SISG New MMC Cap Construction Method
Terry Fritz, Thu Aug 10 2006, 04:12AMHi,
With my SISG computer model finally working, I was able to run a bunch of numbers today and decided my present cap value was too high and should be reduced just a little for optimal performance from one MOT (but the present one has a much wider BPS range!!). So I found I had the parts on hand to make the new MMC!!! This time, I took pictures...
First I wired up the caps with solder, bleed resistors, Kapton tape, G-11 insulator sheet wire terminals...
It does not matter how it "looks" as long as it is sound electrically. You are never going to see it again
) All connections are heavily soldered and the ring lugs are soldered to the brass bolts. Then, I took the plastic gutter pipe and cut it to 8 inches and made holes for the bolt terminals. You then install the guts...
The bolts are held with loctite 271 and are D*** Tight!! They should never come loose, but if they do, the foam holds the other side so they can be re-tightened anyway.
Then I masked off the pretty parts and covered the bench with plastic and tape getting ready for the foam fill.
Then you just blast the inside with foam from the inside out trying to avoid big air spaces:
After a few hours it expands and does its thing as it pleases:
The Dow foam really does not expand super much which is good.
Later tomorrow, I will cut off all the dried excess gushing from the ends. Then let it dry for about a week to be sure all the foam inside is totally dry (but it can be "used" right away) If there are any big gaps, I can always re-foam them again...
After about a week, I will cut out shallows in the foam ends and fill them with two layers of epoxy. Then just mask off and paint the ends with white paint. It should then look just as good as any commercial cap. I may add mounting tabs too with the epoxy step above.
This cap will be used right at all its ratings. But if it blows up, all the caps will be dead anyway so no need to worry about "fixing it". But it should do fine and having the cap part rock solid is simple and cheap with this method!!
The gutter pipe was $9 for ten feet. The can of foam was $6. You can pick out the dry foam from the can tip, applicator, and ream the tube very easily to use it again. It does seem to expand a little less with many uses. Maybe I did not shake it enough though... Be sure to read all the instructions for the foam since it can be nasty if misused...
Cheers,
Terry
Re: SISG New MMC Cap Construction Method
Hazmatt_(The Underdog), Thu Aug 10 2006, 04:20AM
ever thought of making a U shape out of the drain pipe, greasing it up and casting your caps in resin? You can still see inside of it, and there are no gaps if you mix the catalyst right.
Just a thought.
Oh, and it is really strong stuff...but you can crack it if you hit it with a hammer. So I guess in this case you would fix it with a hammer. heh.
Hazmatt_(The Underdog), Thu Aug 10 2006, 04:20AM
ever thought of making a U shape out of the drain pipe, greasing it up and casting your caps in resin? You can still see inside of it, and there are no gaps if you mix the catalyst right.
Just a thought.
Oh, and it is really strong stuff...but you can crack it if you hit it with a hammer. So I guess in this case you would fix it with a hammer. heh.
Re: SISG New MMC Cap Construction Method
..., Thu Aug 10 2006, 04:26AM
Wow, it is things like this that make you stand out
How is the foam thermally? Seems like you could end up cooking the caps to death...
Lets hope it works
..., Thu Aug 10 2006, 04:26AM
Wow, it is things like this that make you stand out
How is the foam thermally? Seems like you could end up cooking the caps to death...
Lets hope it works
Re: SISG New MMC Cap Construction Method
Terry Fritz, Thu Aug 10 2006, 04:38AM
Hi,
I really don't "want" to see inside in this case
Trying to hide the fast constructing of the guts 
Yes, it is a thermal insulating foam. But the caps ratings are at 70C. The "actual destruct" for MMC caps seems to be at twice power or SQRT(2) current. These are GE 330nF caps I got from Dan a while back and he says they are good The coil's guts will be forced air cooled so I "think" it is "ok".......
Cheers,
Terry
Terry Fritz, Thu Aug 10 2006, 04:38AM
Hi,
...You can still see inside of it...
I really don't "want" to see inside in this case
Trying to hide the fast constructing of the guts How is the foam thermally? Seems like you could end up cooking the caps to death...
Yes, it is a thermal insulating foam. But the caps ratings are at 70C. The "actual destruct" for MMC caps seems to be at twice power or SQRT(2) current. These are GE 330nF caps I got from Dan a while back and he says they are good
The coil's guts will be forced air cooled so I "think" it is "ok".......Cheers,
Terry
Re: SISG New MMC Cap Construction Method
Steve Conner, Thu Aug 10 2006, 10:16AM
Yea, that's the only thing I'd be worried about: the caps getting too hot inside their foam jacket. Forced air cooling probably won't make much difference since the caps are going to be insulated from the surface that gets the cooling.
I guess it doesn't matter if you're not planning to run them hard. I tried a commercial pulse cap on my DRSSTC and I've felt it get pretty hot, I'd estimate about 40-50'C. It stays warm for about an hour afterwards, so I guess there's lots of heat trapped inside
The same caps stayed pretty much cold on a spark-gap coil making about 4ft arcs.
Steve Conner, Thu Aug 10 2006, 10:16AM
Yea, that's the only thing I'd be worried about: the caps getting too hot inside their foam jacket. Forced air cooling probably won't make much difference since the caps are going to be insulated from the surface that gets the cooling.
I guess it doesn't matter if you're not planning to run them hard. I tried a commercial pulse cap on my DRSSTC and I've felt it get pretty hot, I'd estimate about 40-50'C. It stays warm for about an hour afterwards, so I guess there's lots of heat trapped inside
The same caps stayed pretty much cold on a spark-gap coil making about 4ft arcs.Re: SISG New MMC Cap Construction Method
Dr. Drone, Thu Aug 10 2006, 03:22PM
Dr. Drone, Thu Aug 10 2006, 03:22PM
Re: SISG New MMC Cap Construction Method
Terry Fritz, Thu Aug 10 2006, 09:55PM
Hi,
I am running these caps at their RMS and peak current ratings and I normally don't do very long runs. Of course, if one's caps are being pushed hard and are getting hot, then the foam would not be a good idea I can pretty much figure out the exact currents and power dissipations so I can run fairly confidently there.
Chistopher's caps have super cooling!!
I see he also puts two fans together:
I always wondered what would happen if one put like 100 fans together... Maybe like a jet engine!!
One of the fan vendors broght a "super fan" to show us one day. Something the guys in their lab cooked up. It was just like a normal muffin fan but ran at like 150 watts.. We bolted and clamped it to a bench and got far back. That thing was frightening!! it was like a mini jet engine!!! I guess they could not actually sell it since it was too dangerous and not real reliable.
I see you use the GE caps too. They are pretty nice even if they are not foil types. The data sheet is hard to find but it is here if you don't have it:
It funny, when the MMC first came out none of the vendors listed RMS and peak currents at 100kHz... Now they all do)) I railed on every cap vender I could get my hands on at the time and I got the other engineers to do the same. I think we "changed things" there Far better than the old tan delta crap... CD was first with help from the Geek group and my work at the time beat up WIMA... It is cool to have specs for parts geared right at Tesla coils )
The SISG can easily push "enormous" peak and RMS currents!!! It like to ride right up there to 1000 BPS!!! So I have to be careful and not get carried away) The IGBTs have been tested to 800 amps peak so that is where I limit them. Maybe they could take more but I was not able to dead short them hard enough to go higher at the time. But that is about the limit for a small MMC anyway...
The box and foam is a whole lot better than just filling with epoxy like I did on the last coil. The foam is almost free and does not get hot or risk crushing anything as it cures. There are a few parts like diode bridges and MMCs that are real fragile without really being in a nice box, so this really fills the bill there.
For traveling with parts, a box also makes them very damage proof!!
Besides, we have to do something to "keep up" with Dan's nice bricks
I suppose you could package and foam a whole coil too... Just a line cord out one side and the top termianal out the other
Cheers,
Terry
Terry Fritz, Thu Aug 10 2006, 09:55PM
Hi,
I am running these caps at their RMS and peak current ratings and I normally don't do very long runs. Of course, if one's caps are being pushed hard and are getting hot, then the foam would not be a good idea
I can pretty much figure out the exact currents and power dissipations so I can run fairly confidently there.Chistopher's caps have super cooling!!
