Beta CA build for R/C controllers

[Kepler]

Managed to get in a heap of testing with the CA RC controller this weekend. Tested it with quite a few motor and ESC combinations with everything tested working without any issues.

Also tested it with a Turnigy servo speed regulator in between the ESC and CA so that ramp up and ramp down could be individually tuned. With this in place, AGain was set at 300 for quick response to the Regulator with the Reg then doing the rest of ramp work. This worked without an issue and is a very useful addition for swing arm fiction drive designs to smooth out the engagement and disengagement. Probably not needed for direct drive setups though as the AGain tunes the start up ramp quite effectively.

A couple of minor items that could be improved.

Low voltage cut out is only adjustable in whole numbers. This is OK for higher voltages, but RC setups tend to run lower at voltage's and need more resolution on the LVC set point. For example on a 5S setup, 17V is too high to get the most out of the pack and 16V causes the pack to discharge too far. 16.5V LVC suits a 5S pack better.

If an adjustment is made in setup, the power needs to be cycled to the ESC. Unplugging and re-plugging the main CA plug will also reset the ESC. At least then you don't get the battery plug in spark. It would be nice if once you left setup, the ESC would also re set.

I found that most current limit changes also needed a corresponding AGain change. The higher the current limit, the lower the AGain needed to be. This meant that I need to access the Advanced menu every time I made a change to the current setting. Perhaps in the future, current limit settings could also scale AGain at the same time. Alternately Current limit and AGain could both reside next to each other in the normal setup menu.

The above are all very minor and are more of a wish list then a bug report. The LVC is for me the most important item that should be considered.

 

[fechter]

The PI tuning is a real pain with RC controllers. That was the big issue with my analog version.

The latency of the controller makes tuning very difficult. I was thinking of having an asymetrical slew rate for the limiter action. Make it respond fast for overcurrent but respond slow for undercurrent. This way might alleviate some of the oscillation without making the tuning so critical. You need to have the output drop fast when you let off the throttle, but if it ramps up a little slow, that is usually acceptable.

 

[AussieJester]

Great news Kepler i am so keen to hook mine up but due to
me and electrocnis not mixing well together i have been waiting for another
member (Danny Mayes) to come by and "oversee" me hooking up
and setting the parameters correctly on my RC Compatible CA unit
I have to say i wish i had gotten oje of these CA's long long ago
i LOVE the large screen design Justin, the quality looks outstanding
also, if i can get this running today i will be happy as a pig in shit....hurry up and get here Danny
or i might have to do the un-thinkable...break out the soldering iron and start on it myself!!!

KiM

 

[justin_le]

Managed to get in a heap of testing with the CA RC controller this weekend. Tested it with quite a few motor and ESC combinations with everything tested working without any issues.

Good to hear that!

A couple of minor items that could be improved.
Low voltage cut out is only adjustable in whole numbers. This is OK for higher voltages, but RC setups tend to run lower at voltage's and need more resolution on the LVC set point. For example on a 5S setup, 17V is too high to get the most out of the pack and 16V causes the pack to discharge too far. 16.5V LVC suits a 5S pack better.

OK, this is added to the V2.24 beta2 feature list.

If an adjustment is made in setup, the power needs to be cycled to the ESC. Unplugging and re-plugging the main CA plug will also reset the ESC. At least then you don't get the battery plug in spark. It would be nice if once you left setup, the ESC would also re set.

This must be something quite particular with your ESC. When the CA enters the setup menu, then the output channel (either the pulse stream or the analog voltage) are turned off until the setup menu is exited and normal operation resumes. At that point the CA will output the PPM signal again. I'm guessing that this particular ESC sees an interuption in the pulse stream as a fault condition?

Does the fault condition also get reset by power cycling the 5V line to the ESC input rather than having to power cycle from the main battery pack? . If so then it is possible for the CA to power down and then re-enable the 5V bus as well at the end of the setup menu. It would just require the code to then re-initialize the LCD module since that also runs on the same 5V rail.

I found that most current limit changes also needed a corresponding AGain change. The higher the current limit, the lower the AGain needed to be. This meant that I need to access the Advanced menu every time I made a change to the current setting. Perhaps in the future, current limit settings could also scale AGain at the same time. Alternately Current limit and AGain could both reside next to each other in the normal setup menu.

This is curious and should not be the case. The current limit will affect the steadiness with which your command current will hold a particular level, but if the CA is told to maintain 10A, the feedback loop doesn't care if that's because there is a 10% throttle with a 100A limit, or a 50% throttle on a 20A limit. What happened when you increased the current limit and felt that the AGain value was too high?

