If you need assistance, please send an email to forum at 4hv dot org. To ensure your email is not marked as spam, please include the phrase "4hv help" in the subject line. You can also find assistance via IRC, at irc.shadowworld.net, room #hvcomm.
Support 4hv.org!
Donate:
4hv.org is hosted on a dedicated server. Unfortunately, this server costs and we rely on the help of site members to keep 4hv.org running. Please consider donating. We will place your name on the thanks list and you'll be helping to keep 4hv.org alive and free for everyone. Members whose names appear in red bold have donated recently. Green bold denotes those who have recently donated to keep the server carbon neutral.
Special Thanks To:
Aaron Holmes
Aaron Wheeler
Adam Horden
Alan Scrimgeour
Andre
Andrew Haynes
Anonymous000
asabase
Austin Weil
barney
Barry
Bert Hickman
Bill Kukowski
Blitzorn
Brandon Paradelas
Bruce Bowling
BubeeMike
Byong Park
Cesiumsponge
Chris F.
Chris Hooper
Corey Worthington
Derek Woodroffe
Dalus
Dan Strother
Daniel Davis
Daniel Uhrenholt
datasheetarchive
Dave Billington
Dave Marshall
David F.
Dennis Rogers
drelectrix
Dr. John Gudenas
Dr. Spark
E.TexasTesla
eastvoltresearch
Eirik Taylor
Erik Dyakov
Erlend^SE
Finn Hammer
Firebug24k
GalliumMan
Gary Peterson
George Slade
GhostNull
Gordon Mcknight
Graham Armitage
Grant
GreySoul
Henry H
IamSmooth
In memory of Leo Powning
Jacob Cash
James Howells
James Pawson
Jeff Greenfield
Jeff Thomas
Jesse Frost
Jim Mitchell
jlr134
Joe Mastroianni
John Forcina
John Oberg
John Willcutt
Jon Newcomb
klugesmith
Leslie Wright
Lutz Hoffman
Mads Barnkob
Martin King
Mats Karlsson
Matt Gibson
Matthew Guidry
mbd
Michael D'Angelo
Mikkel
mileswaldron
mister_rf
Neil Foster
Nick de Smith
Nick Soroka
nicklenorp
Nik
Norman Stanley
Patrick Coleman
Paul Brodie
Paul Jordan
Paul Montgomery
Ped
Peter Krogen
Peter Terren
PhilGood
Richard Feldman
Robert Bush
Royce Bailey
Scott Fusare
Scott Newman
smiffy
Stella
Steven Busic
Steve Conner
Steve Jones
Steve Ward
Sulaiman
Thomas Coyle
Thomas A. Wallace
Thomas W
Timo
Torch
Ulf Jonsson
vasil
Vaxian
vladi mazzilli
wastehl
Weston
William Kim
William N.
William Stehl
Wesley Venis
The aforementioned have contributed financially to the continuing triumph of 4hv.org. They are deserving of my most heartfelt thanks.
Registered Member #141
Joined: Sat Feb 11 2006, 01:14PM
Location: Southern California
Posts: 96
From Steve Ward:
Keep the large filtering capacitors electrically close to the bridge as well. The large electrolytic capacitors serve as an energy resivior supplying hundreds of amps on demand. Too much inductance or resistance in this path will hinder performance. Having the main filter capacitors close to the bridge also reduces the burden on the decoupling capacitors. I typically size these filter capacitors such that they store maybe 20 to 50X the intended "bang" energy. The bang energy is simply given by (input power)/(bangs per second). For example, my DRSSTC-1 uses 880J of storage and runs about a 12J bang (it can really support much more!). Really, I could have used as little as 375J and have done ok (this has been tested) but these capacitors were on hand.
*Edit: Nevermind, I just realized that this does not answer your question.
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
You can't find an on-resistance figure because IGBTs don't have on-resistance.
That's not to say that they're superconducting, though. They're non-linear, so the V-I characteristic isn't a straight line and hence Ohm's Law doesn't apply. You can find a little chart of the characteristic in the datasheet. However, most DRSSTC builders seem to overdrive them right off the chart.
The current in the tank circuit is only limited by the resistance, but in any efficient coil, the resistance is so small it can be ignored. So how come the current isn't infinite? The answer is streamer loading: the sparks consume the power and stop the current from building up too high. If anything goes wrong, the tank current can indeed tend to infinity, and this is what the current limiting circuit is for!
Anyway, this means that you can't use the tank circuit resistance to calculate the primary current. Here's what you do: Pull a primary current figure out of a hat. You chose 1200 amps. Calculate the reactive voltage across your tank cap for this current. Then set your current limiter to 1200 amps and it will also protect your tank cap from overvoltage. Finally fiddle with your primary tap until you actually get 1200 amps.
Registered Member #1900
Joined: Fri Jan 02 2009, 06:44PM
Location: Texas
Posts: 29
That's pretty interesting Steve, thanks for that. So I guess actually calculating the current through the system is mostly futile, and just goes by whatever you set the limiting (as you say). BUT, if for some reason you design the system for too low a limit current and its constantly tripping (probably not a good thing), how would you actually lower the primary current to prevent this? Maybe detune the primary like you say? I guess you could also lower the duty cycle and input voltage.
