M5 Display - setting magnets and PAS sensitivity

[Cyclomania]

So I have an M5 display, a 17mah controller(compatible with 36v and 48v) and a 36 volt battery on my ebike.

It was originally a 48 volt battery on it but I switched it out. Everything is working fine now. But I want the PAS to help me for a longer period. When it kicks in.

It is now kicking in pretty well. I've set the PAS sensitivity (m5 setting 11) to 4. So it kicks in pretty fast when I pedal. But it gives me a strong but short burst of power, and then it stops giving me power. So the throttle is much nicer to use. SInce it gives power continually.

Is this PAS sensitivity also in charge of how long the PAS is giving the bike power for? How do I make it last for longer when it kicks in? Should I set the PAs sensitivity higher or something?

Also, when I was tinkering I by mistake changed the 13-setting which is for magnets of the motor. I cannot remember what this was set to from the beginning.
So could you help me understand, from my pictures, how many magnets is in my motor somehow? Not sure how to figure this out. The number of magnets I mean. Display setting is set to 5 now but not sure if I have 5, 8 or 12 magnets so I appreciate the help figuring this out.

Thanks

 

[Cyclomania]

Either of those circuits can override PAS, and anything else that may use the same 5v input could cause issues if there is an intermittent connection or short.

Ah I tried now this evening. First with no throttle. Then with disconnected hall sensors. I also checked the connections. And as far as I can see there is nothing wrong with them. But I am not sure.

Then I tried the self learning line. It now looks like this video here. Maybe a little bit longer bursts(?). What do you guys think? But not entirely sure. It still works in bursts.

Maybe best to just switch to another PAS-sensor? And try that one out? One specifically made for kt-controllers?

Video Unavailable

This video is unavailable.

streamable.com

 

[fill]

Thats with Everything but pas battery and motor disconnected from the controller??
and its a new controller.. And NO error or upset on the display? what speeds are this happening at?
What does the display show when the motor cuts out?

Most ODD I would have laid cash money on it being a dodgy/shorting brake sensor.

So to rule out the controller a new pas sensor may be a worthwhile investment $2 if you can wait for a shipment from china $10 via amazon..

I have a proclivity to take everything apart. here i would feel justified in opening up the controller to check all the wires are connected as expected and no stray whiskers are shorting connections. .. Its like every printer problem, if you dont check the printer lead first its the printer lead thats the problem..

 

[fill]

fwiw you should be able to 'ghost pedal' in pas mode 5 and the bike should act as it the throttle is wide open. no cut offs until maximum ouput is hit,

checking 'manual b' again i see it suggests a 'default' value for 12a as a current limit? You did change that to match the 17a of your new controller?

 

[E-HP]

Ah I tried now this evening. First with no throttle. Then with disconnected hall sensors. I also checked the connections. And as far as I can see there is nothing wrong with them. But I am not sure.

Then I tried the self learning line. It now looks like this video here. Maybe a little bit longer bursts(?). What do you guys think? But not entirely sure. It still works in bursts.

Maybe best to just switch to another PAS-sensor? And try that one out? One specifically made for kt-controllers?

Video Unavailable

This video is unavailable.

streamable.com

Yes, it seems like hardware. The KT model that fill described should work. I used the same one when I was running a KT controller.

Just to confirm, you did test setting P13 to something low, correct? P13 essentially sets how much the PAS thinks the crank is rotating. If for instance, you have a 16 magnet PAS, the controller will know 16 signals is one revolution. Also, if the signals are too far apart, PAS stops assisting. I noticed in the test, your cadence is pretty slow. For the example, if you set P13 to 8, the controller will think half a revolution is a full one. In other words, you can trick the controller to have PAS kick in faster, or not cutout when pedaling slow. Do you pedal that slowly while riding?

 

[Cyclomania]

fwiw you should be able to 'ghost pedal' in pas mode 5 and the bike should act as it the throttle is wide open. no cut offs until maximum ouput is hit,

checking 'manual b' again i see it suggests a 'default' value for 12a as a current limit? You did change that to match the 17a of your new controller?