I see he also puts two fans together:
I always wondered what would happen if one put like 100 fans together... Maybe like a jet engine!!
One of the fan vendors broght a "super fan" to show us one day. Something the guys in their lab cooked up. It was just like a normal muffin fan but ran at like 150 watts.. We bolted and clamped it to a bench and got far back. That thing was frightening!! it was like a mini jet engine!!! I guess they could not actually sell it since it was too dangerous and not real reliable.
I see you use the GE caps too. They are pretty nice even if they are not foil types. The data sheet is hard to find but it is here if you don't have it:
It funny, when the MMC first came out none of the vendors listed RMS and peak currents at 100kHz... Now they all do
)) I railed on every cap vender I could get my hands on at the time and I got the other engineers to do the same. I think we "changed things" there Far better than the old tan delta crap... CD was first with help from the Geek group and my work at the time beat up WIMA... It is cool to have specs for parts geared right at Tesla coils )The SISG can easily push "enormous" peak and RMS currents!!! It like to ride right up there to 1000 BPS!!! So I have to be careful and not get carried away
) The IGBTs have been tested to 800 amps peak so that is where I limit them. Maybe they could take more but I was not able to dead short them hard enough to go higher at the time. But that is about the limit for a small MMC anyway...The box and foam is a whole lot better than just filling with epoxy like I did on the last coil. The foam is almost free and does not get hot or risk crushing anything as it cures. There are a few parts like diode bridges and MMCs that are real fragile without really being in a nice box, so this really fills the bill there.
For traveling with parts, a box also makes them very damage proof!!
Besides, we have to do something to "keep up" with Dan's nice bricks
I suppose you could package and foam a whole coil too... Just a line cord out one side and the top termianal out the other
Cheers,
Terry
Re: SISG New MMC Cap Construction Method
Marko, Thu Aug 10 2006, 10:07PM
I just like the rectangular tube..
You could use poliester or epoxy instead of foam - much better heat conductor and insulates well.
It would also look 1337 like commercial flybacks, bigger caps and etc. wich are filled with that nice epoxy.. (maybe for 1.1 version of that caps
)
You can make harder cement out of resin by putting talcum powder in it, it you want to 'fix the middle' so you can make the cap stand upwards when you pour the liquid resin.
(I admit it's sometimes scary, ''what if it seeps trough''...
Marko, Thu Aug 10 2006, 10:07PM
I just like the rectangular tube..
You could use poliester or epoxy instead of foam - much better heat conductor
and insulates well.It would also look 1337 like commercial flybacks, bigger caps and etc. wich are filled with that nice epoxy.. (maybe for 1.1 version of that caps
)You can make harder cement out of resin by putting talcum powder in it, it you want to 'fix the middle' so you can make the cap stand upwards when you pour the liquid resin.
(I admit it's sometimes scary, ''what if it seeps trough''...
Re: SISG New MMC Cap Construction Method
robert, Fri Aug 11 2006, 12:24PM
Probably just something overly loud with not much more airflow then a single fan.
Stacking fans of the same type doesnt really improve anything a lot, just it gets much louder.
I only saw stacked fans once, and these were sepcially designed, counter-rotating and with different blade shapes.
These were in a slim server, 40mm size and overly loud and powerful, and drawing over 12W.
robert, Fri Aug 11 2006, 12:24PM
I always wondered what would happen if one put like 100 fans together... Maybe like a jet engine!!
Probably just something overly loud with not much more airflow then a single fan.
Stacking fans of the same type doesnt really improve anything a lot, just it gets much louder.
I only saw stacked fans once, and these were sepcially designed, counter-rotating and with different blade shapes.
These were in a slim server, 40mm size and overly loud and powerful, and drawing over 12W.
Re: SISG New MMC Cap Construction Method
Dr. Drone, Fri Aug 11 2006, 03:25PM
Dr. Drone, Fri Aug 11 2006, 03:25PM
Print this page