-Justin

 

[justin_le]

The latency of the controller makes tuning very difficult....

Indeed indeed indeed. In principle, a current limit is about as trivial a control loop as you can get, since there is little in the way of inertia in the system. However, if the ESC has a built in delay or ramping limit between the PPM input signal and what it outputs to the motor, then it becomes a lot trickier for an external device to exert effective control of the system. The lag time between input /output of the ESC will result in overshoot and oscillations and the feedback loop hunting around the set point value.

Its worth setting the ESC's inbuilt ramp to the fastest setting also so the CA inputs will react more quickly and give better control.

Correct, if there is any way to program the ESC's internal ramping then that should be setup to make it as responsive as possible, and then you'll have much more flexibility in the tuning via the CA.

The only minor issue I have is that there is no ramp down on throttle off. For most applications this isn't an issue but for my drive, it benefits from a ramp down on throttle off so it hits the park end stop more softly. I think I should be able to get around this though by using a Turnigy servo travel speed module in the in between the CA and ESC.

A much easier way would be to install a small capacitor between the throttle signal line and Gnd. Though the hall effect throttles can source a lot of current when forcing the signal high, they do not have much ability to pull a signal low, just fractions of a mA. If you wire a capacitor there it would have no effect on your ramp-up rate, but would cause the desired slow ramp-down effect.

Well done Justin. You are onto a winner here and i think it will prove to be a very popular choice for the RC ebike builders out there.
One last comment. Although the large screen is brilliant to see, its physical size does overpower the overall look of the bike. If a bike is purely an ebike then this is not such an issue. However, if you are looking for a more stealth approach as some of these RC drives are looking to do, the huge screen does end up spoiling the effect a little.
My personal vote is for the smaller screen however I fully understand need to try and standardise a product to reduce manufacturing costs. Certainly the large screen wont stop me purchasing further units however I think the feed back is important.

Feedback is indeed always appreciated. On this matter, it seems that about 70% of people really like the large screen, 20% are somewhat indifferent, and about 10% would prefer the smaller option. So overall it was the right decision but I know where you are coming from. If it didn't cost about $10,000 to pay for the tooling of a custom small enclosure box, window, and tactile push button, then we'd happily continue to have both options!

We can still make the small CA boards with no enclosure box for people who want to do their own stealthier housing.

-Justin

 

[adrian_sm]

I found that most current limit changes also needed a corresponding AGain change. The higher the current limit, the lower the AGain needed to be. This meant that I need to access the Advanced menu every time I made a change to the current setting. Perhaps in the future, current limit settings could also scale AGain at the same time. Alternately Current limit and AGain could both reside next to each other in the normal setup menu.

This is curious and should not be the case. The current limit will affect the steadiness with which your command current will hold a particular level, but if the CA is told to maintain 10A, the feedback loop doesn't care if that's because there is a 10% throttle with a 100A limit, or a 50% throttle on a 20A limit. What happened when you increased the current limit and felt that the AGain value was too high?

-Justin

I can confirm this behavior on my setup. When the MaxAmps is increased, throttle output increases much faster, and the system will want to oscillate around the setpoint.

It makes sense to me since the throttle change is proportional to AGain * Error. So if you go from zero throttle to full throttle, having a higher MaxAmps will result in a higher error, therefore require a lower AGain if you want a similar rate of change in throttle out signal.

The important thing for the friction drives (like Keplers and mine) is that it is important to control the rate at which the throttle increases, as this controls how the motor engages with the tire.

But there was also an impact on the stability of the PI loop when changing MaxAmps. Similar to what is described in the manual regarding the loop being setup for 300-600 watt systems, and for higher power systems it will tend to oscillate.

- Adrian

 

[fechter]

My preference has always been to have the deceleration speed as fast as possible for the event you need to do a panic stop. You don't want the motor pushing into an oncoming obstacle.

The Turnigy servo travel speed module would be a good way to test if slowing down the acceleration ramp will prevent oscillations.
If it does, then it should be possible to write some code in the CA that can do this.

 

[AussieJester]

Danny Mayes and myself managed to get the CA hooked up to the hv 160/ Turnigy
one my 2 speed bike, unfortunately its 12.37 at night and pissing rain out side hopefully
will get to give it a whirl tomorrow, lights up and the wheel spins whenz you turn zee throttle'

Thanks again for your help Danny will post those pics we took in my build log thread tomorrow off to bed now

KiM

 

[Ypedal]

welcome to the modern age of display tech AJ !!!... you should have gotten a CA long long time ago !