Maybe i'm thinking about it too much, but I don't like unknowns :P
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
Well then you have a problem, because the way that streamers load the coil is just that, unknown. And it's the streamer loading that determines the primary current.
Going by my own experience, and that of other DRSSTC builders I've corresponded with, the following rules of thumb seem to work out OK:
1. Build your secondary with the lowest resonant frequency that's practical (ie the finest wire gauge you can handle) and make it good and stubby for increased coupling. Toroid size should follow the guidelines given for spark-gap coils, if you want it to break out without a breakout point. Zo should be about 50k ohms.
2. Choose the current limit according to the size of coil, size of sparks you want, what size of IGBTs you can get and how hard you dare to overdrive them, etc. You can work back from the spark length you want, through Freau's spark length equation to get power, then divide by 120 to get bang energy, then calculate the required peak current from that assuming a 150 microsecond burst. (Longer bursts than this don't really help, nor do higher break rates, so you should use the 150 and 120 figures irrespective of what burst length and break rate you're actually going to run.)
3. Design the primary circuit to have the same resonant frequency as the secondary, and about 10 times the inverter's DC bus voltage dropped across the capacitor when a current equal to the current limit is flowing. From these two constraints you can calculate both L and C. Use the highest coupling possible without flashovers.
4. Put it all together. Adjust the primary tapping point until it works properly.
Note this is the "classic" method, it may even be in the HVWiki. Ward has some newer ideas which I think could work out better.
Registered Member #1900
Joined: Fri Jan 02 2009, 06:44PM
Location: Texas
Posts: 29
That all looks good... but regarding point 3: I was told that streamers (when active) lower the resonant frequency of the secondary... and to design the primary up to 20% lower frequency than the secondary. Of course that may not happen... but I guess you should design in that much "leeway" for tuning purposes.
I read the HVWiki through a while ago but some things were a bit vague, thanks for clarifying.
Registered Member #1169
Joined: Wed Dec 12 2007, 09:16AM
Location: Portland OR
Posts: 251
Steve McConner wrote ...
Well then you have a problem, because the way that streamers load the coil is just that, unknown. And it's the streamer loading that determines the primary current.
Going by my own experience, and that of other DRSSTC builders I've corresponded with, the following rules of thumb seem to work out OK:
1. Build your secondary with the lowest resonant frequency that's practical (ie the finest wire gauge you can handle) and make it good and stubby for increased coupling. Toroid size should follow the guidelines given for spark-gap coils, if you want it to break out without a breakout point. Zo should be about 50k ohms.
2. Choose the current limit according to the size of coil, size of sparks you want, what size of IGBTs you can get and how hard you dare to overdrive them, etc. You can work back from the spark length you want, through Freau's spark length equation to get power, then divide by 120 to get bang energy, then calculate the required peak current from that assuming a 150 microsecond burst. (Longer bursts than this don't really help, nor do higher break rates, so you should use the 150 and 120 figures irrespective of what burst length and break rate you're actually going to run.)
3. Design the primary circuit to have the same resonant frequency as the secondary, and about 10 times the inverter's DC bus voltage dropped across the capacitor when a current equal to the current limit is flowing. From these two constraints you can calculate both L and C. Use the highest coupling possible without flashovers.
4. Put it all together. Adjust the primary tapping point until it works properly.
Note this is the "classic" method, it may even be in the HVWiki. Ward has some newer ideas which I think could work out better.
so the peak current through the primary is determined by the streamer loading correct? lets say i want a 12.5' streamer...
Power in = 240volts x 30amp = 7200watts <-- I am assuming power in for DRSSTC is still the wall power spark length = 1.77 x sqrt(Pin) = 12.5'
60hz = 120BPS
Bang energy = 7200watts / 120BPS = 60 Joules <-- required Bang Energy for 12.5 foot streamers
Bang energy of cap = Ec
Vp = SQRT( 2xEc / C )
Vp = Peak voltage seen @ MMC with a 12.5' streamer load correct? Vp = 10954.45 @ 1uf Vp = 12649.11 @ .75uf <--- Trying out different MMC sizes (voltage peak increases as cap size goes down) Vp = 14142.13 @ .6uf
R = Capacitance reactance of MMC correct? (calculated from online calculator) R = 3.979 @ 1uf R = 5.305 @ .75uf <-- Resistance goes down as Capacitance goes up R = 6.631 @ .6uf
Ip = Peak Current = Vp / R Ip = 2752 amps @1uf Ip = 2384 amps @.75uf <-- Peak Current goes down as cap size goes down Ip = 2132 amps @.6uf
so if I choose a .6uf MMC size it will see a peak voltage of about 14142.3 volts and peak current of 2132amps I am assuming the max amount of amps I can pump through the primary is limited by how much my IGBTs can take?
