Fiido D2 conversion to KT controller upgrade not working (noisy motor and cuts out after 5 seconds)

Hi all, Not sure if anyone can help, but i've been pulling my hair out over a problem with a controller upgrade on my Fiido D2 and wondered if anyone had any pointers or ideas as to what i can try next to get it to work ?

 

The old controller on the bike is a WK3615YS 2019.08 36V 120degree 13+-1a F1 181810110962 and was working just fine.

 

The back wheel hub motor has this written on it 16" 36v250W19H7789 FIIDO 4

 

The new controller is a KT36/48SVPRD-HRD01

 

The reason i upgraded the controller was because i broke my leg in 3 places a few years ago in a bad motorbike crash and its never fully healed and is very weak and makes even walking a bit difficult, I struggle to get up steep hills and thought an Ebike would be the answer, but the Fiido D2 doesn't quite have enough power the assist me up the hill, so i thought i would upgrade the controller for a bit more power (I don't want more speed, I'm not interested in going fast, I just more hill climbing power)

 

The upgrade went well, and the new controller fitted in nicely, I left the pedal assist disconnected as i only want to use the throttle to assist me on the hills and find the power kicking in from pedal assist a bit unbalancing at times. With throttle only, I know where i am when adding power and its easier than pedal assist for me.

 

The problem i now have is that the motor is very noisy when running, I havent test ridden it yet, just test run it with the back wheel off the ground at home, It sounds a bit like a phase mismatch to me and after running the motor for about 5 seconds, it just stops and i have to let go of the throttle and reapply to get the wheel moving again.

 

I have tried swapping back to the old controller to see if the motor was ok, and it runs nice and smooth again on the old controller, but its still noisy and cuts out on the new one.

 

My P settings are:-

P1 100

P2 1

P3 1

P4 0

P5 0

 

My C settings are:-

C1 02

C2 0 (I think this is the phase control settings, and have tried 1,2,3,4 and 5. Each of them makes very little difference to how it sounds and runs)

C3 8

C4 0

C5 10

C6 3

C7 0

C8 0

C9 0

C10 N

C11 0

C12 4

C13 0

C14 2

 

I also dont have a speed sensor wire comming from the motor in the hall sensor plug, could this be the issue ? I was thinking that the hall sensors themselves could detect motor rotation and start point ??

Pedal assist plug is left unplugged as i didnt want it.

I'm using a Wuxing thumb throttle and an KT LCD5 display

 

Help !!, does anyone have any ideas what the problem might be ? any suggestions or setting changes to get it working ? and is the motor even compatible with this controller ?

 

Thankyou

There is no need for the speed sensor wire to be connected. The halls in your motor take care of that on the KT controller. Your motor is either a bafang G020 or a clone. If it’s a G020 then P2 needs to be set to 6. If it’s a clone then you may have to fiddle with P1 and P2 to get it to run efficiently and for the speed to be accurate. P2 is likely to be either 1 or 6. But you can do that after you get your motor to spin correctly.

 

Leave C2 at 0

 

The good news is that you have got it almost working and it looks like you have to go through the process of swapping phase wires to get the motor to run in the new controller.

 

Here is the test sheet. There are only a few combinations that will work.

 

https://www.pedelecs.co.uk/forum/threads/help-needed-for-carrera-crossfire-throttle-convesion-please.45028/#lg=attachment50994&slide=0

Edited April 7, 20233 yr by AGS

There is no need for the speed sensor wire to be connected. The halls in your motor take care of that on the KT controller. Your motor is either a bafang G020 or a clone. If it’s a G020 then P2 needs to be set to 6. If it’s a clone then you may have to fiddle with P1 and P2 to get it to run efficiently and for the speed to be accurate. P2 is likely to be either 1 or 6. But you can do that after you get your motor to spin correctly.

 

Leave C2 at 0

 

The good news is that you have got it almost working and it looks like you have to go through the process of swapping phase wires to get the motor to run in the new controller.

 

Here is the test sheet. There are only a few combinations that will work.

 

https://www.pedelecs.co.uk/forum/threads/help-needed-for-carrera-crossfire-throttle-convesion-please.45028/#lg=attachment50994&slide=0

 

 

 

Ahh, excellent information, thank you so much for replying !!, I have just tried setting P2 to 6 and its slightly worse, I'll have to google P2 again as i cant remember what that one does, but i'll now try swapping the Yellow and Green phase wires and get my mulltimeter setup on it to monitor current draw and will report back.

P2 is the number of magnets in the motor that are used for the speed sensor, could be anything from 1 to 6 depending on the motor.

 

P1 = The number of motor magnets divided by the gear ratio.

