A provocative title, I know.

Those who've been around a while may remember my system not coming back on power if I go over the cutoff speed and continue to pedal and/or forget to operate the brake cut off.

And it's still doing it - if anything it's worse. Today, whilst admiring the wild flowers and stunning views, I forgot to pull the brake on a down ramp. This left me, again, with no power on the other side. Power cut in at some random point (I've never been able to ascertain parameters for this) and slowly ramped up from a lowly 60w to full 700w so slowly that I was down to about 6mph by the time I was getting usefull power. Result is that most of the time the speedo is getting more attention than the country side I'm riding through.

To add to this the speed cut off, that was nicely set at 15.5mph, today seemed to have spontaneously altered to something like 14.8mph. I've altered nothing since my last ride a few days ago.

I'm really hacked off with it. I have a suspicion that something in the controller may have been heat damaged when I was previously trying to run a 48v 328rpm motor at 36v in my 700c wheel. Don't ask why, suffice to say it was an abysmal failure. I'm now running a 36v motor correctly.

I'm running an AKM128 (250w label), T09S controller (set at 20A max) and lcd3 display.

I've tried different pas detector. All sorts of different settings. Speed registration matches a gps satnav, so P1 should be right.

I'm on the verge of ordering another controller but thought I'd give one more thing a go.

Which setting is it to change to speed control? I know it's going to be less subtle but it's what I'm used to from my Ezee days.

Just one more thing. I'm pretty sure the motor is whining more than when it was new. This is a smooth sound that continues after cutoff. There was some thought, from Saneagle, that the motor core was continuing to rotate when no power was applied, thus making the controller think it was over speed ?!?. Some thought that internal grease may be missing or too thick. Possible?

Post all your settings so that we can check them.

20 minutes ago, Benjahmin said:

Which setting is it to change to speed control?

Dunno anything about this hub stuff, but @Nealh says here it's P3=0 for speed control.

https://www.pedelecs.co.uk/topic/31165-kt-lcd-3-power-settings/#findComment-443795

Edited March 30Mar 30 by guerney

P1-168, P2 - 1 I think, throttle acting as walk assist then full when pedalling. P3-1, P4-1, P5-13,

C1-07, C2-0, C3-8, C4-1 (see above) C5-09, C6-3, C7-0, C8-0, C9-0, C10-n, C11-0, C12-4, C13-0, C14-3

Thanks Guerney

Edited March 30Mar 30 by Benjahmin

Just set P3 to 0. Certainly gives differing speeds and may work better when I'm out with the missus on her Big Bear. Should be able to better match her speed.

Only went to the village shop so not sure if this circumvents my problem. From what I saw on this short run, I think not but I'll see on the next outing.

46 minutes ago, Benjahmin said:

P1-168, P2 - 1 I think, throttle acting as walk assist then full when pedalling. P3-1, P4-1, P5-13,

C1-07, C2-0, C3-8, C4-1 (see above) C5-09, C6-3, C7-0, C8-0, C9-0, C10-n, C11-0, C12-4, C13-0, C14-3

They all look fine. If you can hang on a few weeks until the weather is a bit warmer, I'll come to you with a couple of other controllers to try. I'll contact you nearer the time to make arrangements. If you don't hear anything by the end of April, send me a reminder.

Sorry to hear about your Kt controller problem with your AKM motor. This appears to be a known problem and is discussed in the German form with one supplier claiming that he has a batch of Kt controller all suffering from this bug. Link is below. You need Google translate .

. https://www.pedelecforum.de/forum/index.php?threads/hysterese-kt-controller-mit-akm100sx.115911/

37 minutes ago, Sturmey said:

Sorry to hear about your Kt controller problem with your AKM motor. This appears to be a known problem and is discussed in the German form with one supplier claiming that he has a batch of Kt controller all suffering from this bug. Link is below. You need Google translate .

. https://www.pedelecforum.de/forum/index.php?threads/hysterese-kt-controller-mit-akm100sx.115911/

Did he indicate the part numbers or batch numbers? I can't read it because I don't want the cookies.

Edited March 30Mar 30 by D8veh

I read through the thread, as far as I can see the guy Sandwich only turns up once and gives no indication of batch numbers.

My german isn't up to much but it seems like similar answers are being tried as those offered to me on this forum. Particularly the point about the motor never actually stopping but not giving power over the cut off.

