I think my SP5 controller bit the dust!

[JJ]

I just got my GM cadence sensor hooked up last Tuesday. Max speed on a electric assisted bicycle in Minnesota is 20mph (32 kph), so I had previously adjusted the max speed using my USB cable and Bluetooth dongle. On a ride last Thursday, using the bluetooth dongle, I was able to adjust the speed down a little more, modify the ramp-up speed and make sure the pedelec function was turned on, etc. Everything was working, but not as well as it should. I was planning on tweaking the pedelec functions on my Android device on the way home from a rest stop.

Then, it stopped connecting with my phone, so I rode home about 18 miles with the PAS set on full (default, unless controlled by the display or the bluetooth software). I found it was turned on by default (not by pushing the CC button twice, like the instructions say) and that I could pedal up to the speed I wanted to ride at, push the CC button and it would hold there. I wasn't sure I could make it all the way home at 20 mph, so I set it at about 16-17mph, and got home with no more issues (47 mile ride, about 75 kilometers). Once at home, while charging the battery, I found I could connect with BT again, so I thought probably some burglar alarm or some other weird phenomenom was at work preventing connection. This has happened before in approximately the same spot.

So yesterday I stopped about 7 miles into my ride to have a drink and tweak the pedelec functions using my Android phone. But after braking to a stop, the battery lights on the throttle went out. I checked the connector, it was OK, I turned the battery off and back on -- no joy, I disconnected the brake levers and cruise control, still not working, and moreover the power display on my battery only flickered once when pushed. The bluetooth connection was lost again, and when I got home I couldn't even get the USB program to connect to the controller (again, disconnecting everything to check it isn't a brake lever or malfunctioning throttle). However, after sitting overnight, when I switched on the battery, the throttle lights came back on & the power display on the battery worked again.

The last thing I did before it went out was brake to a stop. That's the last thing I did back in July before the controller went out the first time, only that time it shuddered and vibrated something awful as I came to a stop, and the lights on the throttle didn't go out. Gary Salo suggested a bad controller, Luna sent me one and I've been riding without problems until now.

Am I up the proverbial sewage filled water way without a means of propulsion??

[Bikemad]

Hi JJ andto the forum.

If the lights on the throttle have gone out, it is not getting battery voltage to the controller's harness.

This could be due to one of the following:

• A poor connection or break in one of the battery wires/connectors (gently wriggle all battery wires to see if the throttle lights come on again)
• A faulty battery BMS
• Bad contacts on the battery switch
• A direct short within the controller causing the battery's BMS to disconnect the power

I suggest you start by measuring the voltage coming from the battery with the controller disconnected to make sure the problem is not a bad connection inside the battery. (Check out this thread for further information.)

If the battery voltage is reading correctly, then check it again with the controller reconnected and see if it drops significantly.

Alan
 

[JJ]

Hi, Alan. that was quicker than I expected, especially on a Sunday!

Like I said, the throttle lights and battery power indicator are working today. I suppose maybe the BMS shut down yesterday and was reset through something I did last night, like, for instance, charging the battery up to full capacity, or turning the switch on & off a few times while testing it. However, the motor still doesn't work, and I still can't talk to the controller with either Bluetooth or USB cable.

The battery measured 58.3V on my multimeter this morning (52V Luna Bottle). I don't have an easy way to check with the controller connected, since all my connections are shrink wrapped for waterproofing, but after trying over and over again to connect, the voltage is still at 58.3 this evening, so in all the testing I did today, power's getting to the controller, and back out to the throttle LEDs, but the controller hasn't used any power to speak of.

[Bikemad]

If you still have all leds lit on the throttle, it is obviously receiving full battery voltage and the controller should work if it has not been damaged and there are no other wiring faults on the wiring harness.

Check that there are no short circuits between the red +5V supply wire and the black ground on the pedelec sensor connection and also check that you have at least 4V between the +5V supply and ground.

Have you tried unplugging both of the brake switch connectors now the battery gauge LEDs are lit up again?

As you seem to have experienced problems directly after braking on at least two occasions, I'm wondering whether the BMS of the Lunar battery is unable to accept the amount of regen current, causing it to disconnecting the battery while it is still receiving a high amount of charging current.
An instant disconnection of the battery while it is receiving high regen current (initiated by either the BMS or a poor connection on the battery supply cables, connectors, or switch) could cause problems for the controller, as there there would suddenly be nothing to absorb the excess current and voltage being produced by the regen.

I also discovered a video on YouTube of a problematic Lunar 52V bottle pack which was assumed to be caused by a shorted wire inside the pack.

Power's getting to the controller, and back out to the throttle LEDs, but the controller hasn't used any power to speak of.

The battery power going to the throttle LEDs does not come from within the controller itself, it is actually supplied direct from the battery via the multi junction block on the controller's external harness. 

Alan
 

[JJ]

No short circuits on the pedelec cable, and again, I' don't know how to test shrink wrapped crimped connections for voltage. How do you suggest I do that? Besides, what possible benefit would I get from finding if there is at least 4V between the +5V and ground???

Have you tried unplugging both of the brake switch connectors now the battery gauge LEDs are lit up again? 

Yes, I tried unplugging everything -- again. And again when I tried with a different throttle.

As you seem to have experienced problems directly after braking on at least two occasions, I'm wondering whether the BMS of the Lunar battery is unable to accept the amount of regen current, causing it to disconnecting the battery while it is still receiving a high amount of charging current.
An instant disconnection of the battery while it is receiving high regen current (initiated by either the BMS or a poor connection on the battery supply cables, connectors, or switch) could cause problems for the controller, as there there would suddenly be nothing to absorb the excess current and voltage being produced by the regen.

I read on endless sphere about problems GM was having with regen back in 2008-9, IIRC. A high amount of regen coming into a basically full battery needs to be shunted off somewhere, and GM supposedly fixed that problem before coming out with the MP3. So, my SP5 should be able to deal with it, and has on every occasion except for possibly these two. It shouldn't be a BMS issue, IMO.

I took the top and bottom off my battery (and opened the carrier) to check for loose or poor connections, but didn't find anything (that doesn't mean there isn't a problem, just nothing obvious). The maker of the video you linked to supposed his problem was loose connections (in the comments), but was unable to actually find any problems, either. His bottle case and carrier are different from mine, however.

The battery power going to the throttle LEDs does not come from within the controller itself, it is actually supplied direct from the battery via the multi junction block on the controller's external harness. 

Since the 5V is split off in that junction you mention & goes every which direction from there (including into what you are calling the controller), it seems to me to be a part of the whole. Since the cable cannot be removed from the controller without physically breaking the controller, I sorta lump the cable, junction and controller together in my mind. You can't buy them separate, either. Just my point of view, I guess.

I have used this battery with a new TSDZ2 torque sensing mid-motor as well, and haven't had any problems with it, but that motor doesn't have regen, a throttle or cruise control nor does it need brake cutoffs. To me it seems to be a Golden Motor (controller/cable/brake lever/throttle/etc.) issue, not a battery issue, but that may be incorrect.

JJ

[Bikemad]

No short circuits on the pedelec cable, and again, I' don't know how to test shrink wrapped crimped connections for voltage. How do you suggest I do that? Besides, what possible benefit would I get from finding if there is at least 4V between the +5V and ground???

You can use a couple of dressmaking pins or needles to stick into the wires and then touch the meter probes against the pins/needles to check the voltage.

Alternatively, you can carefully measure the voltage between pins 4 and 2 on the five pin motor harness connector (the one used for the Bluetooth dongle or USB programming lead) shown in the following diagram:



I place small pieces of plastic drinking straw over each pin to prevent accidentally shorting the meter probes against adjacent pins. 

If the +5v regulator inside the controller is working correctly you should have a reading of at least 4V between the +5V and ground  If there is no voltage (or very low voltage) it means the +5v output from the controller is either faulty or partially shorted to ground somewhere within the harness.
If the reading is still very low with the control harness unplugged from the motor harness, then the +5V regulator inside the controller is probably faulty.
However, if the reading jumps up to over 4V with the control harness unplugged from the motor harness, then the control harness itself (or one of the controls attached to it) will be causing the problem.

If the voltage on the +5V feed is too low, the controller (and the Hall sensors in the motor) will not operate, and there have been several reports of controllers suffering from problems with the +5V regulated output, therefore I would check it out first.

Unfortunately, the controller will probably need to be replaced if its +5V supply has failed.