Also I want to design my filter caps with about 20x my original Bang Energy?
Registered Member #30
Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706
Well, that's not necessarily true. The whole point of the DRSSTC is that the bang energy can be greater than the amount of energy the tank capacitor can store. Most of the energy comes straight from your filter caps, and the tank cap is mostly just there as an impedance matching device.
That's why I suggest using the "10x your DC bus voltage" rule of thumb to size the capacitor, instead of the energy equation.
In this case your effective DC bus voltage is 677/2 because you have a half bridge, so you design your tank cap to take about 4kV.
Now if we pick a 2500 amp current limit, which is reasonable for CM600s, the tank capacitance is calculated by finding the reactance that drops 4kV at 2500 amps and 40kHz.
You then do a sanity check to make sure that the above setup can deliver 60 joules in 150 microseconds. Well, you can get away with 300us.
So the power delivered is roughly (677/2)*(4/pi)*2500, then divide by 2 to convert from peak to RMS, and another 2 to account for the ringup. The 4/pi is a conversion factor needed because the voltage is a square wave but the current is a sine wave.
Anyway that gives us about 270 kilowatts, which is 27 joules of bang energy per 100 microseconds of burst length. So in 300us, you'll deliver about 54J. Probably more, as the ringup will be complete before 300us.
This is just a rough calculation with a lot of assumptions made, but it shows you're in the ballpark.
PS: I managed to get a hold of Steve Ward, and he said he "didn't strongly disagree" with anything I said in this thread. Still, your mileage may vary
Registered Member #1169
Joined: Wed Dec 12 2007, 09:16AM
Location: Portland OR
Posts: 251
wrote ...
That's why I suggest using the "10x your DC bus voltage" rule of thumb to size the capacitor, instead of the energy equation.
So just to clarify... the capacitance reactance is not a good way to determine the peak voltage drop across the MMC tank. The reason for this is because at resonance the impedance from the L and C cancel each other out? instead pk voltage is coming from the Filter caps.
wrote ... In this case your effective DC bus voltage is 677/2 because you have a half bridge, so you design your tank cap to take about 4kV.
if I had a full bridge (which I will most likely do) the peak voltage rating of my MMC should be 6770VDC.
wrote ... Now if we pick a 2500 amp current limit, which is reasonable for CM600s, the tank capacitance is calculated by finding the reactance that drops 4kV at 2500 amps and 40kHz.
Using R = V/I = 4000/2500 R = 1.60 ohms?
And using XC = 1 / (2 * pi * f * C), where f=40khz and C=2.5uf XC = 1.592 ohms?
2.5uf seems too big...
wrote ... You then do a sanity check to make sure that the above setup can deliver 60 joules in 150 microseconds. Well, you can get away with 300us.
So the power delivered is roughly (677/2)*(4/pi)*2500, then divide by 2 to convert from peak to RMS, and another 2 to account for the ringup. The 4/pi is a conversion factor needed because the voltage is a square wave but the current is a sine wave.
Anyway that gives us about 270 kilowatts, which is 27 joules of bang energy per 100 microseconds of burst length. So in 300us, you'll deliver about 54J. Probably more, as the ringup will be complete before 300us.
This is just a rough calculation with a lot of assumptions made, but it shows you're in the ballpark.
PS: I managed to get a hold of Steve Ward, and he said he "didn't strongly disagree" with anything I said in this thread. Still, your mileage may vary
Going back to the previous Spark Length equation.... The Input power (Pi) is the power from the wall? Or from the Inverter Or from the primary MMC / Inductor. I just assumed 240volts RMS X 30amps = 7200 watts because that is how they would do it for a spark gap coil, but I have a feeling this is wrong.
I do appreciate your help on this Steve. I sent a couple of PMs to our vents here (who I could consider vets anyways). There seems to be quite a bit of confusion when designing MMCs.
From my E-mails with Steve Ward, it sounded like he was comfortable with the idea of setting the voltage peak of the MMC based on the capacitance reactance and current limit.
Registered Member #1900
Joined: Fri Jan 02 2009, 06:44PM
Location: Texas
Posts: 29
I think that both methods work here. The way I see it, Steve's method is more efficient for getting a good MMC size (big enough, but not oversized). Whereas calculating the voltage rating from capacitance/fo would prevent an issue arising from error in estimating the MMC rating to be 10x the bus voltage.
I'm not entirely convinced that 10x is enough... given that the coils I worked with this weekend had an MMC rating of 12kV (with a voltage doubled 208V bus = 400V).
Of course, like he says, these are unpredictable creatures. I recommend you don't worry about it TOO much more... and just get building.
This site is powered by e107, which is released under the GNU GPL License. All work on this site, except where otherwise noted, is licensed under a Creative Commons Attribution-ShareAlike 2.5 License. By submitting any information to this site, you agree that anything submitted will be so licensed. Please read our Disclaimer and Policies page for information on your rights and responsibilities regarding this site.
DRSSTC-2 Design Review (Updated Specs 8/28/09)
