Beta CA build for R/C controllers

[Kepler]

The holes are there for the header pins on the CA board. The PICkit-3 has the header pins protruding from it. Just stick the header pins in to holes on the board. Keep a bit of sideway pressure on the programmer for good contact. Too easy.

 

[Kepler]

OK, Testing has been done. Tested the CA on 2 different types of ESC and pushed it though every sort of throttle up / throttle down combo I could think off. CA didn't miss a beat. No surge issues at all and had no trouble getting a nice smooth acceleration curve on every throttle up.

However one new issue has shown up. The voltage calibration is now out by 6V. Checked this on 2 updated CA's. Both had the same problem. I have emailed Justin with the above and are confident it will be a simple fix to sort out.

All in all, very pleased with the CA's control. 8)

 

[kevo]

As user of Kepler's and Adrian's friction drives and Justin's CA I totally look appreciate the awesome tech support provided by all of you!

 

[justin_le]

OK, Testing has been done. Tested the CA on 2 different types of ESC and pushed it though every sort of throttle up / throttle down combo I could think off. CA didn't miss a beat. No surge issues at all and had no trouble getting a nice smooth acceleration curve on every throttle up.

Yipee! Thanks Kepler for having the first stab of properly field testing this and verifying that it did the fix OK. I was pretty sure looking at what was going on with the oscilloscope, but there's no substitute for testing in a real system.

However one new issue has shown up. The voltage calibration is now out by 6V. Checked this on 2 updated CA's. Both had the same problem. I have emailed Justin with the above and are confident it will be a simple fix to sort out.

I sent you firmware that had a voltage calibration value pre-loaded and that skips the normal calibration routine. You can make this match exactly by going into the advanced setup menu and changing the VoltSense scaling parameter in V/V. The displayed voltage varies linearly with the scaling parameter here, so scale by the same ratio of your display voltage to actual voltage, and it will be spot on again.

-Justin

 

[Kepler]

No problems Justin.

I figured the voltage calibration issue would be a minor adjustment.

If anyone on my side of the world needs their CA flashed, shoot me a PM so we can make arrangements to get it updated.

 

[H3D]

Kepler,

I wonder if the firmware upgrade fixes the problem I have with version B3. Just before I hit max rpm (about 90-98% of it), the motor/wholesystem surges in an oscillatory manner with a period of about 3 Hz (kinda shakes). Just at that throttle position (and any gearing, so doesn't really depend on the load, it's just a little more obvious at heavy load - like 40mph on a calibrated trainer). Throttle at max (2ms output) or just somewhat below is fine. Nothing like this with the servotester.

Also, my ebike.ca hall throttle outputs .82V (off) to 4V (full on). I can set the auxthreshold to 0.9 and the startpoint is fine. But it only takes 0.6 volts of throttle (like up to ~1.5) to get to full output. That's not much and makes it really twitchy of course. Is there a CA setup way to expand the throttle input action range to like 3Volts (.9 to 3.9, say)?

Short of a software fix, I divided down the input throttle voltage with a couple resistors and lowered auxthreshold accordingly. Still, the max-rpm point on the throttle depends on the load (i.e. 1'rst gear has a narrow throttle range start to maxrpm, and 9'th gear has a full range (as set by the resistor divider). I guess that's kind of how it is on a car throttle, but it's quite different from servotester and will take some getting used to.

Thanks for any and all assistance,

H3D

 

[H3D]

Answered my own question,
"Is there a CA setup way to expand the throttle input action range to like 3Volts (.9 to 3.9, say)?"

Didn't notice the full throttle range is essentially zero to Imax. I had Imax (50A) way too high for my trainer, even in 9'th gear, where it takes about 25A to go ~45mph). So in 1'rst gear on the trainer (~15mph top) where it takes ~6A, the throttle range from off to fullrpm was ~1/8 the total thumb twist angle, and that felt really funny (but now that I know what's going on, I can deal with it).

H3D

 

[johnrobholmes]

Got mine in hand, and have some suggestions for production

...

However, the wiring inside on the 6 conductor cable would be slightly different. The yellow wire for getting the speed from the motor hall signal (which is usually not connected in the CA-DPS) would be instead attached to the 5V bus of the CA, as the RC controllers seem to require 5V on the servo input line in order to work. And the green throttle over-ride wire would naturally be soldered to the PWM pad rather than the Th pad:
View attachment 1
...

