Help! Erratic controller behaviour - is it a programming issue?

[Slightlypedantic]

I have a problem with a Lishui controller. At least, I think it's the controller. The supplier is currently sending a replacement, so I'll know for sure soon.

If the old controller is faulty, I'd be keen fix it and use it on another bike.

All is fine except that, if you are out of the saddle with one foot on the pedal and one on the ground (e.g. stopping briefly, doing a U-turn or negotiating a gateway), there is often a sudden but very brief unexpected surge of power. Not good when waiting for oncoming traffic to pass before turning right !

When moving off from rest, it takes about 2/3 turn of the pedals before the power cuts in, which is fine. I appreciate that when coasting and then recommencing pedalling the delay will be less. But the pedal movement when "scooting" slowly or briefly stopping is very small and surely should not trigger the motor? Or could there be another cause?

This has been apparent from new, not something that has developed later. There are only a few settings available via the display (D500) and none that make any difference. Could the problem be "built in" to the programming, in which case could it be altered using a laptop and lead?

NB The setting "Slowly start up" has settings 0, 1, 2, 3. I've tried 0 and 2, but no discernible difference. I assume this is a soft start adjustment but perhaps it is something else?

Any thoughts or ideas, anyone?

Thanks!

 

[saneagle]

There is nothing in the controller that can do that. The only way it can happen is if the controller is getting either a PAS pulsing signal or a signal on the throttle connector of more than 1V. Did you fit a throttle?

 

[Slightlypedantic]

There is nothing in the controller that can do that. The only way it can happen is if the controller is getting either a PAS pulsing signal or a signal on the throttle connector of more than 1V. Did you fit a throttle?

Hi saneagle, and thanks. No, I didn't fit a throttle and left the two throttle settings set to "N". Could it be a PAS sensor or cable fault?

 

[saneagle]

Hi saneagle, and thanks. No, I didn't fit a throttle and left the two throttle settings set to "N". Could it be a PAS sensor or cable fault?

It's possible. Unplug it, lift the wheel off the ground and short the red wire to the one that isn't black. if the motor spins, it's a problem with the pedal sensor.

 

[Slightlypedantic]

It's possible. Unplug it, lift the wheel off the ground and short the red wire to the one that isn't black. if the motor spins, it's a problem with the pedal sensor.

OK. Thanks again.

 

[Slightlypedantic]

OK. Thanks again.

Just remembered: I fitted a left side PAS, the type with a plastic casing that "plugs" into the bottom bracket internal splines. Looks neat and is easy to fit, except that:

The magnet disc is inside the casing and is quite a tight fit on the BB axle. The only visible part of the disc is the plastic spiral splines that grip the BB axle. After seating the casing against the BB, I pushed the disc in a little more by its splines to try and keep it from rubbing against the casing. Unfortunately it's not possible to tell if the disc is exactly square to the axle or pushed in too far, so it could be rubbing the inside of the casing. Could this be relevant?


Error: I said above that the display is a D500 but of course it is actually a C500.

 

[Woosh]

The left-sided PAS fits 16mm in diameter axle. Yours probably has 17mm diameter. You can force fit the PAS onto your bb but it will crack in a few months. best buy a Neco bb of the same size. Neco square taper bb has 16mm diameter axle.

 

[Slightlypedantic]

The left-sided PAS fits 16mm in diameter axle. Yours probably has 17mm diameter. You can force fit the PAS onto your bb but it will crack in a few months. best buy a Neco bb of the same size. Neco square taper bb has 16mm diameter axle.

That's worth checking. I do remember there was something about the axle diameter when I bought the kit so I'll have to review that.

 

[Slightlypedantic]

That's worth checking. I do remember there was something about the axle diameter when I bought the kit so I'll have to review that.

Thanks Woosh, good point and thanks for the tip.

Actually there wasn't anything about axle diameter in the fitting instructions (it wasn't a Woosh kit). I was actually thinking of a different kit (which was a Woosh one) that I fitted for a friend, where this was not an issue. Woosh's instructions etc are very informative!

Anyway, I've had my calipers out. Unfortunately none of the circular part of the BB axle was visible between the PAS sensor and crank although the housing looked well seated. I'd have to remove the crank and sensor to get a definite axle diameter measurement. There was just room to get a measurement across the square taper diagonal (i.e. not across the flats), about 16.2mm. Since this is over 16mm, perhaps it suggests I have 17mm axle, as you suspect. When fitting the sensor, it took a firm push and a bit of coaxing to get the magnet disc in position and it did seem to be on the cusp of being on the circular section rather than the taper... I remember thinking it didn't feel quite right. A slightly longer BB might be a good idea.

Although the bike hasn't been ridden much, the PAS sensor has now been fitted for 8 or 9 months. I suppose that if it is going to break, time is more important than use. I imagine that it would break across a magnet hole due there being less material and higher stress at that point.

Assuming the magnet disc has split, could any of the following conditions might account for the random spurious bursts of power?

• Magnet disc has split and is loose/crooked/slipping
• Magnet disc has slipped back onto the square taper section and is crooked/too far from the Hall sensors
• Magnet disc is split but not loose, however a magnet might be loose
• Magnets are slightly unevenly spaced due to the split
• Magnet has fallen out

Btw, I noticed slight surging when the motor is up to the speed limit for each power level. This is slow and rhythmic, as though the controller is hunting slightly. Could any of the above be a cause or perhaps it's just normal operation?

