Calculating MMC size for full bridge SSTC / DRSSTC... / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Calculating MMC size for full bridge SSTC / DRSSTC...

Move Thread

LAN_403

Nicko

Thu Oct 29 2009, 09:45PM

Registered Member #1334 Joined: Tue Feb 19 2008, 04:37PM
Location: Nr. London, UK
Posts: 615

Most of the MMC calculators are geared towards "traditional" TCs... there seems to be a consensus that DR/SSTCs may be a special case...

Are there any good on-line calculators or rules that can be applied here? What about voltage rating too?

many thanks

MRacerxdl

Thu Oct 29 2009, 10:18PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

For SSTC's, the cap is only for DC block, so you need much higher value you can get, but anything higher than 5uF will rare make any difference. Most uses 1uF and are good.

About DRSSTC, the calculation is the same for common TC, the ressoant formulae is the same. About the voltage rating. I made a document about the voltage rise on ressoant devices:

(Edit: Translated now =) )

You can predict the voltage that will have on your DRSSTC:
You know the operating frequency, and know what is the capacitor value you need.

For an 50nF cap running at 100Khz you can calc the reactance of it:
XC = 1 / (2 * pi * f * C)

Usually the DRSSTC have a control of the primary current. If so, you can know what is the voltage at the cap with ohm law...
The reactance for 50nF at 100kHz will be aprox 31 Ohms, with a 300A primary current you will have 9300V
U = R * I
U = 31 * 300
U = 9300Vrms

but for peak values you multiply for 1.414 (sqrt(2) because its sine wave)
so its 13150Vp

Its simply, my document explains with a few screenshots of the Spice simulation.

Other differences about DRSSTC and common SGTC's, its about the primary frequency needs to be a little lower than secondary, but I dont know why, but it works better. Someone with more experience can say better than me about this =)

I hope I helped =)

Nicko

Thu Oct 29 2009, 10:58PM

Registered Member #1334 Joined: Tue Feb 19 2008, 04:37PM
Location: Nr. London, UK
Posts: 615

MRacerxdl wrote ...

...
About DRSSTC, the calculation is the same for common TC, the ressoant formulae is the same. About the voltage rating. I made a document about the voltage rise on ressoant devices, but its in portuguese:

(I hope I can translate it soon)

Thanks for the feedback - Google is your friend!

lightlinked

Thu Oct 29 2009, 11:11PM

Registered Member #2087 Joined: Tue Apr 21 2009, 08:32AM
Location:
Posts: 115

someone should add the calcs to the wiki

i am not qualified to

MRacerxdl

Thu Oct 29 2009, 11:39PM

Registered Member #989 Joined: Sat Sept 08 2007, 02:15AM
Location: SÃ£o Paulo, Brazil
Posts: 476

Hehe, sorry, I translated it now (I was with nothing to do here)
Updated the link. Now its in english:

A few gramatical errors, but you can understand =D

Nicko

Fri Oct 30 2009, 12:55PM

Registered Member #1334 Joined: Tue Feb 19 2008, 04:37PM
Location: Nr. London, UK
Posts: 615

So,

Now I'm knocking up a quick demo MMC - I've a crate load of 942C .15uF 2000VDC caps and some 10M 0.5W carbon resistors...

So, I was going to make a 75nF 12KV MMC just for a starter. This works out at 3 strings of 6, i.e. 18 caps and 18 resistors.

Question: Why can't this be made as 1 string of 6 x (3 caps in parallel + 1 resistor)? This uses 12 less resistors and could be more modular. Same voltage, same capacitance.

Any reason to avoid this technique and stick to 3 distinct string?

Thanks

Nick

Dr. Slack

Fri Oct 30 2009, 01:49PM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

This uses 12 less resistors

12 fewer resistors

The two arrangements are identical to first order.

Second order effects may or may not be significant depending on what's important for you.

It's easier accidentally to get lower MMC inductance with 3 parallel strings than one, but if you're designing for low inductance then there need be no real difference. While MMC inductance will be resonated out, it reduces efficiency slightly.

You can do an FEMA (Failure Mode Effects Analysis - beloved of miliatary contractors) to see how many extra components you may lose if one goes down in either arrangement, but that's probably more trouble than it's worth.

Less electronic hardware sounds good. But OTOH, 3 discrete strings allows you to change capacitance quickly at the same voltage more easily than a single string of triple units.

Mads Barnkob

Fri Oct 30 2009, 03:05PM

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

Having only one resistor over 6 capacitors will only make the load sharing between your strings instead of load sharing for each capacitor, so the risc of failure is greater!

Remember to design your MMC to withstand the RMS current, in a DRSSTC that is just as important, if not more than the voltage rating.

So you might need additional strings in parallel to obtain the current rating needed.

For my own first DRSSTC I made a spreadsheet with all the DRSSTC related equations I could find, so I made myself a design guide to experiment with different load settings, here is a screendump from my project thread:

There is no guarentee for it being valid or without errors, that is why it is not public shared for now, but I can send it to you if you want, but to be honest, I dont like spreading something that is not done or might contain errors, its so damn hard to get off the internet again :D

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