Brand new ebike conversion dead within 50km

[Heatserve]

Hi all

I'm new to this forum and to ebikes, so please bear with me if I get any of the terminology wrong.

OK, I bought an ebike kit and triangle battery from the same eBay supplier after confirming that all components were 100% compatible with each other. The kit consisted of:-

KT-LCD3 display
1500W rear motor in a 26"*4" double rimmed fat bike wheel
48V 1500W 45A controller
Half twist grip throttle
Replacement brake levers
Pedelec sensor
48V 20ah battery

I then got a local bike shop to do the conversion using a 17" fat bike.

The guys from the shop were a father and son team from the UK and did a fantastic job.

Anyway, the job was finished at about 5pm and I rode it for about 35km and everything was perfect.

The next morning I took the bike out again. I was using the minimum assist level and the wattage was approximately 250W from the display.

After about 10 km, the power seemed to momentarily disappear and then resume again. This happened a few times.

Up until now, the ride was on relatively flat ground but after about 14km I had to go up quite a steep hill.

I changed the assist level to 3, but the bike didn't really seem to perform well, so I changed it to 5 and was getting an indicated 2Kw to the motor, but it really didn't feel like anywhere near that.

At the top of the hill I changed back to level 1 at which point I lost all power.

The display was illuminated, but the battery icon was showing blank. At this point I had well over 50 volts in the battery.

I checked for obvious faults but couldn't find anything. The motor was warm, but not hot, but the controller was absolutely boiling.

Hot as in too hot to touch.

I contacted the supplier who suggested changing the C5 setting to 5, which I have done but this didn't restore any kind of functionality. After a bit of a discussion with the supplier, he has agreed to send out a new controller.

I now have a few queries which I hope someone can help me out with.

It seems, according to Google, that these sine wave controllers run fairly cool. In fact, I can't find any reference to one overheating. The controller is housed in a bag to the rear of the bike, visible in the photo, instead of the supplied bag. I live in Thailand, which is pretty hot, but nowhere near hot enough to adversely effect the controller temperature I would have thought. It was pulling maximum power for less than a minute, but unfortunately I have no idea of what temperature the controller was at prior to the hill. Anyway, does this sound like a bad controller or is excessive heat a feature that I should be aware of?

Secondly, I was sort of hoping that the controller might have some sort of thermal cutout that might have operated, but as it's still broken and in the absence of a reset switch, this seems unlikely. So, my question is, is there any kind of possibility that I can repair or recover this controller? There's no such thing as an ebike shop even remotely close to me.

Thirdly, after endurance testing and making some gearing changes to the bike, I was hoping to use it for touring. I now am concerned about reliability. Generally, are these controllers reliable and have I just been unlucky? Is there a better controller that I should be looking at?

Apologies for the length of this post and the fact that it's been a bit of a brain dump. I'm a little frustrated that an expensive project has failed so quickly.

Regards
Mike

Edit: just to add, I did not at any time have any error conditions displayed via the screen.

Photos attached

 

[Jon NCal]

When you get the new controller, make sure to mount it in the open air on the frame, not in a bag. It needs air circulation to dissipate heat.

 

[Heatserve]

Hi

Thanks for the reply and I guessed that someone would mention this.

Here's what I don't understand. The controller was supplied with a bag and is obviously intended to live there. I haven't seen other builds with the controllers free to the open air, in fact many that I have seen are supplied in plastic boxes.

It's not really practical to mount the controller on the frame. I could however, keep the bag unzipped to provide air circulation, which I'd planned to do in order to regularly monitor its temperature.

So, back to one of my questions, do these things run really hot when pulling just 5 or 6 amps?

 

[Drunkskunk]

See all those fins in the design of the controller? They are there to be... Cooling fins. They aren't cosmetic, but necessary for radiating heat. That controller can reach 200 degrees in normal operation, and needs fresh air over those fins to keep cool.

Yes, some vendors supply a bag or a box to stuff the controller in. and most of the time, a controller won't get more than 10 degrees over ambient. But most of the time isn't all of the time. Under some conditions, it's going to get hot. Many Vendors aren't usually design engineers. they are salesmen. They don't know better.

All that said, it sounds like your controller died from other issues. The key might be when you were pulling 250W on assist 3. 250w would be enough power for ~17-19mph on throttle alone, with no pedal input. If you were going slower and pedaling, you already had a problem.

If the motor was kind of pulsing, thumping, or vibrating while running at that point, you had a blown FET. The controller was bad, and running it made it worse until it died. Just a faulty part.

