New Super Light Tongxin Motor Design/Kit

[Pedalo]

This thread is really taking me back about 30 years or so to a misspent youth

I used to spend hours tinkering with 555 timers to implement police sirens and various other noise making circuits. I'm amazed the 555 is still available after all these years!

 

[benjy_a]

This thread is really taking me back about 30 years or so to a misspent youth

I used to spend hours tinkering with 555 timers to implement police sirens and various other noise making circuits. I'm amazed the 555 is still available after all these years!

An amazingly versatile little chip...I remember before the days that you could easily buy cheap inverters (to give you mains 240V from a car battery) I built one using a 555 timer and some (very) chunky transistors to pulse 12V backwards through a transformer at 100 Hz (square wave emulating 50Hz sine wave).

Unfortunately I calculated the frequemcy incorrectly (10 times too high) and produced 2400v from a car battery, exploding my multimeter in the process!! Once the correct frequency was set I got it working to run an old 240v TV from a car battery.

Children should not be let loose unsupervised....I'm lucky to still be here!

 

[dingrpdl]

Just a thought, Jerry...

If you open up the controller, I suspect you might find an unused contact for a throttle signal input on the PCB. If this is the case, then all that would be needed to convert the pedelec sensor connection to a throttle input connection is to move one wire from the pedelec signal to the throttle signal solder pad. The ground and 5V signal can remain connected to the original pads on the PCB.

If you do choose to investigate this option (and I understand if you choose not to as so many options have been presented to you), take a good picture of the front and back of the circuit board. And note the model of the controller as well as possibly a website screened onto the board.

A long shot. But much neater if you can move just one wire.

 

[jerrysimon]

Thanks guys given me lots to think about. I may open up the controller to have a look but its pretty well sealed. Would also void any warranty, though not sure if they would honor that anyway now I have cut off some of the connectors/wire. Also as I understand this is a new designed controller and as they don't provide any form with throttle option, I doubt the circuit will cater for one.

For now I am going to try it as is.

Another pain is I sent my pictures to Tongxin to show them how I fitted the disc and they have replied to tell me it should be on the non crank side, sigh....

They said the way I have fitted it to the crank side it would not work and would only turn the motor if I pedal backwards. This is of course not true, as I have it working fine fitted like this but it does also turn the motor if I pedal back hard.

Instructions when they sent the motor would have been helpful

At least others now know the correct way to fit the system. I also think lengthening the cable for the sensor has an effect on its performance/sensitivity. With the extra length lead it seems to accelerate up to full speed the moment it detects pedal cadence whereas when I had it stock, it defiantly seemed to be a step change as you increased cadence. I am going to put it back to its original length and retest.

Regards

Jerry

 

[benjy_a]

Another pain is I sent my pictures to Tongxin to show them how I fitted the disc and they have replied to tell me it should be on the non crank side, sigh....

They said the way I have fitted it to the crank side it would not work and would only turn the motor if I pedal backwards. This is of course not true, as I have it working fine fitted like this but it does also turn the motor if I pedal back hard.


Jerry

If this is true then something is definitely screwey. If on the wrong side and working as the manufacturer intends then it should work perfectly, but backwards.

I believe that this will be due to the magnets being spaced irregularly on the disc and the controller should be able to distinguish from the pattern of pulses which direction the disc is turning and operate accordingly.

I find it very weird that a change in length of wire also alters the behaviour. Any wiring should be of such low internal resistance as to have no effect.

All very strange...would be easier if it was behaving correctly but backwards! Have you considered the possibility of a faulty controller?

Ben

 

[laroche]

Jerry

Interesting that the cables with the built-in controller version are each cut to about the right length, even for a big (57cm) frame. But even if they're not quite right I shall not be altering them in any way!

Why, why, why are there no instructions? The kit would not require much explanation; a couple of diagrams and a photo or two and would be enough. And instructions would give them less hassle, never mind the customers!

Happy days are here! The students are now beyond help. One paper done and, after the hols, it's almost wall to wall exams. Anybody got a bike for sale?

One technical question. How did you keep the disk a safe distance away from the bottom bracket, to prevent the magnet side being fouled?

Cheers!

 

[neptune]

There is some interesting stuff going on in this thread. @Benjy_a, it would be interesting to see what the output of a pedal sensor looks like on the oscilloscope. I would not expect the pulse ratio to be 50/50, due to the magnet spacing as you say. I suspect that this ratio is not critical, and all that matters is the pulse rate or frequency. If we could come up with a simple pedalec simulator, it will have widespread use as a diagnostic tool, and a conversion kit to allow throttle operation of non throttle bikes.

 

[benjy_a]

Hi Neptune,
When I get the time I will have a play around with this for sure. However I do really expect the output of the pedelec to be irregularly spaced pulses which will resemble say Morse code....dot dot dash dot etc.

This would be more complicated to emulate. One way to try without building any electronics would be to modify a disk to only have evenly spaced magnets and see if this activates the controller. I think I have a spare pedelec disc and sensor, I could try this quite easily.

I had the same thought that you could potentially have an after market throttle kit for pedelec only bikes.

