Erider torque voltage output consistency issue

[onitubai]

I finally have a fix that's worked for hundreds of miles. Following the experiences of others in this thread taking steps to mediate electromagnetic noise picked up by the signal wiring, I took a different approach by looking for potential sources of that noise.

The most accessible suspect is a 50cm cable run between the battery and controller. The cable that came with the rest of the kit is unshielded, with parallel conductors. Twisted wire sets and/or shielding should be the default for any system handling both high loads and signal traffic. That cable went straight to the scrap pile. I replaced it with a pair of 10 AWG wires wound up by hand as a twisted pair.

The next suspect is the motor cable. It looks a lot like the phase wires are parallel and unshielded. Before going to the substantial effort of fixing that, I temporarily re-routed it through space, away from the frame as much as possible.

The PAS has worked perfectly ever since. I eventually moved the motor cable back against the frame, and that hasn't caused any trouble with the PAS. Twisted pair battery wiring mediated enough electromagnetic noise to solve my PAS issues. No more PAS crashes, no more wacky watt readings, no more feedback effects.100 watts of torque signal doesn't randomly turn into full throttle thanks to power crosstalk amplifying the torque signal in a feedback loop. Pedal assist stops when pedaling stops, not seconds later. PAS is actually safe to use now in traffic or while riding near other people.

I'll probably replace the PAS extension with proper shielded wiring eventually, but first I want to kill the rest of the noise sources.The gmac motor cable is next, because the speedometer and thermistor signals go howling all over the place in a pattern consistent with motor speed and load.

 

[zorbas]

Hello, I read this thread because I have a similar issue with my set up. I connected the -ve side of the battery to the frame. And now as soon as press the throttle, I have a fault with the phaserunner. It says "Fault 9 Instantaneous phase over current".
Did I do something wrong on the wiring? I did a Y split as explained here. I soldered a wire on the negative side of the battery and connected it to the frame.

The issue is resolved by disconnecting the wire of the frame and restarting the system. Nothing is damaged but I really want to solve the original issue output of Erider torque voltage inconstistency.
I am checking here before contacting Grintech support as maybe one of you sees exactly what happens.
I have a 48V battery, a phaserunner and a GMAC with my erider.

 

[mkessy]

I think I'm having a similar issue as zorbas. I've attached a picture of my ground wiring attempt. Another issue I've noticed is when I pedal, even with the system off. I can see flickering in the CA as if it's recieving a power signal of some sort (I was testing at night). Should I make a ground loop or is this y-splitter adequate? I'm thinking maybe I need to separate some of the wiring as everything is pretty bunched together.

Where my issue differs from zorbas is that I've been getting cutout of the e-brake. Power cycling solves that issue but it returns. It's inconsistent and hard to reproduce. Additionally on CA 3.14 and I'm wondering if any of the firmware updates since then would help with these issues. It just feels like maybe I'm reaching the limit of what the CA can handle (E-rider torque sensor, GMAC 10T with prop regen, digi and analog aux, multiple DC accessories) and maybe there's just a lot of interference.

 

[chuyskywalker]

I've attached a picture of my ground wiring attempt.

IMG-9264.jpg

I'll be making the same setup shortly to see if it resolves my issue with the PAS input cycling on/off as I pedal.

 

[chuyskywalker]

100% successful! Will clean up the installation later, but man I'm so glad I found this thread.

 

[zorbas]

.. I connected the -ve side of the battery to the frame. And now as soon as press the throttle, I have a fault with the phaserunner. It says "Fault 9 Instantaneous phase over current"...

Anyone who could help ? I did what is suggested on Grintech site with no positive effect.
"
Instantaneous Phase Overcurrent Fault (2-2 Flash): This is a fault that typically occurs when there is a sudden change in the motor eRPM (such as hitting a pothole, or the motor transitioning from open loop to closed loop sensorless operation), and it usually means that the current regulator bandwidth and/or the PLL bandwidth are not high enough to track the dynamic motor response."

When I change PLL and current regulator settings I get this new fault
"POST Static Gate Fault (1-8 Flash)"

My phaserunner even started to smoke when pressing the throttle!

 

[amberwolf]

When you see smoke, it usually means there has been a serious component failure. It could just be wiring, but it is more likely to be FETs, given the previous overcurrent failure, and power should immediately be disconnected from the system to try to prevent a fire.

I recommend first testing the FETs via the procedures on http://ebikes.ca in the Learn - Troubleshooting section.

If they test ok, then verify none of the phase wiring of the controller or motor shorts to the frame. Since battery negative is now connected to the frame, then any internal motor short to the stator or axle will be shorting a phase directly to battery negative, which shorts out the controller itself and could cause overcurrent readings (and kill FETs too).

If the FETs do show a failure, then still check the wiring to be sure that's not the cause of the failure. Grin can fix the controller, AFAIK, if it is damaged; you'd need to contact them directly about that.

 

[zorbas]

Thanks for this suggestion amberwolf. I did the check and evrything seems fine. I also tested the throttle and the pedal sensor and both trigger a response and make the hub turn.
As I said, I only get this fault when I connect the -tive side of the battery to the bike frame. Besides the issue with my CA showing inconsistent human power, it's fine.
The question is what could cause the fault when I connect that to the frame? I did exactly as everyone here.

 

[amberwolf]

If none of the phases or any other electrical connection of the motor or controller (other than battery negative) connects to the frame at any resistance you can detect with a multimeter, it might be kind of short that only occurs at high voltage, from an insulation failure in the windings.

To test this directly you'd need a meter that uses high voltage pulses to check specifically for this kind of issue, which isn't typically useful for much else and so it isn't practical to get one unless you need to test a lot of motors, coils, transformers, etc. for such problems.

