, and instead of erasing and wasting all those letters and words I try steering it back on course. Whoa sorry I just subjected yall to the quagmire that is my stunted thought process, I'll try and limited that exposure going forward, promise. Rooster
I didn't mean flex was added on. I meant flex was found to exist and rather than try an expensive exercise of eliminating it, companies just go with it, and people try to tune around it. Meh. With or without it shouldn't make a difference if you know how to tune your car.
niznai is offline  
Old 12-17-2016, 01:01 PM
  #67  
Tech Elite
iTrader: (1)
 
Join Date: Sep 2010
Location: My house.
Posts: 3,567
Trader Rating: 1 (100%+)
Default

Originally Posted by robk View Post
No it doesn't. Not in the way you're thinking of anyway, and you've pointed it out in your response. 99.9% of drivers will not have a data acquisition system available to them, or the data scraping skills to use that data to make decisions.

Further, I have actually raced with some guys who have full scale experience. One guy in particular I have seen running a super stiff chassis, and his car was really good, but his base of knowledge was vast in both rc and full size. I'm not sure what he was doing would be the best solution for most rc racers who didn't have his frame of reference. That's why development has gone the direction it has for cars available for retail sale.
Not in that way, when I meant know the suspension is roll distribution which you can do using a spreasheet. One end rolling more than the other means the end that rolls more will twist/flex the chassis right? Get the roll centres to do a handshake and flex becomes less of a influence.
For example last week, the less flex I had in the motor mount area the more consistent and true to the geometry I choose the car behaved. With flex I was chasing my tail so to speak. YMMV
30Tooth is offline  
Old 12-17-2016, 01:50 PM
  #68  
Tech Addict
iTrader: (2)
 
MaxRain's Avatar
 
Join Date: Feb 2007
Location: Calgary, Canada
Posts: 576
Trader Rating: 2 (100%+)
Default

with all this discussion, would there be a way to measure the amount of g-forces a TC goes thru? I think of Forza/Gran Tourismo and the circle graph that is available in the analytic section of the game. I think this would help determine if the car is hitting the maximum g-forces that it can tolerate without losing grip. A stiff chassis would likely reveal that high g-forces would be reached quicker as a car rolls, where not as quick with a softer one.
MaxRain is offline  
Old 12-17-2016, 02:03 PM
  #69  
Tech Apprentice
 
icecyc1's Avatar
 
Join Date: Sep 2014
Posts: 99
Default

Originally Posted by MaxRain View Post
with all this discussion, would there be a way to measure the amount of g-forces a TC goes thru? I think of Forza/Gran Tourismo and the circle graph that is available in the analytic section of the game. I think this would help determine if the car is hitting the maximum g-forces that it can tolerate without losing grip. A stiff chassis would likely reveal that high g-forces would be reached quicker as a car rolls, where not as quick with a softer one.
Here is a system you can use to monitor many things on an RC vehicle while driving. It connects in line between the battery and ESC (Or I suppose you could power it with a separate battery too). You can get an accelerometer for it to monitor the g's. Bob Wright from RC Crew Chief put one on his cars (both on-road and off-road) and it works pretty well. The sampling rate is not super high at 50Hz, but it should be enough to get good basic roll/pitch forces during cornering. It's not a fast enough sample rate to get suspension displacement data which would need to be more on the order of 200Hz or more depending on what properties you are looking for.

http://www.eagletreesystems.com/inde...&product_id=54
icecyc1 is offline  
Old 12-17-2016, 07:50 PM
  #70  
Tech Elite
Thread Starter
iTrader: (2)
 
Join Date: Sep 2010
Location: Chicagoland
Posts: 2,196
Trader Rating: 2 (100%+)
Default

Originally Posted by robk View Post
I'm not sure what he was doing would be the best solution for most rc racers who didn't have his frame of reference. That's why development has gone the direction it has for cars available for retail sale.
I strongly suspect, none of the r/c companies has the money for the sort of FEA, or real chassis work.

I'd expect to see more prototypes. More radical changes in prototypes.

The amounts of adjustment available in chassis makes me think that they expect us to do the engineering work that "should be done by them".

