DRSSTC Tank Circuit Current / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

DRSSTC Tank Circuit Current

1 2 next

Move Thread

LAN_403

UbuntuNinja

Tue Feb 09 2010, 05:40AM

Registered Member #2677 Joined: Mon Feb 08 2010, 03:06AM
Location: Palo Alto, California, USA
Posts: 64

Can anyone tell me how to calculate the maximum current within the tank circuit (to be multiplied by impedance to find voltage rating needed for tank capacitors) for a coil I am designing with the following specs:
Primary Inductance: 5.1774974 microhenries Primary Tank capacitor: 50.03240031 nanofarads Primary Self Capacitance: 15.4661 picofarads Supply power to bridge: ~340V@5A Secondary Inductance: 10.98424351 millihenries Secondary Total capacitance: 23.5904 picofarads System resonant frequency: 312.6564576 kHz Interrupter Burst Length: 9.5 microseconds Interrupter Bursts per second: 2000
My MOSFETS are rated 13 Amps at 500 Volts and I would like to limit the current to around 12A. Could this be done by changing the interrupter burst lenth/frequency? If not I could do away with the top load I am making from dryer duct and then the Primary Capacitor would be only 14.71308059 nF and the system resonant frequency would be 576.3419 kHz. I am aware that the decimal values are strange and could not easily be achieved but I included them just for the sake of accuracy. For instance, if I do not use a topload I will probably make the primary capacitor 15nF. I appreciate the help :).

Hazmatt_(The Underdog)

Tue Feb 09 2010, 06:16AM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

I think you forgot to consider your coupling value
K= 1.41444329333499558320110234

You have all of the important numbers, so you could either simulate it, or look around the forum more and do your homework.

By the way, you'll never have that kind of accuracy with your numbers. When you get down to that kind of resolution, temperature actually plays a big role and you can't get much more accurate then 2 sig figs.

TheBoozer

Tue Feb 09 2010, 10:13AM

Registered Member #1535 Joined: Wed Jun 11 2008, 11:37PM
Location: Northeastern Pennsylvania - USA
Posts: 117

The answer to that question doesn't solely rely on the numbers you've provided...

A very lossy design will only allow the tank to ring up so high due to resistive losses...

Personally, I built 2 identical DRSSTC's with the only difference being the capacitor assemblies. One rings up twice as fast as the other due to higher current carrying capabilities... For these systems, I chose a value for the over-current-detector that I am comfortable with. These systems were 500a. Punching that in to a calculator with my known impedance gave me my answer to choosing a capacitor voltage rating...

Adjusting your BPS will allow you to control how much you are pulling from the wall.....

Good luck.

Mads Barnkob

Tue Feb 09 2010, 02:54PM

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

There is the easy way suggested by Steve Conner elsewhere in another DRSSTC thread, multiply your bus voltage with 10. F.ex. 320V * 10 = 3200V

Another is your primary Apeak * primary Zsurge. F.ex. 700A * 5,84 Ohm = 4091,37V

You have to be sure that your MMC can withstand the RMS current too, I think its (SQRT(ONtime/OFFtime))*0,707*Primary Apeak^2. F.ex. A Cornell-dublier 942C20P15K-F capacitor is rated 13,5A at 100 kHz, so 6 parallel strings will give us a rating of 81 Arms.
My coil will at 700Apeak have a RMS current: SQRT((100uS/8233uS)*0,707*(700A^2)) = 64,87A.

I hope this helps and is without too many errors :)

In the end... its a guideline, we are already pushing / punishing the components, as always its a tradeoff between price and accepted component temperature rise, it might be cheaper to add some fans than f.ex. another 12 capacitors.

Angstrom

Tue Feb 09 2010, 05:50PM

Registered Member #1900 Joined: Fri Jan 02 2009, 06:44PM
Location: Texas
Posts: 29

Mads,
Are you sure about that RMS current calculation? I think if you leave off the square of current it works out.

I used the VDC * 10 method as well and made sure it matched up with the current limit set by OCD. Although I am using almost double the rating given by Apeak * Zmmc. The more rating the better but it can get expensive after 6kV or so.

UbuntuNinja

Wed Feb 10 2010, 05:41AM

Registered Member #2677 Joined: Mon Feb 08 2010, 03:06AM
Location: Palo Alto, California, USA
Posts: 64

Coupling is k=0.1849. I have tired using ScanTesla but it does not generate an output. The output.txt file says that none of the models were tested. I attached the input.txt i am using. Perhaps someone could point out if I made a mistake entering a value somewhere? My problem is that I don't know how to calculate the primary peak current OR the RMS current. I was using the equation Apeak=VoltsIn*SQRT(Capacitance/Inductance) but the problem with the is that when you multiply it by impedance, Z=SQRT(Inductance/Capacitance), the stuff inside the SQRT multiply to 1 leaving you with volts in, right where you started.
]input.txt[/file]

Wolfram

Wed Feb 10 2010, 07:21AM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Mads Barnkob wrote ...

There is the easy way suggested by Steve Conner elsewhere in another DRSSTC thread, multiply your bus voltage with 10. F.ex. 320V * 10 = 3200V

This will only give you the right number if you've chosen your surge impedance so that the voltage across the resonant capacitor is ten times the DC bus voltage at resonance.

WRONG, to be corrected!
You have to be sure that your MMC can withstand the RMS current too, I think its (SQRT(ONtime/OFFtime))*0,707*Primary Apeak^2. F.ex. A Cornell-dublier 942C20P15K-F capacitor is rated 13,5A at 100 kHz, so 6 parallel strings will give us a rating of 81 Arms.
My coil will at 700Apeak have a RMS current: (SQRT(100uS/8233uS))*0,707*700A^2 = 64,87A.

I hope this helps and is without too many errors :)

Here's the right formula

Anders M.

Luca

Wed Feb 10 2010, 08:58AM

Registered Member #2481 Joined: Mon Nov 23 2009, 03:07PM
Location: ITALY
Posts: 134

Anders M. wrote ...

Here's the right formula

Anders M.

I think that this formula is a very conservative estimation of the actual rms current. I performed various simulations both with PSpice and with ScanTesla and I have always found that the actual rms primary current is quite lower than that predicted by the formula.

Regards,

Luca

Wolfram

Wed Feb 10 2010, 11:10AM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Yes, my experience has been the same. That formula gives me 200A RMS where spice says 140A RMS for one of my simulations.

Anders M.

Mads Barnkob

Wed Feb 10 2010, 02:13PM

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

Anders M. wrote ...

Yes, my experience has been the same. That formula gives me 200A RMS where spice says 140A RMS for one of my simulations.

Anders M.

Thanks for the correction, must have missed a couple of ( ) when I wrote the equation from the spreadsheet

Following the experience of bigger is better when talking about MMC capacitance, its usually not a problem covering the RMS current anyway.

1 2 next

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