Many ESCs output between 5.5v and 6v on the RX positive wire. None that I am aware of require power into this line. I would suggest the yellow wire is left for connection to a hall signal instead of using it as a 5v rail. You would probably get many damaged units otherwise. The RC ESC can work properly with only ground and signal hooked up through the regular RX signal wires. Positive RX from the ESC can be left floating with no detriment, I have to do this frequently on my crawlers to prevent ESC damage from external servo power sources. You can even float the ground too and have only signal, but then ground reference starts to move around under load. Normally to no ill effect, but it is good practice to have as many grounds tied together as possible!

For those who've played with and dissected RC controllers, do they typically have a visible or accessible shunt resistor you can see on the ground return line for the current sense? Even if not it wouldn't be too hard to put a ~1mOhm shunt resistor right where the - battery lead connects to the board and use that.

Justin

I know of only one R/C controller with an internal shunt, the ICE series from Castle. They are a pain to reach on the heatsinked versions if I recall correctly. I will take one apart when I get home and see how easy we could access it.

 

[justin_le]

...

However, the wiring inside on the 6 conductor cable would be slightly different. The yellow wire for getting the speed from the motor hall signal (which is usually not connected in the CA-DPS) would be instead attached to the 5V bus of the CA, as the RC controllers seem to require 5V on the servo input line in order to work.
...

Many ESCs output between 5.5v and 6v on the RX positive wire. None that I am aware of require power into this line.

OK Interesting, I'd much prefer having the yellow wire attached to the Sp pad on the CA because that also keeps things consistent too, but was left with the impression that some of the RC ESCs needed the 5V present in order to operate.

I know of only one R/C controller with an internal shunt, the ICE series from Castle. They are a pain to reach on the heatsinked versions if I recall correctly. I will take one apart when I get home and see how easy we could access it.

It certainly is handy to just reuse an existing shunt if there is one present so let us know how this goes. I've also been working on getting a more refined inline shunt tooled up that would really lend itself to splicing inline with the battery lead wire and then covering in shrinkwrap. That ends up being pretty tidy too.

 

[lostrack]

Just flashed my CA. Took less than 5 seconds to write. All default settings to read/write worked. Couldn't be easier.

v2.24 B4

 

[lostrack]

Can anyone help me with this?

The throttle isn't working. Is this wired up correctly?

 

[lostrack]

all done (*edit*)

 

[www.recumbents.com]

Who in the USA can upgrade the microcode on these to the beta 4 release? How much?

Getting tired of the beta3 lurch...

Thanks,

-Warren.

 

[mdd0127]

I'd like to get mine reprogrammed too. I'll probably be buying the pic programmer pretty soon so if you're still needing someone to reflash yours by the time I'm ready, I'd be happy to do it for free.

 

[www.recumbents.com]

Sounds good, thanks. Please PM me when you are ready.

 

[panurge]

Hi guys

I've reflashed mine just few minutes ago....the flash was really easy and quick, everything seem to be ok, this is a full description for who is going to do it himself and has not great E-skills....as me^_^.

The Pickit3 has female connectors as the CA so a header pins strip is needed that usually come in single 6 or 40 units, 'cause the Ca only has 5 headers, with a standard 6 pin header you need to cut out one of them that is not used by the Pickit, indeed. The Physical process as for Kepler description is easy, just mount the headers to the Pickit and hold it softly to the CA board with some inclination...

Using the MPLAB software, I've just selected the correct microchip model and verified in the supported programmer list that mine is allowed to flash this chip, and that the 5v output needed for the CA is present.
Than connected and recognized the CA, erased the old firmware and imported the new one.... success....that's all.

Rebooting it, the correct new firmware is shown and a subsequent blank screen appear....rebooting holding a button and re-entering the original Advanced setup values (Vsense Rshunt ecc), the main screen was now showing the correct read (with good precision, indeed).

I've made a small video during the re-flash, feel free to PM for it.

Thanks Justin!!!!!!!!!

Looking for a V3 with small screen soon if available, I think.

Regards

Jules

 

[justin_le]

Looking for a V3 with small screen soon if available, I think.
Jules

Hey Jules and good work on your first run-though reprogramming the micro.

Chances are that we will re-introduce an option of the small screen devices with the V3 production run, but as a board-only unit without enclosure. So for those who have the means/ability/desire to piece together their own housing, then the small screen will again be a available.

It's rather the inverse of what we had before, where the small screen came in a enclosure and the large display units were just the board+LCD.

-Justin

 

[panurge]

Hey Jules and good work on your first run-though reprogramming the micro.

Chances are that we will re-introduce an option of the small screen devices with the V3 production run, but as a board-only unit without enclosure. So for those who have the means/ability/desire to piece together their own housing, then the small screen will again be a available.

