FIXED: broke my golden motor controller any one fix these?

[whereswally606]

So a while back I hooked up a huge scooter motor rated at 48v and 4000w (4 x the power of my magic pie) to my magic pie external controller. it worked for a while then it died. i opened it up and i cant see anything burnt or crispy. using the programming cable i can change the setting and they remain in the memory the next time i recall the settings so that at least is not broke. I have since replaced it with the exact same model since i built a purpose build bracket for my fram and i liked the functions like regen and programmability.

sooooo can anyone fix these? preferably someone uk based?

also i have a 24volt controller no brand off a cheapo lead sled. I let the magic smoke out of that one properly, caps blew and garage ended up reeking. If anyone wants it they can have it for delivery cost.

 

[whereswally606]

really no one wanting to have a tinker with this?

 

[MadRhino]

I guess your small controllers are not worth the labour, and you need a much bigger one anyway to pull the best of your scooter motor.

 

[whereswally606]

so, after it being on my desk as a paper weight i feel like i should have another go at repairing my golden motor controller. I really like the regen and i have no qualms about max Voltage being 48 volts, infact that works well with my wattmeter, (too cheap to buy a CA, and too broke). This is because I don't believe that the hua tong i have puts any juice back its all take take take and Id like both my bikes to have parity of power so my accompanying rider doesn't just open the throttle on the lipo bike up and do one.

I had a go tonight at looking for solder which looked to have no continuity, i found 3 potential spots so went to work on them, hooked it up to a known working bike but nothing. just lights on the throttle to say juice and a noise in the mp hub as the battery turned on. feels like when the controller is on it adds resistance to the motion of the wheel a little. I know this is the case if phase wire are shorted but i don't know what this means to the motor.

as for programming the chip seems to be fine to flash with the golden motor program.

I have an identical controller which i tried to open up tonight with my security bit screw driver set unfortunately although I have the correct bit two of the screws were so soft they deformed on turning so they will have to be drilled out and replacement screws found. it also delays my looking at a working controller innards to see if there is anything drastically obvious missing.

the reason it bust as i stated previously is that i was using it to test a 4kw emax motor which it seemed to be able to do with no load. (but actually maybe as madrhino says it couldn't really cope)

anyway Ive got a lyen test but i've never used it and I like anybodies input as to what to test and how to test it. thanks in advance.

 

[Ricky_nz]

I guess this is one of the magic controllers.
I killed one probably over a year ago and definitely blew a FET (no visible damage but measured short) It took out some of the gate drive circuit that controls the FET but its all surface mount and although I have no problems working with surface mount parts it just wasn't worth my time to try and trace it out.
The copper buses over the top don't help either although I could have possible removed them.
Maybe they have improved serviceability since my early one but I'm guessing not :lol:.

It sounds Like you may have a dead FET or two. You could look for the shorted FET and try replacing it but the controller may still not work requiring tracing out of the gate drive circuit. There is plenty of information around on testing FETs so I won't repeat it here.

 

[dnmun]

do you have a voltmeter? if you have one we can help with advice. if not then go to harbor freight and buy one or two of the cheap $2.99 specials.

 

[whereswally606]

right did some reading then some testing.

having read mr fechters words on testing fets. I think ive found a culprit. on the yellow and blue phases i get a steady 7.9 k ohms but on the green phase i get something slowly rising up to 3.4 ish. then i tested the actuall fet legs with the continuity checker and only one beeped. so im guessing that is shorted. worried also about the gating circuitry but i just got some new solder tips so im willing to give it ago.

on the back of this fet it reads
NF75
Vb
737
anyone care to translate?

also i have a lyen controller tester so that should make this easier right?

 

[Ricky_nz]

I haven't read fechters notes so I'm not sure exactly what steps you took but you can get a beep from a continuity tester if it goes through a diode so when checking for a short swap the probes to make sure its a real short.

on the back of this fet it reads
NF75
Vb
737
anyone care to translate?

Its a likely 75V N Channel MOSFET.
Not sure if there are part number digits missing or its a generic device.
The STP140NF75 MOSFET N-CHAN 75V 120A TO-220 seems to be along the lines of what is needed. Since FETS are relatively expensive you really want to make sure they need replacing first.

 

[whereswally606]

Checked. Both ways. Beeps both times no other fets showing shorts though, if I replace the fet and not the resistors for gating will the fet just blow again, I thinking yes but I'm not sure.

 

[whereswally606]

If you are going to quote then use quotes if you are going to copy from another contributor then don't do it on the same page of the thread, if you were just pointing out the futility then please don't. Got to be optimistic else I'll never learn how to repair these things

 

[whereswally606]

Guess who can fix them? I can fix them Boo Yaah!!!!

After a message on another thread about upping the amount of volts through the golden motor from 54 v to up to the limit of the caps ie 63v
Cohberg suggested why do I not swap out the caps since the mosfets are probably good up to 75v. He is probs right about it being way cheaper than my DC DC drop down method i came up with but a bit more risky and i thought i only had one controller to play with (read: risk bricking) anyway then i remembered id gotten half way through mending this other one (older but pretty much identical outside the case) controller. I'd sourced the correct fet and i was just worried about gating resistors that may have been fried too.

In light of the fact i wanted a controller i could afford to break i went ahead an did the mosfet replacement and ignored any potential the gating resistors were broke too. Turns out it mended it. Very happy now indeed.

some pics of my ghetto testing setup.

testing setup with lyens device, very handy

older bcm318 above newer bcm328 pcb

 

[Punx0r]

Now you just need to modify it to increase the current

 

[Ricky_nz]

In light of the fact i wanted a controller i could afford to break i went ahead an did the mosfet replacement and ignored any potential the gating resistors were broke too. Turns out it mended it. Very happy now indeed.

Cool, you got lucky with the gate drive resistors and transistors .

Interestingly looking at the pics of your board, it is a little different than my old one that I gave up on.
On mine the bus bars were over the top of the gate drive components making their repair difficult.
The layout of yours looks better with the bus bar on the other side of the board. I'm guessing yours is a later revision than mine.

 

[whereswally606]

Damn, controller broke again. Looks like I well an trully killed the fets. not sure about the gate resistors. dont know whether to start again with resoldering new better fets. or buy another controller. supposed to be going on a biking holiday in the peaks in 2 weeks, have a hua tong in reserve but the braking on that is shockingly delayed.