Originally Posted by MaxRain View Post
with all this discussion, would there be a way to measure the amount of g-forces a TC goes thru? I think of Forza/Gran Tourismo and the circle graph that is available in the analytic section of the game. I think this would help determine if the car is hitting the maximum g-forces that it can tolerate without losing grip. A stiff chassis would likely reveal that high g-forces would be reached quicker as a car rolls, where not as quick with a softer one.
Easy. Any SPI or I2c 6 axis gyro, an arduino or whatever else you want to pick up that data, and a serial port on your PC.

Problematically, if the chassis if flexing, you need to know the shape of the chassis, which means you need to glue an accelerometer to each end of the chassis.
Nerobro is offline  
Old 12-17-2016, 08:56 PM
  #71  
Tech Regular
iTrader: (16)
 
roosterreagan's Avatar
 
Join Date: Apr 2016
Location: music city
Posts: 293
Trader Rating: 16 (100%+)
Default

Originally Posted by Nerobro View Post
I strongly suspect, none of the r/c companies has the money for the sort of FEA, or real chassis work.

I'd expect to see more prototypes. More radical changes in prototypes.

The amounts of adjustment available in chassis makes me think that they expect us to do the engineering work that "should be done by them".



Easy. Any SPI or I2c 6 axis gyro, an arduino or whatever else you want to pick up that data, and a serial port on your PC.

Problematically, if the chassis if flexing, you need to know the shape of the chassis, which means you need to glue an accelerometer to each end of the chassis.
I would think you'd need two gyros, one on an A arm and one on chassis plate. The chassis mounted one would be subject to chassis roll, wonder if that roll acceleration would muddy the waters for a true measure of actual lateral force... the A arm mounted one would no doubt be a pure stat that together with the other gyro might give you even more data points, compare contrast, maybe time deviation between the two and given enough hz might see something like oscillation or if the chassis rolls once and returns to neutral in a constant direction change... I dunno so much possibility.
roosterreagan is offline  
Old 12-17-2016, 10:07 PM
  #72  
Tech Elite
Thread Starter
iTrader: (2)
 
Join Date: Sep 2010
Location: Chicagoland
Posts: 2,196
Trader Rating: 2 (100%+)
Default

What you care about is chassis position. If you know the position of the front, and rear, you can figure out how much flex is happening. Checking chassis position is not closely related to a-arm position, thre's far to much flex in the tire, wheel, knuckle, pins, and ball ends.

Roll isn't a problem to deal with. You just need to integrate two of the axises.
Nerobro is offline  
Old 12-18-2016, 07:44 AM
  #73  
Tech Champion
iTrader: (22)
 
robk's Avatar
 
Join Date: Aug 2001
Location: Macho Business Donkey Wrestler
Posts: 8,069
Trader Rating: 22 (100%+)
Default

Actually, I think they don't expect anyone to to any chassis work. There are a lot of guys who would get a good start by getting the correct camber and ride height on their cars. The people on this thread are a small minority of racers who have the interest in this type of research as part of their definition of the hobby. Most guys just want to have fun racing - at most they are using a XXX main book or the Xray cheat sheet to help set up their cars. Then there are the guys who have raced for years and have a ton of "try it and see knowledge".

One reason why Xray sells a ton of $500 cars is that they try to provide the information that their top tier drivers have developed on the car, and they make it easy to access. That makes a the purchase worthwhile, vs. re inventing the wheel constantly.

http://site.petitrc.com/reglages/xra...6_SetupHelper/
robk is offline  
Old 12-18-2016, 03:28 PM
  #74  
Tech Elite
iTrader: (1)
 
Join Date: Sep 2010
Location: My house.
Posts: 3,567
Trader Rating: 1 (100%+)
Default

Two G sensors seems right to me.

@robk, wait, did you called us nerds?
30Tooth is offline  
Old 12-18-2016, 03:46 PM
  #75  
Tech Elite
iTrader: (3)
 
Join Date: Jun 2003
Posts: 4,351
Trader Rating: 3 (100%+)
Default

Originally Posted by Nerobro View Post
The amounts of adjustment available in chassis makes me think that they expect us to do the engineering work that "should be done by them".
Maybe it is there so we can tune a car to suit a driving style.
rccartips is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information -

Copyright 2021 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.