It looked like what the OP was saying was exactly the symptoms I have, albeit on a speed pedelec.

Thanks for the offer Saneagle, I'll see how the speed control performs in the interim.

1 hour ago, D8veh said:

Did he indicate the part numbers or batch numbers? I can't read it because I don't want the cookies.

He did not give the batch number. On Google Chrome on my old computer, you can right click to open in incognito window, accepts cookies if you have to and right click for translate. The cookies should clear when leaving the tab.

Edited March 30Mar 30 by Sturmey

Been out for a good ride with P3 set to 0 - so speed control.

The ride started on a shared use ex railway track, so relatively flat. System was well behaved and gave various speeds as expected. However this didn't last. As the ride got into some serious up and down things started to get confusing. In, say, pas 3 on the flat I'd get around 12/13 mph - perfectly acceptable. Once going down a hill over the main cut off the power would not cut in till a very low level (5 or 6 mph). Varying the pas levels did not make any difference leaving me not knowing what speed any pas level was going to give me. Also the 'gremlin' was still fully operational, leaving me with little or no power at critical points.

You may also expect that riding below the speed for any particular pas setting would give you full power. Not a bit of it. The only pas setting giving me a full 700+ watts was level 5. The other settings seemed to give a wattage around the same as those given in current control.

I will say that on the flattish track riding on pas 3 gave me a pleasant 12/13mph and a lot of the time I was riding above this so consuming little or no power. I can see this giving a much higher range. Also level 2 was giving around 10mph which is what my misses seems to like riding at, so I'll be using speed control when riding with her.

However ,as we come into main bike riding season, I'm heartily fed up with this and I think I'm going to bite the bullet and order another on before we can't get them anymore. If the new one behaves the same it'll go back quickly.

Just ordered a new controller from TBK using chinapost. We'll see how long that takes. Let's hope the new one's 'proper'.

Controller arrived today, so that's 10 days to come from China.

Can't get to it at the moment as we have house guests. However, will the current display settings transfer to the new controller, or will I have reset them all?

14 minutes ago, Benjahmin said:

Controller arrived today, so that's 10 days to come from China.

Can't get to it at the moment as we have house guests. However, will the current display settings transfer to the new controller, or will I have reset them all?

They should transfer.

Cheers, I'll let you know when I get to it.

Just installed the new controller, the settings did transfer.

Took it out for a run and it seems to be behaving exactly the same. There's a down slope away from the house, I let it go to 23mph pedaling all the way with no brake applied. When I got to the bottom the power did not come back in until around 7.5mph, it came in at around 70w and ramped up very very slowly to 750w.

On a later slope I let it run to 18mph, no brake whilst pedaling. This came back in at 15.5mph, still with a ramp but at a higher start level and a quick ramp up, which I guess is reasonable.

On the last slope I let it run to 23mph without brake or pedaling, this cut back in correctly.

There is another oddity. Every so often the motor cuts on gear change and comes back in after around 1second, as though I've got a Bafang gear change switch fitted. I haven't. Motor plug is fully inserted and this also happened with the previous controller.

So this is two controllers behaving much the same. The common points are the motor and wiring loom, I have no way of knowing if the fault lies there. Is anyone running a 36v AKM128 with a KT T09s that is willing to try my spare to see if it behaves like this for them?

Ok not a fix but a work around ? idea at least.. you say a pull on the brake cut off 'normalises' motor activation with pedal assist below the speed cut off?

have you explored the minimum cut off signal duration that ' normalises the operation? if not? could a too rapid to register actual cut off but correct operation signal be generated?? a diy solution to ground once per crank revolution perhaps varying the length of the contacts and thus time connected?

Ideally a short enough to normalise operation signal may not be so long enough to register as an actual cut off of the motor ?

Y spliter cables exist for adding gear change sensors to the brake cut off lines for a few $'s

555 timer ic

when you say power "ramped up very slowly", how long in seconds did it take to go from 250w to 750w"? Can you test and let us know?

From 70w up to 750w I'd say around 12 seconds, although it's variable.

Operating the brake switch without actually braking is a fairly precise operation. It's possible that the brake switch is accidentally released momentarily over a bump or something. Even such a short blip (over the cutoff) will cause the gremlin.

I have pondered fitting a toggle switch to the second brake connection so I can just switch it off downhill and concentrate on potholes and wildflowers as I wouldn't have to be watching the speedo so much.