Edited April 7, 20233 yr by AGS

P2 is the number of magnets in the motor that are used for the speed sensor, could be anything from 1 to 6 depending on the motor.

 

P1 = The number of motor magnets divided by the gear ratio.

That makes sense, I know its got 20 magnets in it for the drive, and we are starting to get a little bit beyond my knowledge here !, but are there a different set of magnets used for the speed (hall?) sensor then ?

Yes separate magnets for the speed sensor. They are located in the side cover.

Good news !!!

 

Its running really nicely, and sooo quietly too, much faster than any of the wiring configurations tested so far, should i just accept this one and use it ? or do you think i should test all 36 combinations to see if there's an even better one ?

 

The good working one was the 8th wiring configuration tried so far.

 

I've got yellow and green sensor wires switched and yellow and green on the motor phase wiring switched as well, and its only drawing 9amps under full throttle, no load (wheel in the air), the other previous tests have been drawing close to 20 amps and running horribly and very slowly.

 

I'm tempted to just put it back together and give it a test ride ?

 

Also the speed display doesn't work either, hasn't on any of the tests, but i'm not worried about that really as long as it doesn't affect anything else by it not working ?

It’s not worth trying any of the other combinations if it’s running smoothly. There are only 3 combinations that will work and it looks like you have found one of them.

 

You still have to sort P1 and P2 out which is your next challenge. So ride it and see if the speed reading makes any sense.

 

Although 9 amps still seems quite high under no load, so be careful when you do your load testing. You don’t want to blow your brand new controller up.

 

It should be under 100 watts with no load, 3 amps or so.

 

I should have mentioned that you should use gentle throttle when testing, because the wrong wire combination causes a high current to be drawn.

Edited April 7, 20233 yr by AGS

If you don’t get a speed reading with P1 at 100 and P2 at 1 or 6 then try P1 at 87. If that still doesn’t work then you are on your own, because I have run out of ideas.

 

Oh and I think you need to actually rotate the pedals for the speed to display. I don’t think you will get any reading if you are just using the throttle. And also the controller may switch itself off after 5 minutes if it doesn’t detect a pas signal. Although I’m not sure about either of these. It’s just something in the back of my mind.

Edited April 7, 20233 yr by AGS

Also with a more powerful controller you may find pas more manageable now because you can ghost pedal to get it to work. And adjusting C14 to 3 may also help.

Edited April 7, 20233 yr by AGS

I will plug the pedal sensor back in then and see what its like, I guess i can always programe it to off even with it plugged in, I guess i was going the lazy route by simply not plugging it in !

 

And I will give the P1 settings a try, thanks, but i'm not worried if it doesn't work, I guess i could always wire in an external sensor with a single magnet to get it working.

 

And after you mentioned that you though the amps were a bit high under no load, I decided to carry on the testing of all 36 wiring configs, and I have found another wiring config that gives identical readings, nice and smooth, fast and a 9amp draw, so i'm now thinking that as there are 3 smooth working wiring configs, but only 1 is best, I've probably found the 2 smooth running configs that draw high current, so will carry on testing the find the last smooth running one and report back, I'm on wiring config number 22 out of 36 at the moment !!

I think all 3 will give the same result. Are you using a 36v or 48v battery?

 

Just double check on the display to see how many watts it says when you are running at no load. 300 watts seems high to me no load.

 

Edited: just realised the lcd05 doesn’t display watts.

Edited April 7, 20233 yr by AGS

Is there a display that shows watts whilst you are riding then ?, I did want the KT LCD9 but couldnt find anyone with them in stock in the UK and i ordered from TorqueTech in the end but they only had to LCD05 in stock

I think all 3 will give the same result. Are you using a 36v or 48v battery?

 

Just double check on the display to see how many watts it says when you are running at no load. 300 watts seems high to me no load.

 

Edited: just realised the lcd05 doesn’t display watts.

 

 

You're absolutely correct !!, I just found the third smooth running configuration (typically, it was the 36th test!), and they all draw exactly the same current and sound as smooth as each other, so i'm guessing it doesn't matter which of the 3 i use now then ?

Any one of them will work.

 

Do a bit more testing before taking it out now that you have reconnected the pas.

 

Try pas on no load and see if you get a speed reading. It will take several seconds to kick in because it has to count a certain number of magnet passes before it can do its calculation.

 

And you can also monitor the battery current draw at the same time seeing as you have the equipment to do so.

 

LCD03 shows watts while you are riding, LCD08 is a bit over the top, but LCD05 is perfectly adequate.

 

Torque Tech are pretty good and sell sensible stuff. I bought my controller from them. They aren’t the cheapest around, but the service was good and you know your not buying a knock off.

