I have converted a brushed motor, electric scooter to a brushless motor with suitable controller.

Every thing worked fine for months until I overloaded the motor one day. The motor suddenly lost its torque and proceeded to run very rough; little power and very jerky.

 

I assumed, at first that I had burnt out one set of thyristors and it was only running on two phases but investigation has shown that each phase conductor is passing current when the motor runs. I've tried measuring voltages at between the 5 Hall wires to the controller, these all vary to some degree but one seems to have no or much lower volts between it and the other 4.

 

Can anyone suggest the most likely problem? Whether the controller has a fault or, what I beginning to think now, that there might be a problem with the Hall sensors?

 

Can overloading a motor cause that type of fault?

 

Thanks.

It sounds like a failed Hall sensor in the motor which can cause the rough, noisy, low powered running, though a poor connection in the Hall sensor wires can also be responsible.

 

The Hall sensors can often be dug out of the stator and replaced, fiddly but possible with care.

 

I doubt if overloading would cause a sensor to fail, but I've known it to cause a poor hall sensor wire connection, it happened to me when I was doing a cruel stall test on a very steep hill.

.

>>The Hall sensors can often be dug out of the stator and replaced, fiddly but possible with care.<<

 

Ok, Thanks. It looks like that's where I've got to start delving, intto the motor; I'll post back what I find.

Faulty motor or controler

 

I know this discussion was a long time ago but I have only just got around to stripping the motor down and investigating.

 

There is no apparent physical damage to the sensors, ( I have uploaded a close up picture of them at this location: http://clivebillson.tripod.com/nettop/HallEfectDetector01.jpg ) That is: clivebillson.tripod.com/nettop/HallEfectDetector01.jpg

 

I wonder if anyone knows what resistance measurements should be expected, between each of the three legs of each sensor? I have taken these readings but get a variety of differing results. Is there any other way to test them I wonder?

Edited February 10, 201214 yr by xyzzy

Its all on youtube . Search "testing Hall Sensors " by Electric Chronicals . He shows how to test them without stripping the motor .

The hall sensors give a pulse of 5v in certain position during the rotation of the motor, so you need to rotate the motor with the meter on to see which ones flash or not. There's a 5v, 0v and signal wire. Test the signal wire voltage with your other probe to ground. You can get replacement hall sensors quite cheaply, or you can get a sensorless controller and never have the problem again. Honeywell SS41 or equivalent is the designation of the hall sensors.

This post by Andrew Harvey may also be helpful:

 

Hall sensor testing

 

There's also a sequence of posts and links from this one on, where a member changed one:

Fitting new Hall sensor

Ok, I know this is slow progress, I’ve determined that the “green” hall sensor is faulty and I’ve bought a replacement. Before I replace the old one I want to be sure I’ve got the right device. My understanding is that there are two types of sensors a digital position type and a magnetic field type. Am I correct can anyone advise? If so how do I determine which type I need?

 

d8veh recommends a Honeywell SS41. I think this is the digital type. Is it that this is the usual type used for electric bike controllers?

The one I have for replacement is an XF2- 012.

Thanks.

Did the old one you removed have any identification markings or the two remaining ones?

 

Who did you purchase the new one from and was there a data sheet of the device?

Edited March 27, 201214 yr by shemozzle999

Would it be easier to simply replace the controller for one that works without sensors? it would be more reliable in the long run.....

Did the old one you removed have any identification markings or the two remaining ones?

 

Who did you purchase the new one from and was there a data sheet of the device?

 

I haven't removed the old one yet. I've, earlier, uploaded a picture of the sensors in the hub, shown via a link referred to, four posts back. You will see that the sensor will need digging out and there may not be much discernible left after doing so. You can, however, read part of the identification on one, while in situ; it reads 327 and above that something like 541.

 

The new one I’ve bought is from Spiratronics and the product code is XF2-012, with the following description:

 

This hall effect sensor is an ultra-sensitive transducer which varies its voltage based on changes in magnetic fields around it. It is pole independent and is ideal for hand held equipment and speed measurement.

 

Supply Voltage: 4.5V to 6V

Output Voltage: 4.7V

Max Supply Current: 11mA

Output Sink Current: 10mA

Output Bandwidth: 20kHz

Magnetic Sensitivity: 2.5mV/G

 

Does that help?