I'll have to remove the crank and sensor to see what's actually going on but even then I won't be able to see the magnet disc unless I open the casing. Probably best to replace it, making sure of course that it's the right size for the axle! If that doesn't fix it I will at least know to look elsewhere for the cause...

Meanwhile I am awaiting a replacement controller from China... could take a while.

Cheers.

 

[Woosh]

16.2mm is OK, the sensor won't crack, just jiggle the sensor case a bit to fit it better. Post a picture of the splines on the bb cup if you want me to check.

 

[Cadence]

I can concur about a 17mm dia spindle being too big for an "enclosed" type sensor. My symptoms were rather different to the OP though. Occasionally I had a big delay in the PAS working and then it would fail altogether. I had a large gap between the BB and the crank arm and as the sensor was too small for the spindle it would "spring" out and slide towards the crank on the taper flats. I did a temporary repair using rubber O rings between the crank and sensor, but eventually the inner disc in the sensor cracked. Replaced with the split type and not had any more problems.

 

[Slightlypedantic]

The sensor case fitted well into the BB splines, no worries there. It was the magnet disc splines that concerned me and whether the magnet disc is damaged.

I may have measured the square taper diagonal at a point where it is not the same as the circular part of the axle. I can't see very much, so it will be a dismantle job to check what is going on.

I'll certainly post a photo if I am in doubt - thanks!

 

[Slightlypedantic]

I can concur about a 17mm dia spindle being too big for an "enclosed" type sensor. My symptoms were rather different to the OP though. Occasionally I had a big delay in the PAS working and then it would fail altogether. I had a large gap between the BB and the crank arm and as the sensor was too small for the spindle it would "spring" out and slide towards the crank on the taper flats. I did a temporary repair using rubber O rings between the crank and sensor, but eventually the inner disc in the sensor cracked. Replaced with the split type and not had any more problems.

Thanks Cadence, I'll bear that in mind.

 

[Slightlypedantic]

I've yet to remove the crank and check the axle diameter, but when I do I'll probably fit a new sensor regardless, while I'm at it, and change the BB if necessary. That will eliminate one possibility.

Meanwhile, I'm looking at changing to a KT controller and display, which would eliminate any other possible cause.

 

[Woosh]

Meanwhile, I'm looking at changing to a KT controller and display, which would eliminate any other possible cause.

yes, that's a pretty good plan.

I have a problem with a Lishui controller. At least, I think it's the controller.

Your controller is YCSH (JYT) brand, not Lishui. Lishui controllers have product codes starting with LSW-something. The good thing about Lishui controllers is their firmware cope automatically with a lot of variations in motors: Hall or Hall-less, number of magnet poles in the motor, number of magnets on the PAS disc, number of assist levels etc. They are suitable for a lot of kits, simplifying support because customers don't need programming manuals and support technicians do not have to check the stored parameters. On the other hand, the KT controllers let you change stored parameters yourself.

This is a typical behaviour of a cadence sensor and a well made controller, the motor shifts smoothly up and down the power range when you change the assist level, no big 'bump'.

 

[Slightlypedantic]

Yes, my mistake. My controller, built in to the battery base, looked similar to a Lishui in a Woosh kit I fitted earlier for a friend, so I assumed it was the same brand. I later realised my controller is actually different and not a Lishui at all.

Your video was interesting; the speed didn't seem to ramp up/down between levels. Running at 25kph? It certainly seemed smooth. I think I read in another thread recently that Woosh's current range of bikes now have current control rather than speed control - is that correct?

 

[Woosh]

Your video was interesting; the speed didn't seem to ramp up/down between levels. It certainly seemed smooth. I think I read in another thread recently that Woosh's current range of bikes now have current control rather than speed control - is that correct?

current control on bikes.
The kit controllers still have speed control, I can load current control firmware on request.
Your controller is made by JYT. Both Lishui and JYT can supply current control firmware on request.
I'll make a video with Lishui speed control later this week for comparison.

 

[Slightlypedantic]

Thanks Woosh. That sounds great.

With the JYT speed controller, levels 1 and 2 are not much use to me as the speed limits are so low that I'm doing all the work anyway. If I can get it changed to current control I might be a very happy boy! I'd like to try it and see how it goes, so I'll see what I can find out and report back for benefit of any others reading this thread.

 

[Peter.Bridge]

Thanks Woosh. That sounds great.

With the JYT speed controller, levels 1 and 2 are not much use to me as the speed limits are so low that I'm doing all the work anyway. If I can get it changed to current control I might be a very happy boy! I'd like to try it and see how it goes, so I'll see what I can find out and report back for benefit of any others reading this thread.

Yes - the Bafang SWX02 kit that I got from Ali Express is the same - I changed it to 9 PAS levels on my display which is slightly better, but I am usually pedalling faster than levels 1,2 or 3 (apart from starting off), so really use levels 4-9. At each level there is a speed where I can add a decent amount of pedalling effort and the motor will help but the motor really cuts assist above that speed and assists too much below that speed - there is one "sweet spot" per level. It's still really good but at some point I will get a controller , display (+ will I need new PAS sensor, throttle and brake sensors with different connections?) that does current control.
It doesn't seem as pronounced on hub motors as mid drives - I have a BBS02b and the max assist speeds for each PAS level were really on/off (fortunately you can program them to be entirely current based)

 

[Woosh]

this video shows you a Lishui in speed control mode: