July 5, 2025Jul 5 I do have a 48v Dillinger battery of unknown age. Guess I could temporarily strap it to the rack. I'll have to run my 36v battery down to below 48v cutoff first, so that I can reset the controller after testing. you don't need to run anything down. just connect the 48v battery and it should work, but make sure that the 48v one is charged first, otherwise it might think it's a charged up 36v one.
July 6, 2025Jul 6 A good way to clear the controller before swapping batteries is to power up with no battery, That discharges the capacitors and it will try to recalibrate on the next battery. I found the label on my motor. Woo, I have to remove it as it's an 800W AKM, 48V 10.5 gearing. I took it out with a 36V pack. I don't have a 30 mph bike, and only 18mph legs, but my 20A controller in PAS 1 has no problem putting out 100-110W in PAS 1. It ran nice.
July 9, 2025Jul 9 Author Finally got time for more messing fettling. I remembered that the original 12 magnet disc was still fitted to the inside of the granny ring. So I removed it in case it was affecting anything. Doesn't looked like it was. Trying a selection of P1 settings. 96, 128, 160. 128 gave the nearest mph registration. Settled on 125 which seems to correlate pretty well to the satnav reading. However it is still giving no speed reading when not pedalling. However, for the terrain I need it for, this is definitely the wrong motor. I daren't take it on any of my regular more severe rides, which makes it a bit useless. So I'm going to order a 201rpm Q128. Looks like internal speed sensing is somehow up the spout. The T09S has a speed connection, presumeably for an external sensor. Would this one do the trick? https://topbikekit.com/kt-speed-sensor-bs02-for-ebike-p-255.html?zenid=t7hi6qc7bd9lcsd7v4hqueb2m0 Still riding without the brake switch. I really don't like it. Withou it changing down through the gears, as you approach junctions, gives power - which is unerving to say the least. Also riding shared paths (with zombie pedestrians and long lead dog walkers) where one has to be on and off the power whilst still pedalling to keep forward motion, is really iffy. I will be re-instating it.
July 9, 2025Jul 9 Author Been out again on what I thought was just going to be a short bimble around. Ended up having to be rescued AGAIN as the system through an 03 halls error. I'm thinking this is the result of heat damage, just not sure if it's in the motor or the controller. Don't know how to test the controller or the motor. Either way, I am loosing confidence in this system.
July 9, 2025Jul 9 Been out again on what I thought was just going to be a short bimble around. Ended up having to be rescued AGAIN as the system through an 03 halls error. I'm thinking this is the result of heat damage, just not sure if it's in the motor or the controller. Don't know how to test the controller or the motor. Either way, I am loosing confidence in this system. More likely, you don't have the motor connector in far enough. Show a photo of it.
July 9, 2025Jul 9 Author First thing I checked, 2nd was the plug at the controller end. I've taken a photo but I'm no photographer and my dumb phones no camera. But it's all the way in and I've also checked it for pins being bent or pushed out. Can't see any.
July 9, 2025Jul 9 If you tap the white signal lead at the end of the motor cable, and add 5v power/ground, you can power an external speed sensor if your motor sensor burnt out. Ive done this on a G60 bafang and also on motors that had no sensor. The trouble I had was a lot of electrical noise. It could be due to my choice of sensor or setup. When coasting, the display would jitter like mad above 12 mph. Under power, it was accurate. I'm speaking only to KT controllers. I've seen other posters with the same problem,
July 10, 2025Jul 10 There is another option to try and that is the phase /hall wire swapping , as we tell folks it is a gimme that colour to colour will work always out of thr box. I have posted many times the 36 combo chart to follow to find the best possible combo.
July 10, 2025Jul 10 There is another option to try and that is the phase /hall wire swapping , as we tell folks it is a gimme that colour to colour will work always out of thr box. I have posted many times the 36 combo chart to follow to find the best possible combo. That's always a possibility, but my Q128 always works colour to colour, and I've never heard of anybody having problems with an AKM motor before.
July 10, 2025Jul 10 Author Thanks guys. System did work with some glitches that I was trying to get to the bottom of. The 03 error was new. Meanwhile the weathers a wasting and I'm without a reliable ride. So I've bitten the bullet and ordered an new 201 rpm motor, and 8 magnet pas sensor and a speed sensor. With this I can hopefully get back on the road. Later I can get bench time with the previous motor to see if it can be repaired for a different project.
July 10, 2025Jul 10 03 Hall error will either be a wire fault or Hall failure . Test the Halls with a meter to see.