I changed it to 16. So the controller is max 17. So felt good to take it up to 16.

 

[Cyclomania]

Yes, it seems like hardware. The KT model that fill described should work. I used the same one when I was running a KT controller.

Just to confirm, you did test setting P13 to something low, correct? P13 essentially sets how much the PAS thinks the crank is rotating. If for instance, you have a 16 magnet PAS, the controller will know 16 signals is one revolution. Also, if the signals are too far apart, PAS stops assisting. I noticed in the test, your cadence is pretty slow. For the example, if you set P13 to 8, the controller will think half a revolution is a full one. In other words, you can trick the controller to have PAS kick in faster, or not cutout when pedaling slow. Do you pedal that slowly while riding?

You guys talked about a kt-v12l right? So this one would work?

https://www.aliexpress.com/item/1005003973027024.html?spm=a2g0s.imconversation.0.0.39f63e5ffQ1siD&pdp_npi=4%40dis%21NOK%21NOK%20172.50%21NOK%20125.90%21%21%2116.51%2112.05%21%40211b613917102828502503383e145e%2112000027819005579%21sh%21NO%21167266123%21

Amazon's was faster but a lot more expensive shipping for me, as fill said. And we have toll on this stuff above 33 dollars nowadays.

The thing with the p13 is that it does not seem to work if I pick anything else than twelve. If I go with five or eight the PAS don't seem to work at all. Yeah, pretty low cadence. No, I pedal more when I ride the bike. But just wanted to see if it would work as long as the crank is turning.

This happened when I rode the bike also. Pretty annoying these short bursts so I used the throttle instead, mostly.

The throttle does not do this thing. It gives continuos power. Weird huh?

 

[fill]

Its not just pulse count that dictates pas control, The PAS sensor magnet count and pattern of placement will create a particular 'wave' pattern too (I am out of my depth now) as to not get stimulated by any old passing magnetic field, last thing you want when waiting at a red light is for the pas sensor to react to the person next to yous footware with magnets on them and for the bike to jump out into traffic. magnets are everywhere..

Your link didnt work for me, but there is no need to get the same pas sensor get an 8x or13x if you prefer the style or can find it cheaper just set whatever you get in the controller and it SHOULD all be good.. matching the connecting plug/socket is more important but that too can be sorted with a soldering iron and heatshrink.. the one benefit of this type is it shouldn't go wrong!!?? as types with separate magnet disks can get knocked out of line. or even fall off..

I have an amazon prime uk link that sells a v12l for less than 10 euros?? may require vpn shenanigans
other sensors also fall in this price range too.. though ebay (via china) offers the lowest 'local' market place prices 5+ euros.

You can buy one from yose-power in germany for about 20 euros delivered as i had to replace mine with one like this (the replacement was E14 +pnp iirc

since my cheap donor bike had a crank a bit wider than 17mm in diameter causing the V12L to spin on the crank until the wire snagged it.. Took me weeks to work out why the pas started one day with one foot down while the next i would pedal round 2-3 times before the pas kicked in.. ( I would offer my original, but i hacked/broke it trying to widen the bore a bit..) and they shipped my replacement from germany (2-3days iirc)


Your right it is Very Odd it has to be a fault in the wires or a bad solder junction, magnets dont work intermittently afak??? and a hall effect sensor - a 'solid state' component near as damn it, will either work or not work, again to the best of my understanding??..

Even a stray strand of wire not cut/soldered correctly causing a random short inside the controller would repeat randomly and not only under pas control??

Really hope you have the Eureka moment when whatever is the cause gets revealed soon. and you let us know.. .

 

[Cyclomania]

Its not just pulse count that dictates pas control, The PAS sensor magnet count and pattern of placement will create a particular 'wave' pattern too (I am out of my depth now) as to not get stimulated by any old passing magnetic field, last thing you want when waiting at a red light is for the pas sensor to react to the person next to yous footware with magnets on them and for the bike to jump out into traffic. magnets are everywhere..

