Large DRSSTC [update] / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Large DRSSTC [update]

first previous 15 16 17 18 next

Move Thread

LAN_403

Linas

Thu Jul 17 2014, 12:41PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

Mads Barnkob wrote ...

Are you sure the driver is the problem? It sounds like it could be a similar problem that Kizmo and I had, residual MMC charge that messed up the drivers edge detection.

Topics about it:

and

Original design works, i mean simple CT + 74HC14 +UCC, but you know, i want to have better ZCS than have now, as far as i understand, it can yield more life out of CM600. I hade this problem right from the start with predictor.

So all i have to do is to place bleed resistors on my MMC, and try it again ? (or i must add on each capacitor ?, that will give better balancing)

Mads Barnkob

Thu Jul 17 2014, 01:07PM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

Linas wrote ...

and

If you are still using your CDE 942C MMC, it would be a good idea to add bleeder resistors on every single capacitor to insure there is no residual charge. The reason we used one large resistor on the inverter output was to avoid building high voltage resistors to put across the large GTO snubber capacitors we have used.

Linas

Thu Jul 17 2014, 01:31PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

any particular value should be used?

i have 0,56uF 1600V, 5 in series, 6 in parallel, if i add 5W resistor for 1600V, that will be 512K on each capacitor, and at 8KV i will dissipate around 150W into paraffin filler that is inside my MMC capacitor (no air cooling, only large paraffin bock). Sounds ok ?

Steve Ward

Fri Jul 18 2014, 07:46AM

Registered Member #146 Joined: Sun Feb 12 2006, 04:21AM
Location: Austin Tx
Posts: 1055

I think its much better to put the resistor at the inverter output because power dissipation will be much lower (less peak volts there).

You could still add high resistance bleeders to the MMC caps if you feel there is a risk, but dissipating 150W is a lot.

Its funny how many ways the driver can fail to start nicely. I wrote a paper about why the bleeder on the inverter output works:

Hope that helps explain it.

Linas

Fri Jul 18 2014, 08:31AM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

Steve Ward wrote ...

You could still add high resistance bleeders to the MMC caps if you feel there is a risk, but dissipating 150W is a lot.

150W is peak power if capacitor voltage will reach 8KV. in reality worse case scenario will be duty cycle multiplied by peak power, for my system that will be 150W*0.38. That translates only to 57W. now if we consider that i have AC voltage ( which i do ), total power level will never reach 57W/sqrt(2) and that is only 40W. This is nothing, and i can live with that.

I am also using comparator to generate feedback, that can be adjusted in +-100mV range around zero, and i have clamping diodes to comparator power supply. As far as i remember, i never get any problems with oscillation, and instability, i just don't get clean signal in IGBT Vce waveform. And of course my driver use GDT only for insulation, i have additional power stage to drive IGBT fast, and i GDT will always get waveform that integral over one period will be zero.

Anyway, plan is to add 9W 300K resistors on each capacitor, and add inductorless resistor on IGBT output at around 10K, that will dissipate around 11W at worst case running at highest power setting with largest Vbus voltage.

Linas

Sun Jul 20 2014, 10:57AM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

I have idea.

If i feed signal generator to my feedback 74HC14 chip (just after feedback comparator and delay circuit), and i sweep my LC series circuit with CM600 H bridge, at some frequency i will get perfect zero current switching, and this will be my reference how good signal can be ? (since i will not have any delay in feedback circuit, so H bridge at resonant frequency will be forced to work in ZCS mode)

Does that make any sense ? (also, at this point, if i take my CT feedback, i can calculate how much lag i have, and how much compensation i need )

Linas

Thu Aug 14 2014, 07:19AM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

Since i am redesigning my coil driver, i will remake my current transformer as well.
Right now i am using 1:300 transformer, made from single core.

My question is, do i make single 1:500 or 1:1000 transformer, or i should do 1:33:1:33 from two toroid's ? which is better for phase lead application ?
Also, what is best core material for this ? my coil is around 50kHz +-20kHz

1408000969 1143 FT66676 Untitled

Steve Conner

Thu Aug 14 2014, 08:57AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Re core material, highest permeability is usually best for GDTs and CTs. 3E25 in this case, though I think 3C90 would work well too.

I've always used a single CT, 1:33 for smaller coils and 1:100 for bricks, and just made the burden resistance smaller.

Linas

Tue Aug 26 2014, 08:33PM

Registered Member #1143 Joined: Sun Nov 25 2007, 04:55PM
Location: Vilnius, Lithuania
Posts: 721

Steve Conner wrote ...

Re core material, highest permeability is usually best for GDTs and CTs. 3E25 in this case, though I think 3C90 would work well too.

I've always used a single CT, 1:33 for smaller coils and 1:100 for bricks, and just made the burden resistance smaller.

Ok, i will use 3E25. Now if i have 1:100 for my feedback, what would you recommend for CM600 full bridge with 2-3kA peak current ? ( since i had 0Ä® success with predictor, i will better ask what should i do, than try with my own logic, that get me nowhere )

Mads Barnkob

Thu Jan 29 2015, 08:20AM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

Hey Linas

Do you have any news on your coil and very long work process to share? :)

first previous 15 16 17 18 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob