Re-Repairing my Curtis 1231C Controller

[ted.lowe]

i hope y'all enjoyed my story last night of how my "first" repair didn't quite work (OBVIOUSLY A MAJOR MALFUNCTION) and digging into the bowels of the charred critter for evidence of what went wrong.  i'll post more information/pictures as they come.

It's funny how almost EVeryone suggested that i test the unit on the bench before applying full power (120V and hundred's of amps).  i wish i had that hindsight up front :-)  My only other regret is that i didn't have my video camera running when i first powered up the "repaired" controller (i've never seen such pretty colors of flash and flames - eg, lots of melted copper:-)

This leads to a couple of question.  i need a suitable power supply to bench-test my controller.  It needs to be 70+v DC (since the 1231c's under-voltage cutback spec is 64v).  What amperage should the supply have ?  1A, 5A, 10, 20? Fixed or adjustable ?

The test load can be a series of (wasteful) incandescent light bulbs as suggested in the 1231c manual (which is good reading to learn more about controllers, etc.  http://www.curtisinst.com/index.cfm?fuseaction=cProducts.DownloadPDF&file=1209.pdf)

Any suggestions anyone ?  Thanks!

[miodrag.zubic]

5A would be good, but even 1A power supply would be sufficient.
Light bulbs are OK to test it, but they don't really test the freewheeling diodes. Having a small 90V dc motor (or a universal motor, such as those in hair dryers, corded drills and so on) would do that too.

[tim.moore]

Ted so sorry for your loss.  I think this is the curse of the presidency.  You just did this to teach us more about Curtis controller.  You are the best president ever.

[nathan.stowe]

Hey ted,

I know you have Miodrag's controller now, but did you ever get your Curtis back to life?  I have two dead soldiers now.  one has all blown mosfets and the other probably has the a blown logic section.  So I will see what I can do to combine them.  (nothing to loose right?)

Rich Carol took a look at my current fatality, and reminded me of my missing "pre-charge" resistors.    (resistors across the cut off for the controller which let a little current through to keep things semi-ready to go in the controller.  This avoids a sudden in-rush of current when the contractors close)  I had purchased my components used and already mounted to a board from another RX7 and ready to drop in mine.  I noticed the missing pre-charge resistor, but since it had been working for so long for the other guy, I did not worry about it.  It did not seem to bother the old controller that I had, as the failure mode seemingly had nothing to do with it, but rather the common Curtis failure where the mosfets closest to the control circuit smoke first.  The replacement controller from Rich might not have been so lucky.

Can you post how you actually opened the Curtis?  I am going to give it a go now.  I have read boiling them and putting them on a stove to soften the sealer which covers the screws to release the innards.  I think I am going to give a torch a try.   Intense localized heat to soften the goop, while hopefully not heating up the other sections too much due to the nice aluminum case and fins.

[miodrag.zubic]

Even though I don't have a very good suggestion on how to remove those screws, I would try to avoid the torch. There is a quite sensitive material (SilPad) between the aluminum housing and an internal u-channel.
I remember opening a Curtis 1221B and the way I dealt with that was by carefully drilling through the goop until I reached screw heads. Since the "goop" was quite rubbery, I then used a screwdriver and I was able to release the screws just by pushing through what was left of the "goop". I don't know if they still use the same "goop" material, though.

[ted.lowe]

i used a sharp exacto knife and sliced away at the edges of the potting material.  It wasn't that clean but it worked.  i'd avoid the torch but perhaps a carefully controlled heat gun could help soften the material.  The boiling water (or even sub-boiling) would work(help) too i suppose.

Removing the control card is very easy, i think only 4-5 through-hole leads to unsolder.  Solder wick/sucker is a friend :-)

Definitely be careful with the insulating material that Miodrag mentions... i still don't know what caused my 'immense failure' upon re-powering after the 'repair', but a small short through that insulation could have been the cause.

Good luck in your journey!