Sensorless Brushless E-Bike Controller

[Johnbear]

Again I had to see if i could find a reliable controller for my sensorless tongxin motor so I contacted Brett from solarbbq, he sold me this, -$65ish with shipping $35 brokerage fees. I thought it should be fairly straightforward because there is just the 3 phase wires, throttle and power but I cannot get it to work, any suggestions, I am at a loss here. Thank you for any input!

 

[Jeremy Harris]

Interesting, as it seems to have a heck of a lot of connections for a sensorless controller.

The power in cables are easy, presumably the red/black Andersons. The throttle looks to be the three small yellow, red, green wires. The two brown wires with the black connectors are most probably the brake inhibit wires. The motor phase connections look to be the yellow, green, blue Andersons, but the really suspicious looking wires are the ones (five?) going in to the white multipin connector. These look to be Hall sensor connections to me, which doesn't jive with this controller being sensorless.

There is the possibility that this isn't a true sensorless controller, but perhaps a pedal-first controller. Pedal-first controllers don't need the Hall sensors, because they sense the motor back EMF to sync up the phase outputs, but they won't work with a geared freewheel motor like the Tongxin, as the motor needs to spin up under pedal-power before the controller will provide power. No matter how much you pedal a Tongxin, the chances are that the motor won't spin, which means a pedal-first controller may not work.

Jeremy

 

[Johnbear]

Thanks for your reply Jeremy! That was my suspicion as well but Brett assured me it was sensorless, the decal does also indicate sensorless. I added the andersons! It came with generic connectors. I thought it odd that there was the 5 pin connecter that looks like a sensor connection. It would be super if there was a functioning sensorless controller for the tongxin. Maybe this one does not work?

 

[fechter]

It might be pedal first.

You might try spinning the wheel backward on your tongxin while giving it some throttle and see if anything happens.

 

[gmouchawar]

Most controllers also have a key switch. If it is not close it will not power the wheel. If you have a pair of wires that are not label they could go to the key.
Try shorting them.

 

[Johnbear]

I got it to work at 24v on a sensored motor so it would appear as if it is not sensorless after all

 

[diver]

wow do some of the vendors even know what they are selling. tell you what you want to hear to make the sale then most likely will tell you that is not what you told me or i would have not sold you this.Ive been there several times thinking it was me and that i just did not explain it right. Like my 20 amp sensorless controller for a 5304 that was sold to me and then they said oh no it wont run a 5304..lol .On that one i know as i have the email lol.. They just don't listen to the customer i guess and just want to make the sale. some vendors i think go over there sales record and think i did not sell him this motor i sold him that one so he must be mistaken . I don't think they realize that there are more then one vendor out there that maybe one bought from, because maybe that vendor was out of stock at the time..lol

 

[I AM 0Z]

wow do some of the vendors even know what they are selling. tell you what you want to hear to make the sale then most likely will tell you that is not what you told me or i would have not sold you this.

Fellas;

I am a registered ISO 9000 internal quality auditor and if misrepresentation ever happens to you:
1. Inquire if the company you are dealing with is ISO 9000 registered (most large electronic companies are although most small bike shops are not)
2. Ask to see their Registration Certificate or at least email a photo copy of it (this will reveal the registrar who issued the certificate)
3. Email that registrar and report them

i guarantee that if they are registered it is a certainty that misrepresenting the production is in direct violation at least one of their documented quality procedures and inquiring this before hand when talking to a sales rep or salesman for the company will most times get the truth out of them from the very beginning.

Big Pain folks, i know, but if it happens enough it will make them behave.

let the buyer beware Gents

 

[kbarrett]

Real sensorless hall effect controllers are seriously cutting edge stuff ... lots of logic inside to try to deduce what is going on in the motor.

Not something you are going to buy for under $100, IMO.

 

[frank9755]

I got it to work at 24v on a sensored motor so it would appear as if it is not sensorless after all

John, Jeremy,

It definitely is the sensor version. I bought the same from Brett, for my Tongxin, and received the same!
I also thought there were too many wires, and queried it. I sent him the photo back and he said that I'd been sent the sensor version in error and he would get the sensorless one sent out. If and when it turns up, I'll post back with my results.

Frank

 

[frank9755]

Real sensorless hall effect controllers are seriously cutting edge stuff ... lots of logic inside to try to deduce what is going on in the motor.

Not something you are going to buy for under $100, IMO.

Tongxin sells its own controller for a lot less than that!

 

[wrobinson0413]

.

 

[johnrobholmes]

You just answered your own question.


Personally, it is a reliability issue. The startup from zero rpm is the only reason to run sensors IMO.

 

[ZapPat]

You just answered your own question.


Personally, it is a reliability issue. The startup from zero rpm is the only reason to run sensors IMO.

When you talk about the issue of the reliability of hall sensors, have you seen many failures of these? I could see motor heat being a source of concern for the hall sensors, specially with some people who love to push their motor's limits. And I guess humidity maybe, but water shouldn't be in the motor in the first place... Are there many other issues with hall sensors? Experiences anyone?

Almost all the app notes I've seen about the advantages of not using hall sensors for startup mostly focus on lower cost. For industrial giants looking to save a few cents here and there as to boost profits on volume sales this maybe be a concern [read obsession], but for real users, we just want something that does what we want it to, and that's having good reliable low speed startup and operation, halls or not.

