Sabvoton 72150 stuck on hall sensor test.

[thender]

I have one 72150 that works great after finding the proper combination of hall sensor wires to phase wires.

I have another I got - same look, except this one has the multipin green connector on it that goes to an LCD/button control thing. It's pretty useless as all it can display is amperage used/turn it on, but it seems the same.

It works except in the beginning it makes a bunch of strange noises rather than move continuously, until it is in motion at high speed. Very rocky start. I ran the hall sensor test and it failed.

I tried getting an entirely different motor core/hall sensor and it will run the hall sensor test on that one, but it stays in progress indefinitely.

Has anyone here had a 72150 controller that stays in progress constantly or refuses to pass? What are some troubleshooting techniques for this controller when it does such things? Or is it just known to be buggy? Perhaps the 2nd one I got is a dud?

Thanks!

 

[amberwolf]

Does the problematic controller work the same problematic way on the motor that the working controller works with?

Does the working controller work correctly on the motor that the problematic controller doesn't?


Hall problems are usually either broken wires or poor connections, sometimes they are bad halls. Rarely they are caused by an internal controller problem like a bad or missing pullup, or insufficient voltage to the halls, etc.

Your best bet is first to verify the halls are actually operating correctly, and that if they don't seem to be, that they also don't work correctly with the controller you know does work.

 

[thender]

whoops double post, my apologies

 

[thender]

I had to undo some of the waterproofing I had added to check the new motor with the old controller, my apologies for the late reply.

The working controller works correctly on the motor that the problematic controller does not work on. When I try the any of my BBSHD motor cores with the OLD WORKING Sabvoton 72150 that is on my working bike, it passes with 2:test OK

With the new motor core/controller, it fails hall sensor test with 4: test fail4(apparently there are two different statuses for fail, fail3 and fail4).

Also, just to remove another variable, I bought a new BBSHD motor core with fresh hall sensor and a new Sabvoton 72150 - it failed the same way.

I bought a new motor core with a fresh hall sensor and a new fresh Sabvoton 721150 to try and figure it out - I confirmed the wiring that worked on the old bike with BBSHD/72150, which is as follows

T = pin 6 = WHITE
H2 = pin 5 = GREEN
H1 = pin 4 = BLUE
H3 = pin 3 = YELLOW
- power = pin 2 = BLACK
+ power = pin 1 = RED

and then for power to the motor

GREEN PHASE = GREEN PHASE
BLUE PHASE = BLUE PHASE
YELLOW PHASE = YELLOW PHASE

This configuration does not allow my motor core, to work with either of the two new 72150 I have here. It only works on that first original.

There are several possibilities here.

1) Every Sabvoton 72150 has different hall sensor wiring even if they have the same colors. I doubt it.
2) I coincidentally got two bad Sabvoton 72150 from two different vendors. I doubt it.
3) Both hall sensor wiring jobs I did are equally terrible. I am open to it, but I doubt it.

I'm close to stumped.

 

[thender]

I re-ran the test, rather than using 15-25 amps for test given current for hall sensor test, I went all out and tried 45 amps. now it passed. I changed it after realizing that at 15-25 amps it wasn't even turning the stator, just a lot of coil whine. With the other controller, even at 15 amps hall sensor test turned the stator and it passed.

With no load, not actually plugged into the motor housing, turning the stator should NOT take more than an amp if even... so I am concerned as to why this is. I will look into it more tomorrow.

Thank you so much for taking the time to respond to my vague inquiry, I genuinely appreciate it. Have a good night!

 

[amberwolf]

At this point, my best guess is that the Sabvotons are very vulnerable to noise on the hall signal lines, or something like that; if so, separating the hall wires from the phase wires for as much of the length from controller to motor as you possibly can may help, especially if you put a shield (aluminum foil, braided sheath, etc) around the halls that's grounded at the controller end only.

Sometimes adding small capacitors (.1uf or less, ceramic not electrolytic) to the hall power lines, at the controller and motor end, may help with noise.

If you have an oscilloscope you can also examine the signal and power lines for noise, comparing the working and nonworking setups.