If the motor was smooth, you might have had a false positive phase/sensor combo. The color wires don't always match, and some motor controller combinations will run one one of the wrong possible combinations of phase and hall wires. It seems smooth, but thee free running speed is overly high, the torque is low, and the motor will draw far too much power for the work done. It will also cause the controller to burn up.

 

[Heatserve]

Thanks for the reply.

The motor was running fine, all very smooth with seemingly no problems.

Nothing was really under any stress at all and as I mentioned, the motor was warm, but everything is warm in Thailand. It definitely was nowhere near hot.

I'm hoping I was just unlucky but will keep the bag completely unzipped to allow maximum cooling.

When the new controller arrives I'm going, at least initially, limit it to 40kmh and 30 amps and see how it goes.

Fingers crossed!

 

[Alan B]

The controller clearly states on the label that you showed that its rated current is 22 amps, and the maximum is 45 amps. This is a 1kw continuous rated controller. It should not be run over 22 amps for more than a short burst, and it will need lots of cooling. Heating goes up roughly with the square of the current, so operating at 45 amps will produce four times the heat. Actually it is even more because as the FETs warm up their resistance increases and the heat produced is even larger. Unzipping the bag is not likely to be satisfactory, especially if the ambient temperature is warm.

Fat tires take a lot of power to operate. Try pedaling without motor power to see what the motor is working against. Generally the motor and the controller's temperatures will be similar if they are well matched and getting good airflow. Putting the controller in a bag is an invitation to problems unless you get a controller that is rated for a lot more power so it is not working hard.

Good luck with your project.

 

[wesnewell]

If you want a third person to tell you you burned the controller up by enclosing it in a bag, consider it done. BTW, the photo of the controller shows it rated for 22A. That's the continuous amp rating with a max amp of 45A for short periods, not continuous. If you don't want to burn up another one, I'd suggest you mount the controller in the open air or make sure it gets plenty of airflow on all sides of the controller in the bag. Which may prove to be near impossible. Good luck.

 

[Heatserve]

Thanks for the replies. I can reduce the power to not exceed 20 amps I believe.

Has anyone got any suggestions for a more suitable controller?

 

[Heatserve]

I have queried the supplier to ask whether this controller is an appropriate match for the battery and the motor. Remember that I asked the same question before actually buying the kit.

So, can I ask, does everyone here have their controllers completely exposed? I only ask because I have never seen this on an ebike.

One other thing, as I said before, for the vast majority of the time the bike was drawing about 5 amps. Given the air circulation problems by putting it in the bag, would you consider this was enough to start to kill the controller?

Thanks

 

[amberwolf]

If you don't have airflow over the controller, then it needs to be either run nowhere near it's ratings so it doesn't generate heat in the first place, or it needs to be mounted in some way that pulls heat out of the controller to the air.

It doesn't have to generate much waste heat to get damagingly hot inside--soldering irons may only take 15-30 watts to melt solder (which takes between 600-800F depending on the kind), and that kind of heat would be enough to destroy most of the electronics inside a controller, before any of the solder melts, if it were completely enclosed inside something.

If you want to verify if the controller is generating heat, you can put it in the bag, with a remote thermometer sensor (like from an indoor/outdoor thermometer) taped to the flat surface near the fins, and watch it while you ride.

It is possible that it isn't that which caused the failure, but given that the controller was too hot to touch, I'd bet that that is exactly why it died.



Not everyone has their controllers out in the airflow...but some of them have had fried controllers or odd problems that are heat-related (or probably are; most never report back what actually happened).

A number of those not out in the air are inside metal boxes that conduct heat out, or have airflow thru the bag or box that helps at least a little.

One of the commercial ebikes, Cemoto, that I'm working on for someone, has it's controller inside the aluminum toptube, right up at the seattube, but it is also a low-powered geared hubmotor that never draws lots of current, so what little heat it does generate can easily escape thru the aluminum frame.

 

[Heatserve]

Thanks

I think I can get it on the seat post and still conceal the connections in the bag.

It'll also make it very easy to monitor the heat from the controller when I'm on the bike.

And also, on the plus side, means I can put other stuff in the bag :lol:

Also, thanks to everyone who have helped me understand this a little better.

 

[d8veh]

I don't believe that your problem is related to controller cooling. I'm pretty sure that the controller is protected by an internal thermistor, which simply cuts power while the temperature is too high. That wouldn't happen until you were running on maximum power for a while. I've been using KT 12 mosfet controllers enclose in bags for years, and I've never had a problem with cooling; however, I don't run continuously at high power. I wouldn't argue against mounting your controller out in the airflow if you can because it is just about still possible to get the odd warm day in the UK.

