You are here

VESC for RC car application

14 posts / 0 new
Last post
andreas78
Offline
Last seen: 2 years 6 months ago
Joined: 2018-02-09 12:43
Posts: 7
VESC for RC car application

Hello,

my name's Andreas and a few days ago I found the VESC while searching for brushless controllers that are FOC capable.

Well, there are quite a lot controllers for brushless motors on the market, some are reasonable good, some are cheap, a few are both. However, due to the nature of these controllers to mostly work on whatever motor you connect to it, the control algorithm actually can't really be ideal, because the controller doesn't know anything about the motor attached to it. They probably all use block commutation.

Now the VESC can do what I would like a good controller to do, FOC and also current control or RPM control mode. And there is a motor setup procedure (wow!), in order to have an optimized control algorithm and superior efficiency, which is mighty important in racing those little cars. But I have a few questions, as due to being an mechanical engineer myself, I naturally am a bit at war with everything concerning electrical engineering...angel

One thing, I was wondering about, the 3S restriction. At least for the 10th scale cars, only 2S lipos are being used. My guess is, the mosfets need at least around 8V at the gate in order to have a low enough Rds. So what happens, if I would use logic level fets, like the IRL40SC209 https://www.infineon.com/dgdl/Infineon-IRL40SC209-DS-v02_00-EN.pdf?fileId=5546d462557e6e890155a1413329602d ?

I'm not familiar if things like higher tr or tf values or lower Qg would cause any issues. Does the lower voltage cause problems with other parts used in the design? The fets specs sure are nice it seems.

 

Another thing, about the sensors, in the documentation it says, hall sensors are less accurate. But that means, that FOC mode is still possible using those, does it?

 

I might want to use it in a 8th scale buggy, too, just like Benjamin did himself. I already have a motor, with 1700KV, 2 pole pairs @4s (16.8V fully charged) it means it is very close to the 60.000 ERPM, but it should work. I hope, the VESC can handle the power (~400W average and 2000W short peaks at 4S).

I'm eager to test this nice device. I'll probably buy the Hobbyking SK8 (=HW4.10) Version, because it is very cost effective.

Andreas

 

wdaehn
Offline
Last seen: 1 year 4 months ago
Joined: 2017-09-12 17:26
Posts: 65

The Hall sensors of the motor connected to the VESC work very well. I use a LRP Dynamic 8 1600kv motor myself. Quite similar to what you have, it seems. Although it is not a 1:8th car application in my case but a cablecam, I use the parts of such cars.

Encoders start the be interesting in case you want to move your car in the centimeter range. So at extremely low speeds. That would actually make sense for my cablecam but not for an RC car. And even though I use halls, I am okay with the low speed movement I get.

FOC and sensor type is unrelated. You can use FOC even without any sensor.

 

The rest I do not feel qualified to answer.

andreas78
Offline
Last seen: 2 years 6 months ago
Joined: 2018-02-09 12:43
Posts: 7

Thanks wdaehn, that does help a little.

Concerning a low voltage version, I found out, I probably have to change the DRV8302 to the DRV8301 version. Those two have pretty much the same specs, only the DRV8301 needs only a minimum voltage of 6V compared to the DRV8302's 8V. Max voltage is 60V for both.

It seems, what I read in different forums, the FOC mode is a bit unstable and the DRV8302 might fail (especially for high voltage usage up to 12S). Is that also the case for, let's say 4S? Due to the rules in my racing classes, for 8th scale 4S is the limit.

Andreas 

frank
Offline
Last seen: 1 day 14 hours ago
VESC BronzeVESC GoldVESC OriginalVESC Silver
Joined: 2016-12-27 20:19
Posts: 819

Hi Andreas, you want to go for VESC 6 in that case. 4S is possible without issues and 150000 ERPM is tested to work fine. Since you use fast 2 pole inrunners, ERPM is probably the issue with HW 4.xx. Sensors make sense to get the best startup performance. You can define the ERPM limit for sensor usage. After reaching the defined ERPM value, the VESC will not use the sensors any longer. Sensors have their limits at higher ERPMs, and back EMF is more accurate at a certain point. You could also use a AS5047P magnetic sensor for ultimate precision. They are easy to attach to a motor (just glue a magnet to the shaft and place the chip in front of it) and allow to use all sorts of motors that don't feature sensors. I think the VESC 6 is quite good for RC-Car racing. Your competition will not be able to fine tune their cars to the same level. FOC, throttle curves, traction control, nice and very adjustable brakes to name a few features.

Frank

andreas78
Offline
Last seen: 2 years 6 months ago
Joined: 2018-02-09 12:43
Posts: 7

Hey Frank,

thanks for your answer. Unfortunately the all inclusive VESC6 is straining my budget too much right now. Also it is not the smallest device and space is also an issue.  

I bought a TurnigySK8 (=HW4.10). Updating to firmware 3.34 was easy and I already tested some motors at my table. The one for the 8th scale, a Tekin T8 1700KV seems to work nicely, but I haven't tested it in the car yet. Deep winter outside and I'm rather busy at work and there's a family, too. It has 4 poles and will be running just below 60000 ERPM. 

For the smaller 2-poles for 10th scale the ERPM limit shouldn't be much of an issue, either. But they only use 2S batteries, so I'm gonna test the Mini Vesc by Anton Chromjak with some different FETs (logic level). 

I'll keep you guys up to date, but it might take some time...

Andi

arvidb
Offline
Last seen: 3 years 7 months ago
Joined: 2018-01-16 23:09
Posts: 77

Just out of curiosity, how are you going to set the current limits (battery and motor)?