1 hour ago, Benjahmin said:

From 70w up to 750w I'd say around 12 seconds, although it's variable.

Operating the brake switch without actually braking is a fairly precise operation. It's possible that the brake switch is accidentally released momentarily over a bump or something. Even such a short blip (over the cutoff) will cause the gremlin.

I have pondered fitting a toggle switch to the second brake connection so I can just switch it off downhill and concentrate on potholes and wildflowers as I wouldn't have to be watching the speedo so much.

That's not what I asked. How long between 250w and 750w?

Finally got out today. There are two down slopes away from my house. Kept pedaling down both, achieved 23-25mph. At the bottom, wouldn't ya know it, the system behaved perfectly, the power kicked in at 15mph and ramped quickly up to max.

I thought that the only thing I had done differently was that I didn't use the brake, just kept pedaling.

The next hill was a long steep descent into a valley, I had to use the brakes so I didn't pedal at all. Once again, against expectation, it behaved perfectly. Having climbed out of the valley there was a small descent followed by a small rise before going down to the coast. Pedaled down the slope not using brakes. This time on the ascent the power did not kick in until about 8/10mph. As near as I can figure the ramp from 250w up to max (which today seemed to top at 720w) took over 10seconds maybe as much as 12.

This 10-12 seconds ramp from 250 to 720 seemed fairly consistent - when it happens, which I still can't reliably trigger or forecast. As the power comes on at an already low speed and start at around 70w the whole ramp up to full power is excruciatingly slow, especially on a steep climb.

I am thinking I'll disconnect the brake for the next ride - but I have a dim memory that I've already tried this.

1 hour ago, Benjahmin said:

Finally got out today. There are two down slopes away from my house. Kept pedaling down both, achieved 23-25mph. At the bottom, wouldn't ya know it, the system behaved perfectly, the power kicked in at 15mph and ramped quickly up to max.

I thought that the only thing I had done differently was that I didn't use the brake, just kept pedaling.

The next hill was a long steep descent into a valley, I had to use the brakes so I didn't pedal at all. Once again, against expectation, it behaved perfectly. Having climbed out of the valley there was a small descent followed by a small rise before going down to the coast. Pedaled down the slope not using brakes. This time on the ascent the power did not kick in until about 8/10mph. As near as I can figure the ramp from 250w up to max (which today seemed to top at 720w) took over 10seconds maybe as much as 12.

This 10-12 seconds ramp from 250 to 720 seemed fairly consistent - when it happens, which I still can't reliably trigger or forecast. As the power comes on at an already low speed and start at around 70w the whole ramp up to full power is excruciatingly slow, especially on a steep climb.

I am thinking I'll disconnect the brake for the next ride - but I have a dim memory that I've already tried this.

I have the answer now. The reason your controller ramps up the power is because it's slowing down. At the higher end of it's speed range, the back emf reduces the amount of current it can get. That's normal with any DC motor.I'm absolutely sure that what you describe is nothing to do with the controller. At the bottom of the hill, the motor is spinning at max speed, so the most current it can get is around 1 amp. As the bike sl9ws rown, the current will ramp up. The other effect is the clutch disengaging when the bike exceeds the motor's speed limit and the time it takes to re-engage when your speed reduces below the limit.

Is this a thing with AKM128's? When I first posted about this nearly all replies said that they'd never had anything like this with KT controllers. So, if it is the motor is there anything I can do? Does it need some lube? Although these motors are quiet, I ride very quiet roads and can hear the motor. I have pondered why I can continue to hear the same sound, from the motor, over the cut off speed when it is no longer driving.

Is it worth stripping down the motor?

I dunno nuffink bout deez fings, but if you have a throttle, you could test the back EMF theory?

Edited April 25Apr 25 by guerney

Went out yesterday. Downhill from the house I did the same as I did last time. Pedaled all the way down, no brakes. Boy did I suffer for it this time. On the rise the power did not kick in until 7mph. It ramped up from zero so slowly that I was out of the saddle just to keep moving (this is a very heavy bike). It's this inconsistency that drives me nuts. I think I've got the management of it then it throws a curved ball at me.

As this is a new controller I think it's fair to assume that it's the motor as per D8veh's theory/explanation. Not knowing in any detail the way motors are constructed, I find it difficult to follow your explanation D8veh. Would it be beneficial to remove the motor side plate and have a gander?