Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

BLDC motor + "custom" controller = smoke 1

Status
Not open for further replies.
stabmaster

stabmaster

Chemical
May 30, 2005
50
0
0
US
I have a BLDC motor built by ebmpabst. It is magnetically driving a fluid-o-tech gear pump (they came together).

I am trying to run this off of solar panels and I had a custom board built by solarcontrols.com in order to drive this motor.

The ebmpabst motor is driven normally by "variodrive" driver; however, I had a spec sent to me which describes the commutation for those who wish to use their own products.

Matching the pins, it appears that the motor doesn't want to work. It appears that the motor locks into a position when the drive is powered.

solarcontrols told me that this is because the sensors and switches are out of phase, which makes plenty of sense; however, I checked my wiring numerous times. I then tried every permutation (6 permutations)) for the hall sensor configuration but to no avail.

Finally I placed everything back to where they should have been originally. Upon doing so, I spun the motor against its will and saw/smelled burning circuit board. I don't think I let this persist long enough for the motor to have given up the ghost, but I'm not sure.

Is there something I can use to troubleshoot a BLDC motor?? I assume the answer would be: use a different drive.. and to troubleshoot the drive, use a different motor.

Is there any other way to determine what is not working properly? I have studied the spec's carefully and I'm quite confident that I did my part correctly.

On another note, I had the harness soldered for me, and it is possible that there may be a short; however, I tested this with a multimeter numerous times but never found such a short.

Any thoughts on where to start on this?
 
Replies continue below

Recommended for you

Sort by date Sort by votes
Fluid-o-tech MG213

Two spec sheets on this pump/motor:



Here is the manual for the controller:

MANUAL
Model: Ps –Mini
Solar Converters Inc. - Rev. B


Note: This unit is sample unit designed to run a 28 V DC brushless Pump from a 12 V nominal Solar Panel for evaluation purposes

Input connections:

Red: wire: Solar Panel +
Black wire: Solar Panel –

Motor connections

Vc: Motor Phase C
VB: Motor Phase B
Va: Motor Phase A
+5: Hall sensor power supply
Hc: Hall effect sensor phase C
Hb: Hall effect sensor phase B
Ha: Hall effect sensor phase A
Gnd: Hall Effect gound wire

1.0 Specification

Input Voltage: 0 - 25 DC volts PV Array, approx. 15.5V à
Current: 0 - 4 DC amps nominal,

Output Voltage: 28 V PWM suitable for a DCBL Motor @ 2 amps

Nominal Maximum power point tracking to optimize output power.

No User load overvoltage, regulates output voltage to "float” voltage even with no battery connected.

Efficiency: >94% over 20% charging load

Temperature compensated at –4 mv /cell/ deg C

Transient protected - input and output

Temperature range: -40C to +60C

Internal 6 amp Solar panel Fuse


The variodrive commutation diagram was emailed to me. It is in PDF format. Not sure how to share that.
 
Upvote 0 Downvote
It's very doubtful that any harm was done to the motor. Turning the motor shaft by hand should not have caused the motor drive to blow up. The most likely reason for the smoke is that an upper and a lower phase transistor got turned on at the same time.

Possibly, the Hall convention is different between the drive and the motor. The are two conventions, 60 degrees and 120 degrees. This is the separation between the rising edges of the hall sensors. The decoding sequence is different and could cause upper and lower transistors to turn on at the same time.

 
Upvote 0 Downvote
From the motor commutation specs:


It appears (from the diagram, correcst me if i'm wrong) that the hall sensor convention is 120 deg.

I called up Solarconverters and asked them what the convention was for their controls. Apparently they're 120 degrees as well, which begs the question: what is going on?

Further, if the transistors DID get turned on at the same time, and there was a bit of smoking... is that going to cause damage? I imagine eventually something will blow.. but at this point I simply need to get this unit working. :(
 
Upvote 0 Downvote
What a mess!

I "ohmed out" the pins on the motor. Here is the resistance between pins AS SUGGESTED by the diagram in combination with my meter:

1.1 Ohms between: VCC/GND, VCC/RLG3, GND/RLG3

45 Ohms between: L3/L2


Apparently, what is purported to be VCC, GND, and RLG3 are either the 3 Hall sensors or the 3 motor phases. Since I obviously don't know much about BLDC motors I don't know which is which. Further, I obviously sorted VCC and GND. let me write that in bold:

Thanks to the spec sheets as provided, I shorted VCC and GND over that resistor
 
Upvote 0 Downvote
Thanks for letting me pad my own posts as i think aloud;

however, after reviewing the apparent paradox, I determined that the diagram provided to me shows pin positions with respect to the motor oriented from the opposite side to which I had originally thought.

it appears that VCC/GND/RLG3 are actually L1/L2/L3 reading the diagram upside down, or inside out

In other words, the diagram given to me was as shown with the pump not mounted, because the specs are not from the pump company. If i had disassembled this pump I may have known what the motor looks like from the opposite side. Live and learn I suppose. I'll give it one more shot.
 