It's rather the inverse of what we had before, where the small screen came in a enclosure and the large display units were just the board+LCD.

-Justin

Thanks, Justin, was really an easy operation to do, due to how simple and great your unit is. Also I can now well understand the need for a bigger chip....there's no space even for one more digit in the V2 .

Good to hear that The small screen V3 could be an option, even if without the case....this should be perfect for my (slow) project, indeed.

Thanks for all, Justin.

Jules

 

[sn0wchyld]

Looking for a V3 with small screen soon if available, I think.
Jules

Hey Jules and good work on your first run-though reprogramming the micro.

Chances are that we will re-introduce an option of the small screen devices with the V3 production run, but as a board-only unit without enclosure. So for those who have the means/ability/desire to piece together their own housing, then the small screen will again be a available.

It's rather the inverse of what we had before, where the small screen came in a enclosure and the large display units were just the board+LCD.

-Justin

is it at all posible to get the normal screen unit sent with the screen separate? Other than the wiring it might be useful to be able to mount the screen separate from the 'brain'...

 

[justin_le]

is it at all posible to get the normal screen unit sent with the screen separate? Other than the wiring it might be useful to be able to mount the screen separate from the 'brain'...

In principle but there are a few issues with that. You also can't run very long wires between the board to one of these LCD modules (learned that doing robotics at university), more than a few feet and noise issues will mess up the signal transmission. Plus since the CA's circuitboard has a smaller footprint than the LCD module there is little reason not to have them piggy backed. That way it is just 4-6 wires to connect CA to the rest of the system, while you'd need 12 wires to hook up the LCD module remotely.

Other complication is that we need the screen attached to do all of the calibration and testing, and the pins don't stick out quite far enough to use the normal single row female header on the LCD module. That could be sorted, but as it now the LCD screen has to get soldered on before we program/calibrate/verify the device.

-Justin

 

[Thud]

I have a quick question:
Is it possible to drive 2 speed controllers of from a single CA iwth a simple "y" conector? (as i do for my standard set ups)
I think the bully would benifit from this if its feasable.

 

[Bluefang]

Hey guys, I have one of the RC CAs that i bought with another part from FFR trikes and i am unsure if i have wired it correctly. Can you please have a quick look at the pictures and tell me if i have something wrong.

I went thru the advanced menu for the CA and it seemed to be all setup for the RC option, changed the start up speed to 0km/h. And it does not seem to want to run the motor.

When i turn on the power to everything, i get the normal startup tone and then a pause, then 2 loud beeps. From there i get nothing when using the throttle. The ESC is a cheap generic 200amp 6S from hobbyking and the motor is a Aelion 63-74 sized one, will be running as a friction drive.

Here's to hoping its just my stupid errors rather then a dud ESC or something cause i really want to play with a CA as i have never used one or any metering device on any of my bikes

*edit* Tried changing the controller to my CC160 i have, it does not run the motor when everything is hooked up and after afew minutes it emits 5 beeps which seems to mean signal interference. Also when i was testing it with out wiring everything together properly, i moved it at one point and the motor ran(no throttle active), i got excited and disconnected everything so i could go wire it properly to see if it would work then but it did not. So i think i knocked something loose but i have no idea what.

Thanks,
Derek

 

[Bluefang]

Yahooooo

Finally got it all working, turns out i am not stupid and the RC CA is a great product.....Just my cheap hall throttle was somehow buggered. Wired in a new one that i just got to replace a damaged one on my main bike and it worked straight away. To late to play with properly but so much happier now everything works. Good thing is i can now go back to my cheapy ESC as its a low powered system rather then using up my good CC160.

 

[fechter]

I have a quick question:
Is it possible to drive 2 speed controllers of from a single CA iwth a simple "y" conector? (as i do for my standard set ups)
I think the bully would benifit from this if its feasable.

It should work but keep the connection between the two controller battery negative wires as short as practical to minimize any voltage difference between them.

 

[Bluefang]

I have been thinking about a similar thing to what Thud asked, the 2 ways i see it "working" the motors would have to be perfectly balanced much like a set of parallel brushed motors. Ie you limit the system to 400amps, but 1 motor could draw 300 out of that 400 and end up killing itself. The other option would be if you could run 2 separate parallel shunts(assuming its possible) leading into each motor controller, but the same thing would happen as one would hit the cap of say 200amp and the other could only be at 100amp but both would have their throttles cut to limit the power draw.

Only way it have it working "properly" would be to buy 2 CAs. But with the large number of motors thud has to chose from he may be able to find a couple that are balanced really well

Just my 2c