Your link didnt work for me, but there is no need to get the same pas sensor get an 8x or13x if you prefer the style or can find it cheaper just set whatever you get in the controller and it SHOULD all be good.. matching the connecting plug/socket is more important but that too can be sorted with a soldering iron and heatshrink.. the one benefit of this type is it shouldn't go wrong!!?? as types with separate magnet disks can get knocked out of line. or even fall off..

I have an amazon prime uk link that sells a v12l for less than 10 euros?? may require vpn shenanigans
other sensors also fall in this price range too.. though ebay (via china) offers the lowest 'local' market place prices 5+ euros.

You can buy one from yose-power in germany for about 20 euros delivered as i had to replace mine with one like this (the replacement was E14 +pnp iirc

since my cheap donor bike had a crank a bit wider than 17mm in diameter causing the V12L to spin on the crank until the wire snagged it.. Took me weeks to work out why the pas started one day with one foot down while the next i would pedal round 2-3 times before the pas kicked in.. ( I would offer my original, but i hacked/broke it trying to widen the bore a bit..) and they shipped my replacement from germany (2-3days iirc)


Your right it is Very Odd it has to be a fault in the wires or a bad solder junction, magnets dont work intermittently afak??? and a hall effect sensor - a 'solid state' component near as damn it, will either work or not work, again to the best of my understanding??..

Even a stray strand of wire not cut/soldered correctly causing a random short inside the controller would repeat randomly and not only under pas control??

Really hope you have the Eureka moment when whatever is the cause gets revealed soon. and you let us know.. .

Thanks. Lots of good links there.

Maybe the issue is I need to solder it and my connections(you can see my adapters in this video) are not strong enough?

Video Unavailable

This video is unavailable.

streamable.com

Could that give rise to such a problem? I think my connections are pretty good but maybe not.

From five seconds in approximately you can see the quick cable connectors.

 

[Cyclomania]

Your right it is Very Odd it has to be a fault in the wires or a bad solder junction, magnets dont work intermittently afak???

If it is a bad solder connection do you think that one is inside the wires of the pas sensor going to the controller? Or rather inside the actual magnet area, of the PAS sensor glued to the crank, or how do you mean?

I have ordered one of those new ones specifically for KT controllers. But I could also solder a bit in the meantime, when I wait for the new one, to see if I can make the old one work.

 

[Cyclomania]

Yes, it seems like hardware. The KT model that fill described should work. I used the same one when I was running a KT controller.

Just to confirm, you did test setting P13 to something low, correct? P13 essentially sets how much the PAS thinks the crank is rotating. If for instance, you have a 16 magnet PAS, the controller will know 16 signals is one revolution. Also, if the signals are too far apart, PAS stops assisting. I noticed in the test, your cadence is pretty slow. For the example, if you set P13 to 8, the controller will think half a revolution is a full one. In other words, you can trick the controller to have PAS kick in faster, or not cutout when pedaling slow. Do you pedal that slowly while riding?

When I set it to 5 or 8 it does not work at all. Probably because it has to be 12 in this one I guess.

I have ordered a new one. Should be specifically for KT the one I ordered. I guess I have tried most things with the one I have on there now. But you mentioned something about the distance between the magnets and the crank or something like that? I guess I could try to see if that is an issue in the meantime when I wait for the new one.
But how did you mean? You meant the distance between the PAS sensor and the pedal arms or something like that?

 

[AGS]

The gap between the sensor and magnetic disk should be around 1mm. And the magnetic disc should be tight so it doesn’t spin on the crank when the pedal is rotated. Your sensor is too loose and your gap is too wide. Those magnets aren’t that strong, so the disc has to be close enough to the hall sensor so it can pick up the magnetic field.

Just count the number of magnets on the disc, then you will know if it has 5, 8 or 12 magnets and set your controller accordingly.

 

[fill]

If it is a bad solder connection do you think that one is inside the wires of the pas sensor going to the controller? Or rather inside the actual magnet area, of the PAS sensor glued to the crank, or how do you mean?