Upvote 0 Downvote
She works!

So... basically.. i'm an idiot; however, this dialogue/monologue encourages me to contine my quest. Thanks for your help everyone as usual! I hope to one day help someone else out ;)

 
Upvote 0 Downvote
Okay, well there is another problem here.

On solar panels, and on 18V 1.1A power supply (which I was told would be sufficient for testing), the motor runs ONLY when you tap it into the right direction... or in the wrong direction. There are 3 "dead spots" around the motor where there is no commutation if the motor is stopped, even with ample power supplied to it.

The problem is most likely not in the commutation, because I tried switching around the hall sensors in all 6 different permutations and it either ran really "chunky" (crappy), or in reverse (when the hall sensors were set up in the reverse order). Of course, it ran really crappy in reverse due to the 240 degree phase differential.

Is there any hope left out there? It is dwindling on my end. Thanks for any help!
 
Upvote 0 Downvote
I have seen this happen if the Hall signals have an illegal state. For 120 degree commutation, all three Halls high or low voltage (111 or 000)should not occur. Check the state of the Halls at the dead spots.
 
Upvote 0 Downvote
Okay. I will do that right now. For the record, I pulled out a spare "Baldor" 1/6HP BLDC motor and downloaded the commutation information. Hooked it up and it does the SAME thing, leading me to believe that either the two motors are malfunctioning in the same manner, the controller is malfunctioning, or else the two motors are "disagreeing" with the controller in the same manner.

I am getting my $2.99 "Harbor Freight" multimeter out right now.
 
Upvote 0 Downvote
Ok. Roger that.

I checked the voltage on the halls and they all read 6.1 Volts +/- 0.1

Therefore, sreid I believe that you are correct in your assessment.

Since they are in an illegal state, does that mean that the controller algorithm is jacked up a bit?

Woah is me.. woah is me.

THANK YOU SREID!
 
Upvote 0 Downvote
Ok. I talked to the developer of the controller

He said something to the affect that the motors are causing this problem (imagine that).

Well... I've come to realize that I understand electricity and computers just about as much as a really smart silverback gorilla.

Hall sensors are "delivering" this information and they should therefore NEVER deliver an illegal state. There is no doubt that the controller has no idea what to do about this; however, I'm at a loss to understand WHY this is happening.. Ask the manufacturer is what he told me.... Good luck.... i don't speak german... and my baldor contact is really "nice" but he tends to act like the motors run off of magic.. I'm beginning to believe him.
 
Upvote 0 Downvote
So what are the states? 000 and 111 are legal states for 60 degree Halls. There should be six states per electrical cycle (every 60 electrical degrees). There will be multiple cycles depending on the number of poles in the motor.

2 pole-one cycle

4 pole-two cycles

6 pole three cycles
 
Upvote 0 Downvote
34B Brushless DC Series Motors
(1/5 - 3/8 HP)

Motors may be used with a variety of controls, including servo amplifiers, that produce square-wave current for 3-phase, 4-pole brushless motors with 60° commutation

I can't contact any of the reps right now since everything is Eastern... but... I'm guessing that they both have "non-standard" commutation.

 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

Malik009
  • Question
Designing a BLDC motor 1
Replies
2
Views
32
waross
waross
NickParker
  • Locked
  • Question
Ex "Px" Motors
Replies
1
Views
20
Gr8blu
Gr8blu
NickParker
  • Locked
  • Question
Ex "db" or Ex "db eb" 1
Replies
3
Views
15
SilverfoxUK
SilverfoxUK
BlackJackJacques
  • Question
REMOTE START / STOP FOR LVR MOTOR CONTROLLER FIRE PUMP
Replies
14
Views
80
BlackJackJacques
BlackJackJacques
Giuseppe.Ing
  • Locked
  • Question
BESTCOMS_DECS100_11000 - Error "Communication port is not responding"
Replies
6
Views
39
davidbeach
davidbeach
Back
Top