An indirect test for this problem is to take the motor off the bike frame, and mount it to a different frame (or other test stand/fixture that will secure the axle in place so it can't spin out and damage your wiring) that is not electrically connected to the frame the rest of the system is on, then connect it to the controller that's still mounted to the original frame. If the problem does not occur anymore, then there is some sort of problem with the motor (like a partial insulation failure) that is allowing a connection between a phase and the stator or axle, whcih then is grounded to battery negative thru the frame when the battery negative is connected to the frame (but otherwise is not shorting to anything).


If the motor does have this problem, it can often be fixed with CoronaDope or similar high-voltage insulating lacquer; it depends on where the problem is (which is nearly impossible to find out without unwinding the motor). I fixed an MXUS 450x with this issue by drizzling CD down into all the windings of the phase with the problem. (I did the rest of them too once that had dried and tested fixed). But it doesn't always work, if the winding insulation damage is too close to the metal it's shorting against for the CD to get in there and insulate it.



Having had smoke actually come from the controller is still an issue--it means something got so hot it began vaporizing something, be that wire insulation or controller potting. I recommend examining the controller carefully in direct sunlight to see where the smoke may have come from. If it's insulation you can then repair it so no further problems occur from that kind of damage. If it's the potting, then it means whatever is at that point in the potting got very very hot, and may have been damaged from the heat

In either case, knowing where the smoke came from specifically will guide you to the source of the current path, which may help you find the root cause of the problem so it can be fixed.

 

[zorbas]

Thank you for these insight to my problem and the suggestions for investigations. I'll test that and hopefully find out the origin of the issue (fingers crossed). I have already removed a dodgy connection from the system and detached the controller from the frame (it's now hanging in a bag). Next step is testing. If nothing comes out then I'll test the hub on my truing stand. :thumb:

 

[amberwolf]

Note that the controller cant' shed it's heat within a bag, so if it normally gets warm out in the open air, it may get hot in a bag. If it gets hot in the open air, it may overheat in a bag, and if it's thermal rollback feature is setup / working it would then decrease available power to protect itself as it heats up. I don't know if it would shutdown completely, or if it could be damaged since it does have the rollback.

If it is normally not even warm in the open, it will probably not heat up in the bag.

 

[FuzzyWuzzy]

[youtube]bi3IVHanQYg[/youtube]

100% successful! Will clean up the installation later, but man I'm so glad I found this thread.

I love your solution.

 

[zorbas]

Thank you for these insight to my problem and the suggestions for investigations. I'll test that and hopefully find out the origin of the issue (fingers crossed). I have already removed a dodgy connection from the system and detached the controller from the frame (it's now hanging in a bag). Next step is testing. If nothing comes out then I'll test the hub on my truing stand. :thumb:

I wanted to give a little feedback after a few months of testing. I resolved my issue by simply insulating the L1019 connector (Hub to controller) with electrical tape. It's as simple as that. I realized my issue was coming from the contact of that connector to my bike frame, after keeping the cable away from the frame. After that, I still had some electrical noise into the Cycle Analyst (switching the display on its own while cycling). I grounded the neg cable (going to the CA) of the torque sensor on the frame, and it resolved that issue as well. I cross my fingers but apparently, everything is stable now.

 

[Chiasmata]

I'm almost too embarrassed to take a photo. It's just an 18ga wire "tied" around one of the negative prongs and then a loop of bare wire wrapped around a water bottlecage screw that goes into the frame.

If you are using a Cycle Analyst it is easy and effective to get the Ground signal from the Cycle Analyst and connect it to the bike frame. [Update: another suggestion I heard about later was picking up the Ground signal from the 6 pin PAS plug of the controller. This idea was described as, "a cable that goes in the 6 pin PAS plug of the controller with a ring terminal for grounding to the frame at any nearby eyelet."]

I installed an ERider T-17 torque sensing pedal assist system (PAS) and GMAC-10 kit from Grin Technology on a Motiv Duet tandem bike. The ERider was installed in the rear bottom bracket and was connected to a version 3 Cycle Analyst using two PAS Extension Cables (SKU: CAP-PAS_Ext) from Grin.

At first the assist did not work properly, it only gave power in momentary bursts and the Cycle Analyst reported erratic human-watt values including negative 200-600W. The solution turned out to be making a ground connection between the Cycle Analyst and the frame of the bike. Grin customer support said. “It seems like quite a few folks especially with long frame bikes and Erider torque sensors have had success with this now.”

Grounding the frame —
Grin customer support advised me, “one easy spot to get Ground signal would be from the Cycle Analyst's 5.5x2.1mm DC Power plug's outside terminal (inside is B+) if you're not using that connector for anything else.
That or any of the un-used JST connector available on the Cycle Analyst. Ground is Pin #2 on all except the 2-pin Thermistor input.” I purchased a “DC Power Cord, 2.1x55mm Jack to Bare Leads, 6ft/22awg” from a local store, we-supply.com. I had read in this discussion on Endless Sphere, subject “Erider torque voltage output consistency issue,” about grounding the bike frame and that Grin Technologies recommended including a 100mA fuse in the ground wire. What I found available at We-Supply was an Inline Fuseholder 3AB/3AG, 18AWG and a 125mA fuse of the 3AG form factor. I soldered the fuse holder into the ground wire of the above mentioned power cord. I left the positive wire of the cord intact for mechanical strength, but trimmed back and taped the bare end to make sure it doesn’t come in contact with anything. I put a crimp-on spade terminal on the bare end of the ground wire and attached it to a water bottle holder lug on the bike frame. That grounding made all the difference in the world. The torque sensing assist started working perfectly.

Motor GMAC 10T
Battery 52V 20Ah (14S) DT Reention
Controller V5 Baserunner_L10
Cycle Analyst CA3-WPx