It's also possible that Sabvotons have dropped in quality, and manufacturing is randomly bad. Or that there are "genuine" ones, that have passed QC, and "rejects" that didnt' pass QC and were sold to third party sellers (since the Chinese don't ever actually seem to destroy anything, even if it's totally defective, even hazardous, they just sell it to someone else to recycle...who then sells it to someone else as the genuine class-act article). :/

 

[leffex]

I re-ran the test, rather than using 15-25 amps for test given current for hall sensor test, I went all out and tried 45 amps. now it passed. I changed it after realizing that at 15-25 amps it wasn't even turning the stator, just a lot of coil whine. With the other controller, even at 15 amps hall sensor test turned the stator and it passed.

With no load, not actually plugged into the motor housing, turning the stator should NOT take more than an amp if even... so I am concerned as to why this is. I will look into it more tomorrow.

Thank you so much for taking the time to respond to my vague inquiry, I genuinely appreciate it. Have a good night!

Nice that you solved the issue uping the amps (A) in the sabvoton hall test!



epic controller and you've got two. Used mine for a year now at high level and its alive without hicup.

 

[thender]

Should it be troubling that it takes 45 amps to run a hall sensor test? 45 amps is what I am usually using to go 40 MPH with the load of the motor being in the housing/carrying my bike and an actual human - and it is using that for a hall sensor test with the motor/stator outside of the housing with no load

The motor worked with the controller before, the issue I was having was that at low speeds it was stuttering/choking and turning itself on and off repeatedly rather than working smoothly. If the hall is only going to work when 45 amps is going through the motor core, that is a problem.

I bought the second controller that had issues from here https://www.aliexpress.com/item/4000065034692.html?spm=a2g0s.9042311.0.0.27424c4dfdYUtx and the third one with issues from here https://www.amazon.com/gp/product/B07D7TLQVN/ref=ppx_yo_dt_b_asin_title_o07_s00?ie=UTF8&psc=1 The first controller that worked perfectly I purchased from laebike which looks as geocities as it gets but it actually worked great, they do not sell it anymore it seems. https://laebike.com/blogs/information/76664133-sabvoton-sinewave-controller

Here are the settings I have chosen for this. It's a 149SP battery pack that can do 42-45 amps continuous, 55-60 amps peak. For those who may ask, I got this controller because the stock Bafang controller is junk(square wave and kills stators even at low gear going down a hill at 18 MPH), I could never get a Kelly to work right. The ASI BAC800 is great but the company won't give you access to codes/software/credentials to program it even if you throw money at them so I dumped it. This seemed like a great controller - smooth feeling, powerful, efficient.. if only I could get two or three of the same model to behave even remotely similarly to one another we'd be in business. `

 

[amberwolf]

Should it be troubling that it takes 45 amps to run a hall sensor test?

Are you measuring that with a wattmeter or ammeter, or is it a readout from the sabvoton software?

If the latter, I'd suspect a problem in the software; you should check it with the former, because any motor drawing that much current with no load ought to be getting real hot, along with the phase wires and controller, with any kind of actual load.

The ASI BAC800 is great but the company won't give you access to codes/software/credentials to program it even if you throw money at them so I dumped it.

Was it bought from ASI, or from ERT? If from ASI, they should have provided all that stuff with it. If ERT, well, ERT appears to have "issues" providing the support they should.

 

[thender]

Sabvoton software - I will check it again tomorrow with the battery BMS app and see for sure. I gave it about 20 miles of work today with a passenger riding on the rear seat and it worked great, so all seems to be set. This combo kicks ass. Much smoother than the ASI BAC800... no shakiness starting up, quicker response when I hit the throttle. I have to get a switch to put on the XT90S going to the controller from the battery, if I turn the battery BMS on with it plugged into the sabvoton, even if the sabvoton is off it will trigger the BMS to turn straight off. So, I unplug the Sabvoton XT90S, turn the battery on, plug it in, and then flip the ignition switch and it works.

The way it was explained to me was ERT can get in trouble for sharing access codes with customers by ASI. This is a poor situation, and likely why they shied away from doing so. I was wondering why it took so much prying to get access to the thing. I asked ASI if I could pay them for an access code and got crickets. At the time, they had no ASI BAC800 eval kits in stock, and they had no means to purchase a support plan separately. Someone else wound up sharing a code with me that worked for a few days and then died after hearing my experience, and I was done with it after that. Far too much hassle.

If you're locking controllers that people are clearly buying for custom use/DIY market, I don't want it on my bike.