Edited April 7, 20233 yr by AGS

This is above my pay grade, but it may help?

As others have said, 9A with no load doesn't seem right? Carefully does it!

The ammeter may be reading incorrectly.

Right, back again, got called away for shopping duties with my wife !, but now its dark i wont be doing the test ride until tomorrow, One thing i have just found is that my back hydraulic brake was binding and making the wheel stiff to turn, I don't think the caliper liked being upside down for so long, so i've just removed the caliper and its drawing 4.5Amps with no load now, so it looks like i've been fighting a sticking brake caliper for all my tests today.

 

I'm using a DC clamp meter for the readings, its a UNI-T UT210E and has been pretty accurate and reliable in the past, much easier and quicker than breaking into the circuit to put my Fluke meter probes in (and its only rated for 10A anyway)

 

Thankyou for the Video link Mikel Its a visual method of determining the correct wiring which is the same as the link AGS gave me in the second post, but it was interesting to see that i didn't need to do all the 36 different wiring combos !

Yes brake binding makes sense with the current reading you initially got and 4.5A no load is much more sensible.

 

If you are still running a 36 volt battery you will find range limitation a problem due to voltage sag. You could upgrade to a 48 volt battery with your new controller. Although finding a 42 volt battery that fits into your frame may be a challenge.

 

You could even push it to 52 volts, but the controller can’t deal with the minimum voltage cut off for a 52 volt battery and you will need to rely on your battery BMS to protect the battery from discharging too much.

Edited April 8, 20233 yr by AGS

Back again, I've been out for a test ride now, and its running nicely, slightly nosier than the old controller, but it doesn't have a much torque for climbing hills, less than i had previously, and on measuring the current under load (very tricky!) its maxxing out at 19.5 amps but seems less powerful than it used to be, and when i try to move from completely stopped, it feels like its cogging and struggling to start and i have to give a little push off with my feet to get it going properly.

 

The speedo started working after about 30 seconds of riding and seems fairly accurate, it maxed out at about 15mph on the flat, got to 21mph going slightly downhill, but on uphills it slows down more and more until it just runs out of power.

 

The motor and controller and battery stay completely cold, even after 3 miles of riding at full power, so i'm not sure what the issue can be ?

 

Could it simply be an incompatibility between my new controller and my existing motor ?

It’s probably due to voltage sag on the 36 volt battery. When you ride it next keep an eye on the battery voltage. I think it’s sagging too low and the BMS is kicking in because of the higher current demand from the controller.

 

You can limit the controller current to prevent this, but then it will offer no benefit over the original controller.

 

P1 and P2 also play a part in motor efficiency, so you may need to work on those as well as previously discussed.

Edited April 8, 20233 yr by AGS

Back again for an update.

 

The bike is riding ok now, but it has less torque for hill climbing than on the old controller, but does seem to have a little bit more top speed on the flat.

 

I dont think its voltage sag or the bms holding the power back as i'm drawing a solid 19.7 amps going uphill at full throttle with a steady 35.8v showing after about a minute of hill climbing at full power as the lowest voltage, this recovers is seconds back to about 38.5v once off the throttle.

 

I've played with P1 and P2 now, and only P1 seems to make any difference and it very very minimal, It seems best set at 90 - 100.

 

I've also been playing with C2 and this makes quite a big difference to the smoothness, power and quietness of everything with C=3 and C=5 giving the best results, setting C to 1,2 or 4 makes it cut out a few seconds after applying the power and you have to release the throttle and reapply to make it work again for a few more seconds, C=0 seems to be worse than 3 or 5 in terms of noise and power.

 

So i'm now thinking that my motor (as detailed in my first post) just isn't compatible with the KT controller i've chosen and wonder if the old controller was a square wave and moving to this more powerful sine wave controller has caused all the issues.

 

So i'm now wanting to keep the KT controller as they seem to get good reviews and now buy a new rear hub geared motor and i'm thinking that the MXUS seems to fit the bill for best compatibility ? what does anyone think ?

I have 16" wheels, so i'll get suitable spokes and a rim and lace it up myself and am hoping to get to the 500W of power that i'm after for really good hill climbing (i'm not worried about top speed and don't want to go fast at all) and reading lots of reviews and forums, it seems that as long as i can find a decent and KT controller compatible hub motor, then everything should be perfect !!

 

So does anyone have any motor suggestions ? or other ideas that i might try ? and if the battery does prove to be not up to it, then i am experienced at battery building and happy to build a better spec battery, even going to 48v if necessary (but would prefer to stay at 36v to use the existing battery i guess)

 

Thankyou for all your help and suggestions with all my troubles with this bike, and it was really interesting to learn about and figure out all the different wiring configurations !