Instead, there was some other problem in the wiring to the motor, like a bad connection on the hall or phase wires, or you had a false positive phase/hall connection. I'm not sure what happens with a KT controller when you have a false phase/hall connection because, when using PAS, it limits current to a fixed value, while other controllers don't.

A warning to anybody. Do not ride your ebike if it isn't running properly. Consequential damage to the controller will follow.

The fact that your current was high when the motor power was low implies that it wasn't firing properly on all mosfets. What sort connector/s do you have on the motor?

P5 has no effect on the working of the controller. All it does is dampen the movement on the LCD battery display.

The setting that can affect or contribute to your problem is P1, which is the reduction ratio x the number of magnets in your motor, which means the number of magnets if it's a direct drive motor. I don't think that P1 has a big enough effect to make your motor cut out, but it can make it lose sync with the controller, which would make it noisy and rough. If you carried on like that, damage could follow.

 

[Heatserve]

It's the C5 setting that I've been advised to change. I'm going to start with a setting off 4, which, according to the supplier is the lowest I can go.

I wasn't running at high power. According to the display I was running at about 250W ~ 5 amps.

The motor seemed to be running great. With me pedalling as well, the bike was running nicely at about 20mph.

Nothing was hot except the controller.

Maybe I just had a bad one. The new one is on it's way to me now.

If I take the bag off the trayless rack, the controller can be nicely installed with 4 cable ties using the controllers mounting points. A better solution will be to get an aluminum rack with a tray and mount the controller directly onto it, hopefully the rack will then act as an additional heat sink.

The controller is a hallomotor one. I don't think it's possible to wire it incorrectly, but of course that doesn't mean all the connections are good.

 

[dogman dan]

The bag certainly didn't help, once that controller started running hot.

But I do smell a rat here, something else was up. Not sure what, but a false positive forward wire order would do er easy. Running, seeming fine, but wrong phase wire order will blow a controller pretty quick. Just as likely, a random darn defect in the controller.

Do mount the new one out in the air, and if it starts to seem hot, start looking for that rat. Not later, immediately.

 

[Heatserve]

I'll update this thread when the replacement controller is fitted.

 

[d8veh]

The fact that your supplier told you to change P5 implies that they don't have a clue what they're doing. I'm not surprised that it didn't work.

You can't always connect motors to controllers by going colour to colour because they don't all use the same sequence standard; however, if the supplier has sold similar kits to other people that didn't have the problem, then it shouldn't be an issue. If they're selling mix-and-match kits, then it could.

 

[amberwolf]

The fact that your supplier told you to change P5 implies that they don't have a clue what they're doing.

Please note that the OP has stated in the first post and later on that it is *C5*, not *P5* that is what was told to be changed.

I don't know what any of the settings mean or if C5 is any better to change, but just want to make it clear what was said so the troubleshooting is accurate.

 

[beast775]

Check to see how much air youre running..my fatbike will eat my controller at 6lbs air pressure on a steep hill.

 

[Wolfeman]

Heatserve, the controller really needs to be in the breeze. I attached a pic of how mine's mounted, right on the lower front part of the seat tube with the wires running straight up into the rear opening of the frame bag. Even flying in the wind, it gets pretty warm to the touch when I'm asking it to move a bunch of electrons back and forth.

 

[Heatserve]

The C5 setting limits the power that the controller draws.

I don't know where the confusion between the P and C setting comes from.

I have repositioned my controller as the photo shows. The box below it is to conceal the connections, so I'll just need to waterproof it I think.

Still waiting for the new controller, but any comments regarding my positioning welcomed.

The tyres were pretty low, so I've brought them up to 20psi

 

[wesnewell]

That's where I mount my controllers too. Plenty of airflow there.

 

[dogman dan]

And mud, if you ride in the wet. I sort of like to put a rear rack on bikes, just to carry the controller, when they are that big.

Then a home made mud guard can keep most of the shit off it, except for actual rain. But a muddy ride after a rain no problem.

 

[Heatserve]

I'll be happy if it just works haha

 

[Heatserve]

Just to update this thread:

Through a strange series of events I have ended up with 3 controllers being shipped to me.

Due to a delay via DHL, the Chinese supplier very generously sent yet another new controller via air freight at no charge and this arrived today.

So, I've installed it and set the power to about 2/3 of maximum and set the maximum speed temporarily to 40kmh whilst I monitor the heat generated by the controller. If all goes well, I'll remove these constraints later.

I took it out for a quick 16km ride and all was well. Very smooth and no excessive heat from either the controller or the motor, which seemed to be more or less the same temperature.

Hopefully, this will be an end to my problems.

Many thanks to everyone who helped me.

Sent from my ASUS_Z00AD using Tapatalk