Epic 7kw+ 12 fet controller...or there abouts :)

Even just one on that back end plate would do a lot of good just by getting the hot air out of the controller. Since there was nothing to get hot on the bottom of my boards, I blocked that off as a path for air to flow to help focus flow where it was needed most. These days, since I use 6 phase motors, I don't need super powerful controllers, so I run them stock and at conservative settings. 100A each from the battery with less than 2:1 phase/batt current limits at voltages from 74V nominal to 115V nominal is typically plenty...especially since I have the modded MadAss running at 30kw+ peak input for thrills.

John in CR said:

I use 6 phase motors

Click to expand...

Where does one get those 6 phase motors?

shaman said:

John in CR said:

I use 6 phase motors

Click to expand...

Where does one get those 6 phase motors?

Click to expand...

Depends on the power you want...

John is talking about hubmonsters. He has several of them, but they are not cheap. I have 2 of them. It took looking in local for sale sites a long time to find them on existing electric mopeds.

Just about all Motenergy axial flux motors are dual stator or 6 phases. Again not inexpensive.

You can take any outrunner and wind it for 6 phases. I have an Alien Power 12090 that's burnt out. I'll eventually get around to winding it for 6 phases.

Well it's definitely not going to be waterproof or water tight with those fans... After a 14mi ride on my controller pulling anywhere from 600-1500w...more around 650W consistently I had absolutely no heat from my controller. Pretty amazing! Well see what happens when it's 90F out but I'm not worried at all pulling 2K through mine. I also added new thermal paste to the mosfet heatsink block.

I better be seeing wheelies from this controller!

I posted back a few pages in this thread about losing a few mosfets in the controller. I was running an Alien Power 12090 when it happened.

A couple of days ago, I connected that exact same motor again and sure enough, popped mosfets again. Grrr!

Back a few months, I assumed there was a couple of weak mosfets...it happens form time to time.

I took the 12090 apart and found I have a partial phase short to the stator. The motor is effectively brand new as it's never seen more than bench use. This is why I'm blowing mosfets. There's no evidence that a motor phase is damaged.

So from your extensive experience with modding the controller where do you see the weakest link to be? I'm considering adding a little more solder to the shunt for 55-60A on mine... Im thinking the 14-16awg phase wires going into the motor that are only about 6" long won't like those phase amps...

Philaphlous said:

So from your extensive experience with modding the controller where do you see the weakest link to be? I'm considering adding a little more solder to the shunt for 55-60A on mine... Im thinking the 14-16awg phase wires going into the motor that are only about 6" long won't like those phase amps...

Click to expand...

Are you talking about the 12 fet PV controller or something else? No PV controller uses 14 or 16 awg wires anywhere.

Adding solder to the shunts...
just don't. If you want more amperage do it right and use better shunts.
Also, if you are using a PV controller, this is the wrong way to go about doing things. These controllers are fully programmable.
"buttering up" shunts is what you do to cheap Chinese dumb controllers that you buy on ebay for $40.

ElectricGod said:

Philaphlous said:

So from your extensive experience with modding the controller where do you see the weakest link to be? I'm considering adding a little more solder to the shunt for 55-60A on mine... Im thinking the 14-16awg phase wires going into the motor that are only about 6" long won't like those phase amps...

Click to expand...

Are you talking about the 12 fet PV controller or something else? No PV controller uses 14 or 16 awg wires anywhere.

Adding solder to the shunts...
just don't. If you want more amperage do it right and use better shunts.
Also, if you are using a PV controller, this is the wrong way to go about doing things. These controllers are fully programmable.
"buttering up" shunts is what you do to cheap Chinese dumb controllers that you buy on ebay for $40.

Click to expand...

Yea I got one of those cheap ass controllers your talking about. It does have soldering points for a plug to program it but those aren't used at all.
It's kinda strange though that the controller doesn't really even get warm... Maybe it's just in such a great location on my bike? It's behind the downtube between the rear wheel and downtube. It gets airflow from the front and I believe also some from the wheel.

Interesting thred :wink:

Has there been any progress with this project?

j bjork said:

Interesting thred :wink:

Has there been any progress with this project?

Click to expand...

No much...until a couple of days ago. The AP 12090 with the shorted phase blew up the controller. I've had many things on my plate recently so this just sat. I got it pulled apart a couple of days ago and found that 10 of the 12 mosfets are cooked. I've replaced all of them, but need to test it out now.

That was a lot of money on new top end mosfets that lasted just minutes! GRRR!

Well that controller is a lost cause...
Not only were the mosfets toast, but the control electronics were fried too. That Alien Power motor really messed things up badly!

I have another of the 12 fet PV controllers. Last night, I got it mostly modded. The 8 awg wires and big shunts are added. The phase busses still need some beefing up. I'm going to rework my copper heat spreader a little before installing it in the beefed up shell.

It's been a very long time since I worked on this project.

I had to shelve it for a while and over the weekend got back to it.
You could call this a failure or a success...depending on point of view.
I'd say it is a success...I got 7kw and the controller held up quite nicely with the added cooling.

I was measuring 100 battery amps at 82v...so 8kw.
No idea about phase amps. I was about to check that when the controller blew up.
I think I've found the maximum limits I can achieve with this controller!
As you can see all I changed in it is the mosfets and added larger shunts and wires.
7kw was working pretty nicely with the added heat sinking.
8kw exceeded the limits of the board and mosfets.
I was already thinking 7kw was pretty impressive!

I'm not going to try and repair this controller!
There's circuit traces burnt off the board.
I have no idea how many mosfets are cooked, but many exploded.
Probably the MCU and driver components are fried too.
You will notice that I'm not using ceramic insulators.
I had a few of them break so I stuck with kapton for this test.
I had a temp sensor on a mosfet tab and another outside the case on the heat spreader.
Temps on the mosfet spiked suddenly and then this happened.
7kw...doable...8kw...no...not really.

This has been an epic journey, thanks for sharing it with us!

Would you do a short recap of the mods, and what configuration worked best? And shunt mod, how did you get to 7kW specifically?

That's some impressive toast there. I'd guess one of the FETs shorted, but sort of hard to see with all the carbon. Looks like it blew one trace off the board. It's remotely possible the trace blew first then things cascaded.

fechter said:

Looks like it blew one trace off the board. It's remotely possible the trace blew first then things cascaded.

Click to expand...

It could have happended that way.
When i repaird my two 12F back then, i noticed the issue and beefed up those traces. They supply batt minus to one of the two low-side FET's so it has to carry around 50% of the phase current.