July 10, 2025Jul 10 Author Just looked up the GrinTech test instructions, I'll give it a go. This is what I got: R-Bla 4.3v constant W-Bla 5v constant Bla-Blu Bla-G Bla-yel All switching 5v-0v Couldn't get 03 error to come up whilst on the stand. Tried powering the wheel under a braking load but no error. Edited July 10, 2025Jul 10 by Benjahmin
July 17, 2025Jul 17 Author Got my new motor, 36v 201rpm, built into wheel and given initial test today. This motor is much happier on hill climbing. Just did a fairly long, gradually steepening climb on full power. A lot torquier than the last one. Stopped at the top of the hill to check controller temperature, it felt cool. Speed registration when pedalling is now pretty much spot on. However it is still registering 4-5mph low when coasting downhill. This I could live with. BUT there is still this problem. After a descent, when coming back onto power, it will only deliver around 150w for an unspecific period. I have no idea what eventually triggers it into full action. As there are many V bottomed valleys around here, I need to get this one sorted. Here are the current settings P1=168 This assumes 16 magnets and uses quoted reductuin ratio of 10.5 P2=0 P3=1 P4=1 P5=13 C1=07 C2=0 C3=8 C4=3 C5=09 C6=3 C7=0 C8=0 C9=0 C10=n C11=0 C12=4 C13=0 C14=2 Any idea's anyone?
July 17, 2025Jul 17 Got my new motor, 36v 201rpm, built into wheel and given initial test today. This motor is much happier on hill climbing. Just did a fairly long, gradually steepening climb on full power. A lot torquier than the last one. Stopped at the top of the hill to check controller temperature, it felt cool. Speed registration when pedalling is now pretty much spot on. However it is still registering 4-5mph low when coasting downhill. This I could live with. BUT there is still this problem. After a descent, when coming back onto power, it will only deliver around 150w for an unspecific period. I have no idea what eventually triggers it into full action. As there are many V bottomed valleys around here, I need to get this one sorted. Here are the current settings P1=168 This assumes 16 magnets and uses quoted reductuin ratio of 10.5 P2=0 P3=1 P4=1 P5=13 C1=07 C2=0 C3=8 C4=3 C5=09 C6=3 C7=0 C8=0 C9=0 C10=n C11=0 C12=4 C13=0 C14=2 Any idea's anyone? P2 should be 1 or 6, not zero. That should sort out your speed reading as long as you have a working speed sensor.
July 17, 2025Jul 17 Author I'll try on next ride. On the stand 6 seemed to be giving an odd reading. I fitted an external sensor that I bought with the motor. I couldn't get any reading from it on any setting. May have been too much space between wheel magnet and sensor. Zero seemed to give the best. It's this power delivery delay that's the real problem.
July 17, 2025Jul 17 I'll try on next ride. On the stand 6 seemed to be giving an odd reading. I fitted an external sensor that I bought with the motor. I couldn't get any reading from it on any setting. May have been too much space between wheel magnet and sensor. Zero seemed to give the best. It's this power delivery delay that's the real problem. I think that when you pedal hard past the speed limit the controller decides that it has massively overshot the setpoint and then over compensates by reducing the power for a period. I don't think that changing the settings will make the blindest bit off difference because it is hard coded in there. I'm currently experimenting with modifying a Lishui controller to behave like the KT. The hope is that it will get rid of this particular problem and also be more adaptable. It is working on a bench test but I haven't had the time to install it and do a road test.
July 17, 2025Jul 17 Maybe brake to reach the dip ghost pedalling at 14mph so that the controller's all set for the climb? On mine brake cutoffs respond faster than PAS or speed limit. * I'm currently experimenting with modifying a Lishui controller to behave like the KT. WLT hear more. * doh, there was a similar suggestion 90 posts ago.
July 17, 2025Jul 17 Maybe brake to reach the dip ghost pedalling at 14mph so that the controller's all set for the climb? On mine brake cutoffs respond faster than PAS or speed limit. * WLT hear more. * doh, there was a similar suggestion 90 posts ago. I'm keeping notes. When (and if) I get it fitted I will start a thread.
July 17, 2025Jul 17 I'll try on next ride. On the stand 6 seemed to be giving an odd reading. I fitted an external sensor that I bought with the motor. I couldn't get any reading from it on any setting. May have been too much space between wheel magnet and sensor. Zero seemed to give the best. It's this power delivery delay that's the real problem. Have you still got the white wire on the motor connected to the white on the controller or did you hijack it for the external sensor? Let me explain how that setting works. if you have a speed sensor, you set P2 =1, or 6 if your motor has 6 magnets in the hub for its internal speed sensor. I've never seen an AKM with 6 magnets, though they could always change for some reason. If you have no speed sensor, you set C2=0, then it'll use the motor halls for the speed signal, but that shows the motor speed, not the bike speed. When you freewheel or have the motor switched off, the speed should show zero, though the motor will still be spinning at its max speed when you go downhill if the power is still on, so that's the speed you'll see on the LCD, not the bike speed. If your external speed sensor is all wired up, you set P2=1 and adjust the gap to 1mm so that it works, then you get the correct speed at all times, assuming that you've set the correct wheel size. Even if P2 is et to 1, the speed signal will still default to the hall sensors if there is no ouput from the speed sensor.
