12V DRSSTC dimensions / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

12V DRSSTC dimensions

Move Thread

LAN_403

DÃ©dÃ©!

Sat Jan 30 2016, 01:06PM

Registered Member #4932 Joined: Thu May 17 2012, 01:42PM
Location:
Posts: 59

I'm thinking about building a 12V DRSSTC with a fullbridge of P-channel and N-channel Mosfets and no GDT.
I've already selected the mosfets and the primary capacitors:
Primary capacitors
P-channel Mosfets
N-channel Mosfets

For the capacitor I'll use 10, maybe 15 in series and 2, maybe 3 in parallel for a total of 20 to 45 capacitors at 0.235ÂµF
For the mosfets I'll use 4 to 8 mosfets in parallel which gives a total of at least 16 mosfets and at most 32. At first I don't want currents to get too high. maybe 400A, max. 600A. I haven't yet selected the bus capacitor.
I think I'll drive each pair of N-channel and P-channel mosfets with their own dedicated driver with the outputs of the drivers at one side of the H-bridge connected to eachother using resistors and the total current to the drivers limited on the input with another resistor. Is this a good idea?

For the zero crossing detection for the intterupter and OCD I'll use this optocoupler

At first I'll power all of this using a large 12V power supply, but later on it should be powered by lots of Li-ion batteries.

I've got a few things figured out, but I can't figure out what dimensions I should use. I want to make a smaller coil, not taller than 1m, preferably even smaller and a resonance frequency of at least 40kHz, preferably more than 50kHz. I've tried a few dimensions in JavaTC, but can't find the right ones. Do you know of anyone who has done something similar to this before? If so, do you know what dimensions they used? or should I just go for a higher resonance frequency with a smaller primary capacitor?

Is there anything else I should know? Any tips and is it even possible?

Thanks in advance,
Dehim

DÃ©dÃ©!

Sun Jan 31 2016, 02:43PM

Registered Member #4932 Joined: Thu May 17 2012, 01:42PM
Location:
Posts: 59

I think these dimensions are alright:

I don't know if 3000 turns is a little too much or if k is maybe a little on the high side.
Ofcourse the topload isn't gonna be a giant sphere, but it's just to get an idea of the required capacitance. Around 29pF.

JavaTC doesn't take into account the skin effect does it? Anyways, I'm gonna use hollow copper tubing for the primary, so primary circuit resistance is closer to 30mOhm or something. I decided to only use 6 caps in parallel for the tank capacitor as 10 or 15 would have been a little bit overkill so the primary tank cap will have a capacitance of 0.282ÂµF.

Uspring

Mon Feb 01 2016, 02:03PM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

JavaTC seems to calculate AC resistance including skin effect only for secondaries. A primary DC resistance of 30mOhm or perhaps more adding skin effect is a big problem for a bus voltage of 12V. 30mOhm at e.g. 500A implies a voltage drop of 15V. You will never reach that current, since all of the input voltage will be eaten up by the resistance. An arc load on the secondary will show up as an extra resistance in the primary, reducing max current even more.

The extra induced resistance can be more or less freely chosen by the coils parameters, like primary and secondary inductance and coupling. If it is too high, you won't get enough current into your primary, i.e. little power. If it is too low, the efficiency of the coil will suffer, since most of the power will be consumed by the copper resistance of the primary.

The situation is similar to a power supply with an internal resistance. When you load it with a large resistor, there will be little current and little power in the load. If you load it stronger and stronger, the power in the load will increase for some time but then drop again, when the voltage drop along the internal resistance becomes too large. More than 200A in the primary in your case would not increase output power.

I couldn't read the data from your javaTC calculation. I've assumed a working frequency of 50kHz. If you run 200A at this frequency through your MMC of 0.28uF, there will be a voltage drop of 2300V, more than your caps can handle.

Rampup of primary current will be slow with small bus voltages. Omitting resistances, you will gain 24V (i.e. input peak to peak) for each half cycle. That amounts to 50 cycles or about 1ms rampup time for 200A. If your power supply can't handle that, you need big caps to buffer it. On the average there will be a current draw of 100A. For 1ms, this'd be 0.1C of charge from the caps. If you allow for 1V drop during a burst, the capacitance needed is 0.1F.

The biggest problem is really the primary resistance. Steve Ward has built battery powered coils and he always needed to step up the voltage before feeding it into the coil.

Perezx

Tue Feb 02 2016, 05:04AM

Registered Member #54263 Joined: Thu Jan 15 2015, 09:54AM
Location: Perth
Posts: 35

FullBridge for 12V? Sounds like not the best idea...
12V usually means E-klasse, sizes around 15-20cm height and frequencies around 1MHz (at least not less than 500-600 KHz) with power level at 200-300W.
If you want to use, for example, 360W, that means 30A DC input current. Imagine what should be at your FETs.
I have some experience with almost 1KW 12V SGTC... Was VERY hard to build.