I have ordered one of those new ones specifically for KT controllers. But I could also solder a bit in the meantime, when I wait for the new one, to see if I can make the old one work.

IM just 'thinking aloud' above and guessing, but inside your sensor everything is enclosed in potting compound or embedded within plastic requiring destructive surgery to expose if anything like the one i had to replace,#

I dont think its a short or anything at the controller end as if it was the same error would occur when under throttle control.

Your sensor is all in one and has no separate magnet disk, however the magnets in yours are tiny and the wobble in your sensor could be the root as suggested by @AGS smaller magnets may have weaker magnetic fields and require significantly closer tolerances perhaps??

Soldering the connection to the controller would rule out any clamping on insulation not wire errors that remain hidden in the connector block?? but if used correctly those blocks make good connections.just not at all water tight ..

Ever played with an arduino?? if so there is a fairly straight forward way to bench test your pas sensor, but it might be a tricky first look into arduino? and if you need to buy one, the new sensor will probably have arrived by the time its delivered. I think almost everything is covered in the first 5-6 mins of this utube that recently landed in my view..

 

[Cyclomania]

The gap between the sensor and magnetic disk should be around 1mm. And the magnetic disc should be tight so it doesn’t spin on the crank when the pedal is rotated. Your sensor is too loose and your gap is too wide. Those magnets aren’t that strong, so the disc has to be close enough to the hall sensor so it can pick up the magnetic field.

Just count the number of magnets on the disc, then you will know if it has 5, 8 or 12 magnets and set your controller accordingly.

I am at work now so I dont have the PAS sensor in front of me. Thing is this pas sensor looks a bit weird? I installed another one recently but that one was completely different from this one.

Where is the magnet and sensor on this one you think?
I am going to check when I get home as well. But maybe you can tell me from this picture? It is not one of those regular pas sensors.

Do you think both the sensor and magnets are inside the green area in the second picture? Or are sensor in the red area somewhere? I am going to investigate later also.

 

[Cyclomania]

IM just 'thinking aloud' above and guessing, but inside your sensor everything is enclosed in potting compound or embedded within plastic requiring destructive surgery to expose if anything like the one i had to replace,#

Your sensor is all in one and has no separate magnet disk, however the magnets in yours are tiny and the wobble in your sensor could be the root as suggested by @AGS smaller magnets may have weaker magnetic fields and require significantly closer tolerances perhaps??

Ah yes makes sense. Looks like everything is in the green circled area right? So maybe they wobble around in there creating problems for me? Not an easy fix.
Probably simplest solution is to change into the KT-PAS when it arrives(?)

 

[Cyclomania]

Ever played with an arduino?? if so there is a fairly straight forward way to bench test your pas sensor, but it might be a tricky first look into arduino? and if you need to buy one, the new sensor will probably have arrived by the time its delivered. I think almost everything is covered in the first 5-6 mins of this utube that recently landed in my view..

Ah this one is above my head. Interesting to watch but havent been tinkering with those before .

 

[AGS]

Ah, didn’t realise it’s an all in one, but it shouldn’t be rattling around on your bottom bracket.

 

[fill]

Within the green yes, 2 parts the spinny bit with magnets embedded within and the stationary bit with the sensor and wires.. if the spinny bit wobbles too far off its intended 'track' - perhaps thats it? its as good a suggestion as anything else proffered

 

[docw009]

I had a situation two winters ago where the PAS would cut out during a half pedal rotation. I was using an all-in-one sensor with a KT controller. Apparently, the sensor splines were slipping on the pedal axle. I replaced it with an old sensor/disk out of my junk box. They probably weren't even sold as a matched pair, but it worked ... for a while.

Last summer, the same situation re-occured. At first I thought it was something wrong with the controller, but when i went to adjust the disk, I saw it was actually slipping on the axle, I took it off ane planned to bolster it with a layer of epoxy, but the magnets started falling out too. Age, I guess. Maybe the magnets had fallen out of the integrated sensor too. Never looked inside.