July 17, 2025Jul 17 Author Back of the net ! Somehow I'd got the info that AKM had 6 magnets. P2 now set to 1. It's sorted it all. Coasting speed registration now correct. Power pick up also now correct. Thanks once again Saneagle.
July 17, 2025Jul 17 Back of the net ! Somehow I'd got the info that AKM had 6 magnets. P2 now set to 1. It's sorted it all. Coasting speed registration now correct. Power pick up also now correct. Thanks once again Saneagle. That motor is a torque monster. When you upgrade to 48v, it becomes a winch.
July 18, 2025Jul 18 That motor is a torque monster. When you upgrade to 48v, it becomes a winch. This true. I rode our Biria/AKM128 yesterday. It flew up the hills in PAS 3. Well, it's pulling 800 watts too.
July 18, 2025Jul 18 Author I can confirm this. Been up one of my longer climbs at a surprising speed. Speed registration is now consistant and seems accurate. Power cut off and back in is smooth and subtle. I do still have one issue, however. On my ride there's a long descent. I reach 30-35mph going down it. The bottom is v shaped. Obviously, as I start to climb, I'm over the cut off and am going through the gears as I'm slowing down. I was down to around 8or9 before the power started to kick in, but it ramps up very slowly, so that by the time it gets to around 350-400w (useful power on steep hill) I'm down to about 6mph. As it reaches full level 5 power speed does pick up. On another milder descent I reached around 25mph, so was still pedalling. Same thing again, as I hit the rise I was down to about 12 before power started to slowly ramp up. I can't find what evntually triggers pas to respond. Tried going to zero power then back up to 5. Tried the throttle. I have two questions. Could it be anything to do with C14? Not sure what this means: C14 is the parameters of power-assist tuning setting, with the default value of 2. The power-assist is between1-4 gear, and it is invalid until P3 equals to 1. The setting range 1-3, and press (UP) button or (DOWN) button for short to make selection. C14 parameter definition table: C14 Value Assist strength of intelligent pedal motor 1 Weak assist strength of motor 2 General assist strength of motor 3 Stronger assist strength of motor Would setting 3 help? Could it be because I'm using a 12 magnet one piece sensor? I do have a 10 magnet two piece that I could put on.
July 18, 2025Jul 18 I can confirm this. Been up one of my longer climbs at a surprising speed. Speed registration is now consistant and seems accurate. Power cut off and back in is smooth and subtle. I do still have one issue, however. On my ride there's a long descent. I reach 30-35mph going down it. The bottom is v shaped. Obviously, as I start to climb, I'm over the cut off and am going through the gears as I'm slowing down. I was down to around 8or9 before the power started to kick in, but it ramps up very slowly, so that by the time it gets to around 350-400w (useful power on steep hill) I'm down to about 6mph. As it reaches full level 5 power speed does pick up. On another milder descent I reached around 25mph, so was still pedalling. Same thing again, as I hit the rise I was down to about 12 before power started to slowly ramp up. I can't find what evntually triggers pas to respond. Tried going to zero power then back up to 5. Tried the throttle. I have two questions. Could it be anything to do with C14? Not sure what this means: C14 is the parameters of power-assist tuning setting, with the default value of 2. The power-assist is between1-4 gear, and it is invalid until P3 equals to 1. The setting range 1-3, and press (UP) button or (DOWN) button for short to make selection. C14 parameter definition table: C14 Value Assist strength of intelligent pedal motor 1 Weak assist strength of motor 2 General assist strength of motor 3 Stronger assist strength of motor Would setting 3 help? Could it be because I'm using a 12 magnet one piece sensor? I do have a 10 magnet two piece that I could put on. Anecdotal evidence because I haven't tested it thoroughly. Setting C14 to 3 reduces the occurrence of this behaviour. It also makes it a bit too powerful, so a reduction in C5 is needed. This behaviour definitely happens with C14 is set to 2. Setting C14 to 1 doesn't work as advertised. It stays fixed at a set power (say 80W for example) no matter what you set C5 at.
Join the conversation
You are posting as a guest. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.