I'm working on an 8th scale RC car with a VESC too. I originally found the VESC because I was looking for a higher-voltage ESC than the normal "4s-6s" ones, to get rid of the frankly quite ridiculous battery currents of about 100 A that's common on a normal 8th scale car. I'm going to try to run mine on 12s 18650 cells (Sony US18650VTC5) and probably limit battery current to 25 A, for 1 kW of power. I've ordered a Keda SKF-4068/5Y (966 rpm/V) motor and will limit motor current to 40 A. (And speed at maybe 28000 rpm since I'm running a VESC4 too.)

I guess you will be running a higher motor current to get the full power out of the car?

andreas78
Offline
Last seen: 2 years 6 months ago
Joined: 2018-02-09 12:43
Posts: 7

Hi,

yes, I will be stuck with 4S due to racing regulations. I'd like the rules to change from a fixed cell count to a total power limit (just like in real motorsports for the WEC (= Endurance Racing). So it would be up to the each driver, how many cells and voltages one is using. But that's another topic and it will probably never happen....

First,I will solder some copper brackets to the drain tabs of the FETs. I already did one for testing and it looks promising. After that I will build an aluminium case, based on a u shaped extrusion profile 45mm wide and have those copper brackets attached to the case with non-conducting heat tape. A fan will cool everything actively. I hope, that this way I can crank up the current limit and have the power and reliability that I need. Quite frankly, the placement of the FETs of the HW4 version is everything but perfect for effective heat dissipation.

With active cooling, the FETs might stand the higher currents. There's still the possibility to change the FETs to more powerful ones in the 40V range.

I don't know if a Vesc can be used with 4S in a racing buggy, but I figured I just give it a try. Due to it being well documented I might be able to repair it by myself, if the magic smoke appears. I just like fooling around a bit.

Andi

arvidb
Offline
Last seen: 3 years 7 months ago
Joined: 2018-01-16 23:09
Posts: 77

Yeah, it's fun to try these things out! I'm wondering how far my 1 kW will get me. I did run the car with the stock setup and that pulled 930 W max, and that was plenty of fun, so I have good hopes. :) Of course, I'm not racing - maybe more is needed for that?

I've tried to model the motors (and batteries) in octave. The speed-torque characteristics becomes quite different with the VESC and its active current limit:

So in theory I'll only get half of the torque of the original motor at low speed (but much better efficiency). I hope it'll be enough to spray some gravel around though. Well, we'll see - I can't wait to try it. :)

andreas78
Offline
Last seen: 2 years 6 months ago
Joined: 2018-02-09 12:43
Posts: 7

Interesting!

For offroad use, I'm not too concerned about lack of power. Real 1KW is plenty. Actually, I'm a die hard nitro guy. But electric buggies become increasingly interesting lately. Right now, on a real dirt offroad track where the grip is rather low, nitro buggies are often quicker. They don't have so much torque down low which makes them easier to drive. Well, that's when using the standard controllers in electric buggies....

So, that's what I'm interested in, real torque control and also better efficiencies.

Andi

arvidb
Offline
Last seen: 3 years 7 months ago
Joined: 2018-01-16 23:09
Posts: 77

Sounds good then! :)

Good luck with your VESC modifications; keep us posted regarding the results.

Are you building your own Mini VESC or have you found a ready-made one somewhere that you will modify?

andreas78
Offline
Last seen: 2 years 6 months ago
Joined: 2018-02-09 12:43
Posts: 7

I'm building my own Mini VESC, the pcbs I ordered from OSH. Let's hope my soldering skills are sufficient.

Andreas

arvidb
Offline
Last seen: 3 years 7 months ago
Joined: 2018-01-16 23:09
Posts: 77

Cool. I've been thinking about doing the same. I built a VESC4 which works fine. But the FETs on the Mini, as well as the FET drivers (LM5113), will all require hot air soldering. So I'm wondering if it's even worth trying without a reflow oven, or at least a proper rework station. (I did the VESC4 DRV chip with a small hot air gun - it worked, but it wasn't ideal.)

s-light
s-light's picture
Offline
Last seen: 4 years 7 months ago
Joined: 2018-03-08 14:57
Posts: 2

@arvidb regarding hot air:
sparkfun (Nate) has a good comparison about 'home-use' SMD soldering technics.
they found that a hot plate / hot skillet worked very good....
if you look to build a pizza-reflow-oven yourself look that your oven has heater elements at top and bottom.
at my last project a 'professional' china oven that only heated from top failed: it burned the leds but they did not reflow.
my pcb was designed to transfer the heat from the led-pads to the back where it will get glued to a heat-sink...

regarding the mini4:
for my next idea/project i will need a small low voltage (2s) & low power (i think i need something in between 2-24W) ESC that will be used in positional control mode and also needs to speed up to ~1800rpm
as fare as i have understand the VSEC FW can do all this :-)
and the mini4 HW seems to be a good starting point too -
i also found a post from antonchromjak regarding low-voltage FETS in the old forum

so if someone have some ideas / news regarding low voltage versions please post theme :-)

sunny greetings
stefan

arvidb
Offline
Last seen: 3 years 7 months ago
Joined: 2018-01-16 23:09
Posts: 77

Thanks for that link stefan, it made me want to go straight off and buy a hot plate. :) I guess the problem with that method is that it will only work on boards with the tricky components on one side?

I have a solder oven project in the works. My first thought was to rebuild a pizza oven (I bought one and started to rebuild it), but realised that it would be very difficult to get even heat with it. So now I'm thinking it would be better to build a hot air oven, using some fireproof insulating board and a fan of some sort. A hot plate would be so much easier though!