 

[AussieJester]

Just running through the settings am stumped how i increase the max current setting
past 99amps i need at least 160 ... :-S

KiM

 

[Kepler]

Change the ratio on the shunt to the lower setting might do it.

 

[rodgah]

lower

RangeMode should be in the high range mode if you are using that big fat shunt. Setting it in high should also take that restriction away I think because thats about all I changed and I had no limitation on the amps. Also set your Rshunt value so that it reads half accurately.

edit....Rshunt = 0.5mOhm for the big external one.

 

[Kepler]

Well I figured it was in the shunt setting anyway

 

[AussieJester]

Cheers fellas got that sorted, i am using only the small shunt though as my peaks are about 160amp breifly with the HV160 on lowest settings... TBH the shunt looks near identical to the one out of my Turnigy meter i have been using till now... most of the time riding is well under 100amp so not concerned about melting it LoL...

I have noticed similar finding to you Rodger, there isn't a slow start up it turn the throttleslightly and it revs its head off :-S this is obviously not aqcceptable for my own use and will be unable to ride it until its rectified, hopefully i can improve this with some tinkering with settings?

i have been reading and re-reading the instructions eventually it might all sink in...Still rainy outside so unable to take it for a spin yet & i wont be able to until i have another person here anyway, i have no wheelchair tow hitch on the bike and i am not about to get stranded from home with non wheelhair again...(happened once never again) thus far aside from the look of the CA, i'm un-impressed with the difficulty in setting it up, I new this would be the case going in just confirming it now haha Definitely not 'plug and play' and will need a heap of tuning by the looks, a servo tester and bec and Turnigy meter is a shit load easier to instal as it works straight away without the need for the messing with settings definately wouldn't recommend this CA for those like me, that struggle with this sort of stuff, again i knew this going in not complaining just confirming haha... OH, why does the RC CA have default settings suited to 600watt setups? does a 600watt rc setup even exist, they don't start to get interesting till after 4000watts miniumum :-|

... Shall persist with it, very easy to revert back to older setup in 10 minutes as i made that all plug and play which at this point seems more likely until more have these units and i can copy others settings, prolly was unwise of me jumping straight in i guess but
hind sights a wonderful thing... ...

Cheers for help guys im sure i will be back with more questions sooner rather than later...

kiM

 

[adrian_sm]

I have noticed similar finding to you Rodger, there isn't a slow start up it turn the throttleslightly and it revs its head off :-S this is obviously not aqcceptable for my own use and will be unable to ride it until its rectified, hopefully i can improve this with some tinkering with settings?

Hey Kim.

Remember that the CA is now trying to set a current level, not a motor speed. So if the bike wheel is not loaded, it will keep ramping up the throttle trying to get the motor to draw the Amps requested by the throttle position. Hence the motor reving its head off.

Try dropping the AGain setting real low if you want to slow down how quickly the motor spools up. Then keep increasing AGain until you are happy with the throttle response. But unfortunately this really needs to happen with the bike on the road.

Hope the rains ease up soon.

Cheers, Adrian

 

[rodgah]

hey kim since you have it on a bike try this for me......with the back wheel off the ground of course...


even though it will go flat out, in theory 5% throttle should mean 5% amps of the max setting you input.....so:

set 50 amps, open the throttle ever so slightly to just get a reaction, apply the brake and see how much force it takes to slow the wheel down, im guessing not very much.

though i do agree getting someone else to test it is the best option. LMFAO at it being hard to install.....you had to plug it in and change a few settings, lucky you never got a sevcon :lol:

 

[Kepler]

Dont forget to setup ESC for fast throttle response so the CA has something to work with.

I think you will be pleasently surpized once you get use to it. I certainly was. Of cause I was only using a lowly 600W

 

[AussieJester]

Dont forget to setup ESC for fast throttle response so the CA has something to work with.

Cheers for that Kepler shall do

I think you will be pleasently surpized once you get use to it. I certainly was. Of cause I was only using a lowly 600W

I believe your right matey, thanks to Rodgah's patience with my retarded short tempered self, i now have a working Rc Cycle Analyst, still some minor tweaking to get it 100% but i have very very slightly better understanding of things only & running miserley 100amps max atm but, after 5 hours of changing and testing settings with Rodgahs input via MSN this afternoon we got it functional. I will post the exact settings for the CA with my type of combo first thing tomorrow before i start the permanent install, Danny and myself last night 'jury' rigged everything just to ensure it was working, so i now need to redo all the wiring and incorporate a few connectors etc so everything is modular and easily changed in event of failure on the road, i can carry the old servo tester in my tool kit, aside from a motor and throttle i can change the rest road side if it breaks now :: fingers crossed:: this will not ever need to be done...