 

[Cyclomania]

I had a situation two winters ago where the PAS would cut out during a half pedal rotation. I was using an all-in-one sensor with a KT controller. Apparently, the sensor splines were slipping on the pedal axle. I replaced it with an old sensor/disk out of my junk box. They probably weren't even sold as a matched pair, but it worked ... for a while.

Last summer, the same situation re-occured. At first I thought it was something wrong with the controller, but when i went to adjust the disk, I saw it was actually slipping on the axle, I took it off ane planned to bolster it with a layer of epoxy, but the magnets started falling out too. Age, I guess. Maybe the magnets had fallen out of the integrated sensor too. Never looked inside.

Aha.

The PASes that has worked the best for me so far are these ones that are very simple to just put on the axle after removing the pedal arm:

https://www.aliexpress.com/item/1005006229882210.html?spm=a2g0o.order_list.order_list_main.52.21ef1802ec9QLi

Do you think this one might work with a KT controller? It is jhrak another brand. But could not hurt to try it on right? Since I am going to remove it anyway, when the new one arrives, and I have one of these at home. Do you guys think it might work with my KT controller?

 

[fill]

IM confused now, Where is the KT controller? KT or Ketung? brand controllers Use KT displays (kt-lcd1, kt-lcd3, etc...) and would not work at all with a M5 display afaik?? KT brand controllers are popular since they have a wide range of compatible display options And provide a good set of configurable parameters.

But if referring to your current controller? Yes the above linked to pas sensor should work fine with your controller and a KT controller should you invest in one in the future too.

 

[Cyclomania]

IM confused now, Where is the KT controller?

I have a KT controller on this bike But this PAS (I have at home right now) is not KT. It is a brand called Jrahk. But in my experience these types of PASes are usually pretty straight forward to install and usually works fine. So I was thinking about perhaps testing it, in the meantime when I wait for my KT-Pedal assist (that is sent from China).

But I wanted to get some opinions on this idea before I put time into testing it . So I don't do anything unnecessary.

 

[docw009]

Wait. You said you have a KT controller and an M5 display. You're talking about a P13 parameter? No P13 in a KT display.

KT controllers only work with KT displays. If you've hooked up a non KT display, I believe the controller powers up in its default mode where throttle will work with a single level PAS. It may not work with some PAS units if the signal isn't matching the default shape,

You shouldn't be able to see the speed value if it is not a KT display,

 

[Cyclomania]

Wait. You said you have a KT controller and an M5 display. You're talking about a P13 parameter? No P13 in a KT display.

KT controllers only work with KT displays. If you've hooked up a non KT display, I believe the controller powers up in its default mode where throttle will work with a single level PAS. It may not work with some PAS units if the signal isn't matching the default shape,

You shouldn't be able to see the speed value if it is not a KT display,

Hmm, seems to work at least. Because I changed from 36v to 48v in the display and then I got a correct reading. Also possible to switch magnets in this display. Only works at 12 for example.

So seems most of this stuff works between controller and display. I think the issue is the PAS-sensor. Everything else works fine.

 

[fill]

Post a pic of the controller/label, that way we can help you identify it, but thats not a real problem, just a hiccup/misunderstanding..
(Google for example a ktlcd3 manual for a quick double check yourself if still in doubt)

Yes the sensor you have with the non kt brand name should work fine.. the only reason the one i linked to originally had a kt prefix was chance, it was the first cheap option a quick google revealed. Im sorry if that led to a misunderstanding.

The only criteria you need to worry about is the magnet count. Which afaik should be the same for all sensors of that design, but i could be wrong about that... EDIT* AND the position on the crank Left or Right side.. .
When your no name sensor works , or gets us one step closer.. , at least you have another spare on its way ..

If however the problem persists with the replacement no name pas sensor then 'pause for thought' - tripple check wiring, continuity test every wire while waggling the loom - extreme distress testing.. and if that all passes focus has to shift to the controller imho.

 

[Cyclomania]

Here it is