DÃ©dÃ©!

Wed Feb 03 2016, 06:07PM

Registered Member #4932 Joined: Thu May 17 2012, 01:42PM
Location:
Posts: 59

What if I were to use a higher resonance frequency like 250 kHz and a larger primary capacitor like 0.423ÂµF made out of 18 or 27 strings in parallel of 2 or 3 capacitors in series.

These are the specs I've come up with with JavaTC:

J A V A T C version 13.2 - CONSOLIDATED OUTPUT
â€Ž3â€Ž-â€Ž2â€Ž-â€Ž2016â€Ž â€Ž18â€Ž:â€Ž59â€Ž:â€Ž39

Units = Centimeters
Ambient Temp = 20Â°C

------------------------------------------- ---------
Secondary Coil Inputs:
------------------------------------------ ----------
Current Profile = G.PROFILE_LOADED
2 = Radius 1
2 = Radius 2
0 = Height 1
18 = Height 2
2300 = Turns
0.007 = Wire Diameter

---------------------------------------- ------------
Primary Coil Inputs:
------------------------------------------ ----------
Round Primary Conductor
3 = Radius 1
3 = Radius 2
0 = Height 1
2 = Height 2
3.5 = Turns
0.2 = Wire Diameter
0 = Ribbon Width
0 = Ribbon Thickness
0.423 = Primary Cap (uF)
0 = Total Lead Length
0 = Lead Diameter

---------------------------------------- ------------
Top Load Inputs:
------------------------------------------ ----------
Toroid #1: minor=2.5, major=18, height=19.5, topload

----------------------------------------- -----------
Secondary Outputs:
----------------------------------------- -----------
280.81 kHz = Secondary Resonant Frequency
90 degÃ¯Â¿Â½ = Angle of Secondary
18 cm = Length of Winding
127.78 cm = Turns Per Unit
0.00826 mm = Space Between Turns (edge to edge)
289.03 m = Length of Wire
4.5:1 = H/D Aspect Ratio
1284.2449 Ohms = DC Resistance
74930 Ohms = Reactance at Resonance
0.01 kg = Weight of Wire
42.468 mH = Les-Effective Series Inductance
44.344 mH = Lee-Equivalent Energy Inductance
42.866 mH = Ldc-Low Frequency Inductance
7.564 pF = Ces-Effective Shunt Capacitance
7.244 pF = Cee-Equivalent Energy Capacitance
12.913 pF = Cdc-Low Frequency Capacitance
0.1401 mm = Skin Depth
5.979 pF = Topload Effective Capacitance
1544.7655 Ohms = Effective AC Resistance
49 = Q

----------------------------------------------- -----
Primary Outputs:
----------------------------------------- -----------
254.12 kHz = Primary Resonant Frequency
9.5 % high = Percent Detuned
90 degÃ¯Â¿Â½ = Angle of Primary
65.97 cm = Length of Wire
3.62 mOhms = DC Resistance
0.371 cm = Average spacing between turns (edge to edge)
0.897 cm = Proximity between coils
0 cm = Recommended minimum proximity between coils
0.927 Ã¯Â¿Â½H = Ldc-Low Frequency Inductance
0.34642 Ã¯Â¿Â½F = Cap size needed with Primary L (reference)
0 Ã¯Â¿Â½H = Lead Length Inductance
46.935 Ã¯Â¿Â½H = Lm-Mutual Inductance
0.235 k = Coupling Coefficient
0.122 k = Recommended Coupling Coefficient
4.26 = Number of half cycles for energy transfer at K
8.08 Ã¯Â¿Â½s = Time for total energy transfer (ideal quench time)

Maybe I could use a counter to turn it off after a predetermined amount of RF cycles so I can determine the optimal amount of RF cycles.

Is this idea maybe a little less farfetched or should I just drop the idea?

Perezx

Thu Feb 04 2016, 01:36AM

Registered Member #54263 Joined: Thu Jan 15 2015, 09:54AM
Location: Perth
Posts: 35

First of all, do not trust all these calculations so much. They are about 10-15% inaccurate just because they know nothing about surroundings of your TC

Uspring

Fri Feb 05 2016, 03:13PM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

Yes, a higher f wild lead to a faster rampup. The main problem are resistances in the primary circuit. A quick calculation yields a primary AC resistance of about 13mOhms for the coil only. I'd consider a thicker primary wire. You also need to include the resistance of the wiring to the primary and the FETs and also the ESR an ESL of the bus caps. May be manageable, but as Perezx said, hard to build.

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