Thanks again to you Roger, and to you Mr Kepler SiR for your input and patience with me And of course Danny Mayes for his input and guidance with the wiring last night, last but by no means least the guru Justin for this fine product. Bending, cutting and welding steel is sooooo much easier for me than this electrical business :lol: maybe one day i'll get it i have always found it difficult to learn anything i have very little interest in, bad trait of mine i know :-(

KiM

 

[Ypedal]

You are not alone man... .. i get by.. like a monkey i can see and do... but daaamn some of this electrical stuff gets to me some days lol.

 

[justin_le]

I believe your right matey, thanks to Rodgah's patience with my retarded short tempered self, i now have a working Rc Cycle Analyst, still some minor tweaking to get it 100% but i have very very slightly better understanding of things

Hey KiM and glad to hear this. I should have put a bigger forewarning perhaps about what being part and parcel of a "Beta Tester" means!

The plug-and-play that I alluded to was assuming you got the molded shunt which we had terminated appropriately, sorry if that wasn't clear enough. The idea was that ebike throttle plugs straight into CA, CA plugs into special RC shunt harness, and RC shunt harness plugs into the battery pack and the servo input of RC ESC. We're looking at retooling the shunt molds over the fall and in the process we could make a version with 10 AWG wire that would be suitable for 100A setups, so then at least the more power hungry types like yourself could have a similar solution without all the custom wiring.

-Justin

 

[AussieJester]

Absolutely no issue with the product Justin I was fully aware going in there would be some jigging about to set up, don't mind me buddy I'm just easily frustrated with anything electrical, I wish I had 1/1000 the knowledge you and Luke have with electronics, never the less thanks to ES and the support of its members i usually bumble me way through these things hehe..you product is 100% not the problem here I just want to make sure this is clear with everyone.

Shall drag me ass out of bed soon and make a couple small tweaks and report back with final setup

KiM

 

[Danny Mayes]

Hey AJ,

i am delighted to hear that you are 99% of the way there mate. The thought of getting so close with the mega elegant solution of the RC CA and having to pull it all off again and revert to the old set up, i did not want to confront

Much happier now it is panning out and paying off.

D

 

[Kepler]

I plan to fit a button (or buttons) throttle to the CA. I have a nice compact 3 button unit that I plan to adapt. The setup uses tactile buttons arranged with one large button and 2 small buttons above the large button. Its small enough to basically fit on anything, even road bike bars. So the plan is that the large button is the "main power"button and will proved 2/3s power. One of the top buttons will become "boost power" or the equivalent to 100% throttle and the the other small button will be "low power" or 1/3 power. Current limit will be around 800W.

Circuit is a simple voltage divider designed to mimic a 1-4V signal from a Hall throttle. Hopefully should work OK.

 

[AussieJester]

Cheers Danny, even better news to report this eening after Rogers guidance today
from 10am till 5pm this arvo we played with different settings while riding the bike.
...We have it very very close to be on par with a servo tester throttle now
one last thing to do, The problem is as Rodgah stated it to me "....from 0.065 to 0.1 is a very tiny movement of the throttle....not linear compared to from 0.2 thru to 4.0v" so a lil mechanically adujustment of the the pot to increase the voltage slightly and then adjusting the Aux Threshold settings accordingly on the on the CA should :: fingers crossed :: fix the last lil issue with rapid surging off the start with only extremely minimal throttle movement, Will also definitely fix the ~3-5% dead spot at the beginning of the throttle movement ...,I found out today the front wheel DOES come off the ground under motor power alone haha scraed the crap out of me wasnt expecting, Rodgah acted surprised to when i told him but secretly, i think he new it would do it and was doing it for a laugh

I have been recording the best setings we have come too for the mofo 80-100 130kv Turnigy/CC HV160 and rc CA setup will post these soon as we have it dialed in 100% as is now i would be happy to keep it as is if
this is the best we can get it as it is extremly rideable on the road, i'm not confident it would go so well low speed technical riding in the bush, you would likely end up running straight into a tree or rock if you bumped the throttle 2-3mm at the wrong time...with these lil throttle tweaks will be spot on and its simple mod to do with the Magura done it twice already for the other bikes haha...

Very clever idea there Kepler SiR :wink: , you could go one step further and have a lil remote control
to change the settings to oooh say a 200watt that would be unique idea now wouldn't it :: wink wink ::
...

I would like to finish the adjustments tonight but been at it all day legs up and head back now
i think it will be it for today i'm beat...

More impressed than ever Justin you ARE da BoMB

KiM

 

[Danny Mayes]

Great to hear it AJ.

Cant wait to have a test ride