Would it be easier to simply replace the controller for one that works without sensors? it would be more reliable in the long run.....

 

I had thought about that but I have read, elsewhere on this forum, that they don’t always run that well on a motor that is originally designed to run from a controller reliant on sensors, particularly at slow speeds. Also I am not sure how the speed control (twist grip) would interface with a new controller.

There is always the economics to consider as well; approx. £50 for a controller versus £3 for a new sensor.

 

I haven’t ruled out the possibility however.

Yes it does, your old one is an S41, the 327 is either a date or batch code.

I Googled XF2-012 earlier and came up with Spiratonics but that is there own product code but doesn't tell me the actual device code so I can not check the pin outs of the device are the same as the old one.

If you are not in a hurry you could contact them and ask for a data sheet or the manufacturers part number or take a chance and fit it.

 

If you do, I suggest you use wire cutters to cut the leads of the device about half way up the lead then de-solder each lead one at a time by heating joint on the top of pcb and drawing down the lead out of the hole in the pcb with a pair of pliers being careful not to drop the lead inside the motor when you remove it. It then leaves you free to dig out the old device.

When fitting the new one insert the leads into the holes underneath the pcb and the insert/glue the device into the motor slot. Wait for the glue to set and finally solder the leads and any wires back onto the pcb.

 

Your picture link only works for me if I cut and paste the address directly into my browser removing the space after http://:

 

http:// clivebillson.tripod.com/nettop/HallEfectDetector01.jpg

Edited March 27, 201214 yr by shemozzle999

Thanks for the helpful tips on how to change it; I’ll follow your advice.

I'll ask the Spiratronics, supplier for a data sheet but for the sake of a few pounds I may as well just buy another and get the correct product. I’ve searched in the relevant places, like e-bay & RS, but it doesn’t come up with an S41 specifically. However there are so many in the RS stock I wonder if there is an equivalent.

Hi On ebay

 

eBay - The UK's Online Marketplace

 

I would always change all 3

 

Frank

Nearest RS equivalent is ss413A Part No.181-1479

I have used it myself and it works great.

 

They are made by Honeywell link to their datasheets - type in ss41 in the search box, then view product,

then spec tab. Do the same for ss413a the important data is the output voltage and the operating point strength in Tesla.

 

Honeywell Sensing and Control Product Search

Nearest RS equivalent is ss413A Part No.181-1479

I have used it myself and it works great.

 

They are made by Honeywell link to their datasheets - type in ss41 in the search box, then view product,

then spec tab. Do the same for ss413a the important data is the output voltage and the operating point strength in Tesla.

 

Honeywell Sensing and Control Product Search

 

Ok, thanks. I've ordered three. As another member advised, it may well pay to change all three.

 

Thank you for all your help.

NP let us know how you get on.

Hi you have to chip them out using a small screwdriver as a chisel then get in with a Stanly knife to clear out slots i then use super glue This guy on ebay is CHEEP

 

50g SUPER-GLUE ADHESIVE. HIGH STRENGTH. HIGH SPEED. | eBay

 

Frank

I had thought about that but I have read, elsewhere on this forum, that they don’t always run that well on a motor that is originally designed to run from a controller reliant on sensors, particularly at slow speeds. Also I am not sure how the speed control (twist grip) would interface with a new controller.

There is always the economics to consider as well; approx. £50 for a controller versus £3 for a new sensor.

 

I haven’t ruled out the possibility however.

 

Controllers are about £18 from E-crazyman on Ebay, not £50, they run well sensor-less but you may need to modify the connectors to fit your existing ones

NP let us know how you get on.

 

Will do.

They arrived this morning (brilliant service RS). It's not immediately obvious which way round they go. They are bevelled one side and a round spot on the other; any clues?

Hi they can only go in one way when you remove the old ones you will see the slot is beveled

 

Frank

As Frank states, the rounded face of the device with the component id marking is on the outide (exposed to the motor magnets) of the motor.

The face with the spot to the inside (hidden from view when installed).

Edited March 28, 201214 yr by shemozzle999

Ok, thanks to both of you. I'll keep you informed.

Just did this with help from the Forum, success! Recent thread refers. If I can replace one anyone can. Symptoms sound identical to mine.

 

Good luck- AG