[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Small DRSSTCs



Original poster: Steve Conner <steve@xxxxxxxxxxxx>


Question - are IGBTs available that can work well at much higher Fs?

My coil- which is just small enough to run on a tabletop- runs at 220kHz using the well known HGT1N40N60A4D devices from Fairchild. (The so-called 40N60s or miniblocks.) I have seen other builders go up to over 400kHz using other devices in the Fairchild ultrafast range, like the FGH50N6S2D. Of course the higher the frequency you run at, the more benefit you get from using a PLL to overcome gate drive delays. ;-)



2) Suppose one wanted to buiild a twin coil DRSSTC.  Does anyone see a
problem in using feedback from just one of the primaries or secondaries?

I have successfully run mine as a master/slave twin, with the two secondary bases connected together but not grounded. One resonator was driven by a primary as normal, and the other driven only by the base current from the master resonator. I got streamers about the same length as with a single coil, and they would connect. But I tended to get very high voltages at the secondary bases if one coil arced to ground and the other didn't. I used a ball gap between the secondary bases and RF ground to stop too heavy arcing to the primary.

I have posted about this setup on the list before, and also experiments where the two coils were not connected at all- the slave was driven capacitively. I could not get connecting arcs with this one.

However I think you would get results as good, if not better, using the classic configuration that has two identical primary/secondary systems, with the two primaries connected in series and sharing a single tank capacitor and driver. Feedback would be from the single primary current. I have tried using about 3ft. of #14 wire to connect the H-bridge to the coil, and it made no visible difference to the performance, so I don't see a problem in wiring the two primaries together with relatively long leads. Of course they should be as close together as possible to minimise inductance.

It's still good to connect the secondary bases together instead of grounding them separately, as it suppresses the "Roll mode" where both top voltages oscillate in phase and you get streamers that avoid each other. That's one less mode to worry about. Modes can be a problem as a twin coil system seems to have at least three resonant frequencies. Again I found my PLL driver a big help as I could just tell it which mode I wanted to drive, and sit back and watch.

Steve Conner
http://www.scopeboy.com/