Sensorless speed controllers

[Andy Day]

Hi Folks.

I've got a Nano powered Falcon, which after 4 good months use has lunched it's speed controller. Simon at Systems at Work assures me Tony and he are dealing with the problem. I'm looking at alternatives however. Has anyone found an alternative sensorless speed controller which works?

In a different interest, model aeroplanes, I know most electric powered models use a brushless dc motor, and some type of pwm speed controller. These can often handle up to 80 or 100 amps, the most powerful motors rated at well over a kilowatt, ie over 1 1/3 horsepower! These motors develop their power at high rpm, which is of no use to us without some gearing - well a lot of gearing! However it occurs to me they would operate at reduced rpm delivering less power if restricted in some way, and give a performance more suited to a bike.

Has anyone given this technology any attention, or am I as I suspect barking up a gumtree?

 

[john]

The controller comes from TongXin in China. You could try to get one direct from them.

CHINA HANGZHOU ZHEDA TONGXIN ELECTRIC CO,. LTD.

 

[Ian]

Has anyone given this technology any attention, or am I as I suspect barking up a gumtree?

The Austrian member kraeuterbutter has oftem made comparisons to RC aeromodelling, he's obviously an enthusiast himself. However as far as I know he hasn't yet tried any RC technology on a bike.

It certainly sounds attractive, the motors used in models are small, light and very powerful. As I see it the main obstacle would be the gearing, getting a high enough ratio while keeping it efficient and quite would be a challenge, but if it could be overcome...

 

[frank9755]

I'm not aware of any ready made alternatives, although someone who was an electronics whizz could probably adapt an RC one to take 36v.

However, Kraeuterbutter actually has Tongxin controllers - drop him a line and see if he has the type you need. If not he will probably offer to get them in. Or else do try Tongxin direct. They were happy to sell to UK consumers but that may have changed recently.

The problem with these controllers is that they don't last long so you might want to get a few!

 

[Andy Day]

I've been in touch by email with Diana Lin at Tongxin over a week ago, and had no response, which is why I was looking for an alternative. Is there some special way you need to talk to them before they respond?

I don't think you could use an rc controller with the Tongxin as although brushless they do operate at significantly higher speeds, up to well over 10,000 rpm, to change this would I guess at least the ability to reprogram the controller, not a facility available on most BEC programming cards.

 

[flecc]

I've been in touch by email with Diana Lin at Tongxin over a week ago, and had no response, which is why I was looking for an alternative. Is there some special way you need to talk to them before they respond?

It seems that's what everyone else has found, she was communicating well while sales were in the offing but went silent when controller spares and the like cropped up. Tells it's own story.
.

 

[frank9755]

I think she's pretty busy so emails don't always get answered (especially 'hard' ones). When she went quiet on me once I just called her mobile and got an answer.

Frank

 

[Andy Day]

I don't think my email to Tongxin would have been problematic at all!

I merely requested a price for a motor and two controllers, plus shipping costs. No criticism of reliability. I think they have other problems, a large shipment to Systems at Work is "delayed" apparently. Certainly no-one is admitting there may be a reliability problem with anything other than a small batch of controllers which had high infant mortality, and I don't think my failure was one of that batch. It had been pasted almost every day of the week for 4 months, so it's my feeling there may be a bigger issue with poor unreliability.

I'm sure the guys at Systems at Work are honest men, I've had excellent service in the past, but I would think any real solution is largely out of their hands, and down to Tongxin.

Inscrutability rears it's ugly head!

 

[Miles]

You can certainly use RC controllers for bikes. Castle Creations make good high voltage ones: Castle Creations | Phoenix HV High Voltage Brushless Motor Controls

You'll need a servo tester to control the throttle function.

The outrunner type motors which predominate in the RC world aren't so ideal for our usage... Certainly, you need to downrate a lot for the reduced/radial airflow....

Miles

 

[Ian]

You'll need a servo tester to control the throttle function.

Miles

Or an R/C transmitter

 

[Miles]

Or an R/C transmitter

Your life...

 

[john]

It had been pasted almost every day of the week for 4 months

Have you been on the same motor for 4 months too?

If so, how many miles have you had out of it so far?

 

[Andy Day]

John, I must have done about 600 miles in the 4 months I've had it, and the motor itself has been fine, apart from the washers which are meant to prevent the wheel spindle rotating losing the internal shape that locks them to the spindle. I made a new pair out of 1/8 steel sheet, EN16 if you are interested, and after hardening they have given no trouble. I think the originals are the traditional far eastern monkey metal!

Ian, I've got the RC tx and rx, but I find it difficult steering with my feet while operating the tx throttle Seriously, a solution is a simple clock driving a one shot with a pot controlling the one shot pulse width - thats all a simple servo tester is! My concern about using a RC ESC to drive the Tongxin Nano was whether the ESC could cope with the motors lower speed of rotation. It's my understanding that the firmware inside the ESC must have some limits to it's operational parameters, but maybe I'm wrong?

 

[Ian]

J
Ian, I've got the RC tx and rx, but I find it difficult steering with my feet while operating the tx throttle Seriously, a solution is a simple clock driving a one shot with a pot controlling the one shot pulse width - thats all a simple servo tester is! My concern about using a RC ESC to drive the Tongxin Nano was whether the ESC could cope with the motors lower speed of rotation. It's my understanding that the firmware inside the ESC must have some limits to it's operational parameters, but maybe I'm wrong?

Andy, the beauty of using the full RC kit is that you can put all of your effort into operating the throttle while someone else rides the bike.

Your quite probably correct about RC controller firmware not being suited to a low speed hub motor. I can't say for sure though as the last time I dabbled with RC, brushless motors had only just started to appear.. in new fangled VCR's!

 

[Miles]

My concern about using a RC ESC to drive the Tongxin Nano was whether the ESC could cope with the motors lower speed of rotation. It's my understanding that the firmware inside the ESC must have some limits to it's operational parameters, but maybe I'm wrong?

What's the gear reduction ratio on the Nano? I thought it was over 10:1, no?

Obviously, you're not going to get much help from a standing start with any sensorless controller.

 

[john]

www.ebikes.ca Homepage of the revolution have some sensorless controllers in stock.

 

[frank9755]

www.ebikes.ca Homepage of the revolution have some sensorless controllers in stock.

I just checked the site in excitement but didn't see them.
Do you mean the Crystalite pedal first ones? They're no good for a Tongxin as the freewheel means it won't generate the necessary back emf.
If its something else could you post an exact link?

Frank

 

[Jeremy]

Miles wrote:

Obviously, you're not going to get much help from a standing start with any sensorless controller.

Curiously, the standard Tongxin sensorless controller does work fine from a standstill. It seems to start up well, with lots of initial torque, even with the wheel held off the ground.

I believe that it uses the same, or a similar, starting strategy to the RC brushless motor controllers, which is to send some short initial pulses to the motor phase wires and then immediately read the emf from the adjacent windings. From this, it can deduce in a fraction of a second what the required phasing is to make the motor spin in the required direction.

Jeremy