On the other hand, having the correct voltage reading on the +5V wire does not necessarily imply that the controller will still work, as there are many other components within the controller that can also prevent it from working. 

Alan
 

[JJ]

Thanks for the explanation, Alan!

Since none of the wires is available to put a pin in, I went with using the five pin motor harness connector. First I did a resistance test with the battery power off, and there was no continuity (no short between +5V and ground). Then with the power on I got a pretty consistent 1.5 volts between pins 2 & 4 with the control harness connected. I then disconnected it and got right at +4v. Connected again, I tested the brake cables, the CC, and the throttle, and only disconnecting the thumb throttle made any difference, showing +4V again at the five pin motor harness connector. It also showed +4V with my twist-grip throttle connected, so there is probably an issue with the thumb throttle.

With the throttle cable disconnected or with my other throttle connected, the Android program on my phone still cannot see the bluetooth signal from the motor, and with the USB cable, I still get a timeout on the PI-800.exe program on my computer.

My conclusion, based on what you've written and my results is that my thumb throttle may be whacked, but replacing it still doesn't solve the problem of the motor not working. So, probably an internal controller issue, eh?

[Bikemad]

Hi JJ,

Either something has damaged the Hall sensor in your thumb throttle causing it to now overload the +5V supply, or the +5V supply may have been damaged as a result of a failure (or a momentary high voltage surge) within the controller, and is now too weak to power the throttle.

Here are some empirical results obtained from a simple test I conducted on a good controller's +5V output five months ago:

P.S.  I have just checked one of my controllers to see what voltage readings you should expect to see:
The +5V supply was reading a constant ~4.9V with the throttle and Bluetooth dongle disconnected.
When I plug just the throttle in, it drops to 4.23V.
When I plug just the Bluetooth dongle in, it drops to 4.52V.
When both units are plugged in, it drops to 4.22V

As your motor doesn't run with the twist throttle (even with ~4V on the +5V supply) it is quite possible that the controller may have sustained additional internal damage. 

It might also be worth checking the Hall sensors in the motor to confirm that they are still working correctly (when the controller is removed) as they are fed from the same +5V supply as the throttle, Bluetooth dongle and Pedelec sensor. The +5V in the brake switch wires are not normally used (but if you also have the GM hydraulic Brake levers with built in switches, then they might be.)
Take a look at this post and this video for more details on testing the operation of the hall sensors.

I suggest you carefully check the unused headlamp and horn wires on the control harness to ensure that none of the insulation on the four unused wires has shrunk back and exposed the copper wires, allowing them to touch each other.
When this happens, a direct short can easily occur between the battery supply to the headlamp/horn and the ground wires if the horn button or lighting switch is pressed while the wires are touching.
If shorted, the thin wires will get very hot very quickly and can melt through the insulation on adjacent wires in either the throttle/horn and/or the main control harness.

If there is any sign of the four unused wires touching each other, I suggest that you perform a thorough continuity test between all of the wires on the control harness to make sure there are no short circuits before reconnecting it to a replacement controller. 

Unfortunately, a short circuit with full battery voltage can do a lot of damage very quickly, and the damage is not always obvious.

The black and blue wires for the headlamp and horn ground connections will obviously show continuity as they are both connected to the same common ground circuit as the throttle unit and brake switches.

Alan
 

[JJ]

I've removed the Smart Pie system from the bike and done the following:

It might also be worth checking the Hall sensors in the motor to confirm that they are still working correctly (when the controller is removed) as they are fed from the same +5V supply as the throttle, Bluetooth dongle and Pedelec sensor.

My multimeter shows .3V at the Hall sensor 5V pin with control harness connected or disconnected. Without the 5V it seems hard to tell, however when I called Luna, the tech there had me check for cogging, which he said might indicate the Hall sensors were out. Wait, maybe he said if it wasn't cogging... I don't know. It isn't cogging, for what that's worth.

I suggest you carefully check the unused headlamp and horn wires on the control harness to ensure that none of the insulation on the four unused wires has shrunk back and exposed the copper wires, allowing them to touch each other.

The insulation has shrunk back (they're aluminum wires, BTW), but the silicone sealant or whatever was covering it was keeping any shorts in check. There is no sign of overheating, and I've done a continuity check. So, I guess it's time to order a new controller.

JJ