I think Wayne is right in wondering about the importance of sensorless control for ebikes. From what I've seen most existing ebike controllers are far away from even making the most of existing hall sensor control techniques as it is, let alone any type of advanced sensorless operation... Exept maybe Justin's new controller!

 

[wrobinson0413]

.

 

[ZapPat]

Maybe another question to add would be how well the halls are aligned on hub motors? I have no experience with the hub motors because my bikes have DC motors. Has anyone spun one of these motors without a controller and looked at the backemf and hall signals to see how well they sync up? If they were badly aligned, I could see a justification for adding back EMF commutation to increase the motor efficiency.

It seems to me that what sensorless really means is using the zero crossing to do your drive phase timing, instead of using the hall sensor signals in the sensored setup. Then we can drive the motor any way we want with this timing info, no matter if it came from halls or zero crossing. It just happens that sensorless is often implemented with better design features and uses more advanced switching schemes, than controllers using hall sensors. What do you think?

Any hall sensor scew could easily be calibrated out in software, if needed. I don't have much experience as an ebike builder yet, so I don't know if these hall sensors are well mounted or not in general. Probably depends on what day of the week the motor was build, for one! I have one cheap hub motor here (golden motors), so I'll take a snapshot on the scope at a constant no load speed and post what I find here. I'll try to compare the halls with the BEMF voltage waveforms too if possible to see the phase offset. We also have to remember that zero crossing detection is also limited in resolution in it's own way, depending of the method of sampling used. AD inputs are nice and simple, but you only sample so often...

 

[wrobinson0413]

.

 

[dirty_d]

well really there is a limit to the highest rpm that you can go too, if your ADC can only do 15ksps, and the motor is a 12 pole at 6000rpm, thats 3600 zero crossings per second or 278 microseconds per zero crossing. each ADC sample takes 67 microseconds, so you would only get 4 samples per each zero crossing, so the actual detected zero crossing would be +/- 15 degrees of the actual zero crossing, pretty terrible. i think you would want at least 50 samples per zero crossing to get good enough resolution, or you could just use analog zero cross detection and have it output hall signals.

 

[wrobinson0413]

.

 

[Johnbear]

wow do some of the vendors even know what they are selling. tell you what you want to hear to make the sale then most likely will tell you that is not what you told me or i would have not sold you this.Ive been there several times thinking it was me and that i just did not explain it right. Like my 20 amp sensorless controller for a 5304 that was sold to me and then they said oh no it wont run a 5304..lol .On that one i know as i have the email lol.. They just don't listen to the customer i guess and just want to make the sale. some vendors i think go over there sales record and think i did not sell him this motor i sold him that one so he must be mistaken . I don't think they realize that there are more then one vendor out there that maybe one bought from, because maybe that vendor was out of stock at the time..lol

Great discussion on the sensorless brushless controller theory!

But, I have still not heard from Brett at solarbbq, I don't like being ripped off - even if it is a small amount of money. I cannot believe he did it to others as well. The reason I bought from him was because I thought he had a good reputation. Not in my book.

 

[docnjoj]

I wouldnt worry, Johnbear. He has always done right by me, and many others. Im sure he will attempt to make it right!
otherDoc

 

[Johnbear]

I got it to work at 24v on a sensored motor so it would appear as if it is not sensorless after all

John, Jeremy,

It definitely is the sensor version. I bought the same from Brett, for my Tongxin, and received the same!
I also thought there were too many wires, and queried it. I sent him the photo back and he said that I'd been sent the sensor version in error and he would get the sensorless one sent out. If and when it turns up, I'll post back with my results.

Frank[/quote

Have you heard from brett, solarbbq2003, yet about replacing your controller with the right one. I have not and I have sent him several e-mails that he has not replied to.

 

[solarbbq2003]

I'd better say something here, since I sold the controller, there was a mix and a sensored controller was sent by mistake, I didn't realise there was a mistake until I saw a picture of what was sent, my guy in china unfortunately didn't pick up that it was the wrong controller before sending out. The controller that should have been sent only has the three thick phase wires, and a few other wires for power, brake cut off etc. I have assured john I will be sending correct one out, at the moment I cant send anything electrical in nature via ems freight until the olympics finish, but the correct controller seems very good, it will start the motor from any position, was a simple mistake on this occasion which shall be rectified shortly.

 

[solarbbq2003]

I'll give you more details on what happened.
I found a supplier of sensorless controllers to replace a controllers that people were getting from tongxin for their sensorless controllers, I tried the controller on tongxin sensorless motor and it worked well. The supplier I get from is not the factory that makes, a middle man, that person ordered some more controllers for me and my workman in china picked them up.
I assumed they would be exact same as the first ones I tested, my fault for not checking they were correct ones, and a few got sent out, it wasn't until I recieved some pics from the people that were sent them that I realised it was wrong controller.
So the controllers that I had got from supplier have been taken back to him for replacing with correct model.
Very sorry for the mixup, certainly wasn't trying to rip anyone off, was an unfortunate mistake of being supplied wrong controllers, which wasn't picked up by my guy in china. My hands are tied for 2 weeks cant send anything out of china due to crack down on electrical items going out by plane during olympics. But will rectify as soon as olympics are over. I'm way behind with emails at the moment, a backlog of about 100emails so am about one week behind answering emails, hence no replies as yet, but trying to get through the emails at present. sorry guys was simple mistake wasn't picked up before they were sent out

 

[Johnbear]

I recieved the controller, I can't wait to try it. Will post results when I get a chance. I am not at home so I won't be able to test it till later.