Ben

 

[jerrysimon]

I thought the magnets were evenly spaced ?

Well having done another test with the stock wire lengths there is no difference to the performance so all in my imagination. I have tried the motor again on a standard Tongxin control and whilst it works with thumb throttle, it is pretty rough through the thumb throttle movement and does not seem to achieve full speed at full throttle.

So...

What I now plan to do is fit it all stock with original controller and pedelec sensor. I have put in a switch to the brake cutoff cable (so I can switch on assistance when I need it only) and one brake to cut power as well. I am then going to run it every day to see how long this motor/controller lasts. My only slight concern now is that the motor clunks a little if at full speed you stop pedaling. Anyway lets see how long it lasts.

If one of you develop a thumb throttle in the meantime, then I will be happy to try it on my setup.

This has taken up way too much of my time, so now I am just going to fit it as is and use it on my daily commute

Regards

Jerry

 

[benjy_a]

I would suspect that not all of those are magnets...otherwise there would be no reason at all for it to not work exactly the same in both directions.
Ben

 

[jerrysimon]

Why, why, why are there no instructions? The kit would not require much explanation; a couple of diagrams and a photo or two and would be enough. And instructions would give them less hassle, never mind the customers!

One technical question. How did you keep the disk a safe distance away from the bottom bracket, to prevent the magnet side being fouled?

Well we knew it was in early stage of release and they are never good at including instructions.

The disk is a tight fit and I guess as it fits over the raised boss on the crank its pretty secure. I could add a couple of dabs of epoxy where fits flush against the crank I guess.

Regards

Jerry

 

[neptune]

@Jerry. When you get time, will you please test your spare magnet ring to see if all the apparent magnets are in fact magnets. You could test with a sewing needle.
We need to recognise that different controllers may work in different ways . It is likely that some are designed to prevent motor switch-on when back peddling, while others are not.
The other thing is we are hijacking Jerry`s thread , and perhaps we need a new thread dedicated to matters 555.

 

[benjy_a]

Yes, apologies for hijacking the thread its all gone a bit geeky!

Having said that if I get anything working Jerry will be more than welcome to have it at cost price.

 

[jerrysimon]

No problems guys its all part of the same thread as far as I am concerned and very interesting reading. The lack of throttle (amongst a few other things like the pedelec sensor) is a limitation of this kit. Maybe also that the motor seems controller specific. It will be interesting to see laroche's setup given with his the motor has the controller in the motor itself and he has the power gauge/limiter to use on it as well.

I will whip the crank off and test the magnets to confirm which are and which are not. I had assumed they were all magnets

Thanks for the offer Ben. By all means start a new thread when you start the design. I will be following it with interest. I may be able to get the guys at work to put an oscilloscope on my setup later.

Regards

Jerry

 

[jerrysimon]

I have been out on a few rides and everything is working ok. The only thing I noticed was that the motor does make a "clunk" sound if you stop pedaling when running at full speed. I have now also received a reply from Tongxin about which side to fit the disk.

"We haven't tested to put the magnetism disc on right side pedal, our technical worker suggest you to put it in correct way. Cause there's a clutch inside, When you put the speed sensor on the correct way, when you start pedaling, the clutch close, then the motor runs, but when you stop, the clutch open."

As we said a set of instructions would have been good.

This "clunk" noise could be the clutch. If you slow cadence down rather than just stop pedaling then the shut down of the motor is then silent when you stop pedaling.

I am going to try to fit the disc and sensor to the other side of the BB.

Sigh.....

Regards

Jerry

 

[shemozzle999]

Hi Jerry, after looking at your pictures, what about making a short box section carrier for the pedelec pick-up sensor together with a locking screw and glue that to the frame, rather than glueing the pedelec pick-up sensor direct to the frame, that way you will be able to adjust the gap with some shims, remove it for cleaning or future replacement.
It is only a suggestion - the build looks good.

 

[jerrysimon]

Thanks Shemozzle999,

That's worth thinking about. The real issue with mounting it under the BB is it already fouls the fold at its current size so making a container for it would make that worse.

The good news is that I managed to mover the magnetic disk across to the other side (modified the disc more and the left hand crank a little) and have some good progress to report later.

Tongxin have also been very fair and agreed to send me a spare sensor and magnetic disc given the butchering I have done to the current ones

@Jerry. When you get time, will you please test your spare magnet ring to see if all the apparent magnets are in fact magnets. You could test with a sewing needle.

Tested and everyone of the metal inserts (ten) is magnetic!

I also spoke to the engineers at work and they said it would be easy/ok for me to connect it up to an oscilloscope at work to see the wave form/pattern

Regards

Jerry

 

[daniel.weck]

I thought the magnets were evenly spaced ?

Twice as many magnets as the typical Bafang/Tongxin type (the old ones have useful arrows to indicate the correct rotation)

 

[daniel.weck]

My solution using five minute epoxy. Its very secure and I don't think it will be moving anytime soon

Somehow a lot cleaner than my (admittedly temporary) solution

 

[daniel.weck]

From this side it nearly looks like a hub dynamo!
Great-looking white frame by the way