Micro DRSSTC (EDITED) / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Micro DRSSTC (EDITED)

1 2 next

Move Thread

LAN_403

StormInABottle

Fri Dec 13 2013, 09:31AM

Registered Member #9252 Joined: Fri Jan 04 2013, 06:27AM
Location: Andromeda
Posts: 253

This was inspired by THIS

But because i do not like the idea of inserting a test tube down the tesla coil. My dimensions will be 4 inch wide / 4 inch long 34 AWG wire
Schematic? I chose the extremely simple circuit

And is the current transformer really connected in series with the secondary coil?:o

The bridge is this

But using FGH60N60SMD. and the GDT from the control circuit

Can you recommend me a simple interrupter that would require no fibre optic cable?

Dr. Dark Current

Fri Dec 13 2013, 04:38PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Nope, this circuit is really not correct from the topology point of view. It might work, but the efficiency will be poor.

StormInABottle

Fri Dec 13 2013, 07:12PM

Registered Member #9252 Joined: Fri Jan 04 2013, 06:27AM
Location: Andromeda
Posts: 253

The website says it efficiency is 90%.
And. I have seen something similar to a blog. Which showed him running that schematic but using a better MOSFET and inputting 170V. Aka rectified Us mains

How about this schematic? With the same mosfet mentioned above

Goodchild

Sat Dec 14 2013, 03:40PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

It looks like you have all your basic elements here for a class E, FET, feedback source, choke coil, etc.

The STW30NM50N is marginally better if you look in the datasheet at it's current handing and speed. Think about it this way assume that it was 90% efficient (i'm a little skeptical of that for a hobbyist circuit) That means that you are still going to losses 10% in heat in that FET, Thats 70W if you plan to run 700W. That's a lot of flippin power to loose in heat in a single tinny FET!

Now lets do the thermal math, if you look in your datasheet under Figure 3 you will find the thermal impedance curve. Assuming you plan to run CW at roughly 50% duty cycle you would look at that curve and grab a thermal impedance that is on the 50% duty curve at about the desired frequency. I'm seeing about 10 C/W. Now assume that your thermal pad/grease adds another 0.6 C/W and you have a heat sink that is another 5 C/W or so.

Sum up all the thermal impedances and you get about 15.6 C/W. Sanity check, if you look the datasheet also specs a DC value for die to PCB dissipation of about 30 C/W, so our numbers are reasonable for a good heatsink with airflow. Also at 50% duty, 70W can be expressed as about 35W, all though this is not technically correct it will do for this basic example.

Now we use ohms law, (ohms thermal law) the same way I=V/R -> q=dT/R where q=dissipated power dT = temp difference and R = thermal resistance. We know R and q we want to know dT. So MATHS!

35W=dT/15.6 C/W -> You will experience a nice toasty die temp of about 546*C

StormInABottle

Sat Dec 14 2013, 03:54PM

Registered Member #9252 Joined: Fri Jan 04 2013, 06:27AM
Location: Andromeda
Posts: 253

That's for the second circuit right?
546 C O.O.. That's literally a DIE temp.
Could you recommend me a mosfet?

Goodchild

Sat Dec 14 2013, 04:02PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Both are about the same. At that power level you more or less want to go with an interleaved topology that uses multiple FETs. Look in the datasheets for the FETs you pick and you can calculate your die temp the same way I did.

You are looking for a FET with fast switching and low thermal impedance etc.

Also don't use ether of control methods in those circuits, spend the time to design a dedicated driver with a real dedicated gate drive chip and feedback circuit. Part of a good class E circuit is a good control circuit that properly drives the gate. Do some research on class E switch mode PSUs.

StormInABottle

Sun Dec 15 2013, 08:16PM

Registered Member #9252 Joined: Fri Jan 04 2013, 06:27AM
Location: Andromeda
Posts: 253

After reading about Class E tesla coiling. It seems it is VERY easy to detune the 2 inch resonator. I would rather have a DRSSTC approach.

StormInABottle

Tue Dec 17 2013, 12:44PM

Registered Member #9252 Joined: Fri Jan 04 2013, 06:27AM
Location: Andromeda
Posts: 253

Found an interrupter in steve ward's website.
How did i not notice that..
I checked the wiki's article for DRSSTCs and the recommended MMCs are expensive.
Is there any alternative? The coil is really tiny..

Goodchild

Tue Dec 17 2013, 03:46PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

For a small coil an MMC should be relatively cheap. Your MMC must satisfy both the peak current rating and the RMS rating. These ratings must be followed for different reasons. The peak current rating must be followed to help partly with heating but mainly so that you don't poke holes in the dielectric.

Yes voltage is the culprit that actually puts the holes there; however the inductance of the capacitor will dictate the peak current. At a high enough level of peak current the self inductance of the cap will be enough to produce a voltage capable of exceeding the dielectric insulation. Be sure to follow the peak current specked for the caps.
The RMS current must be observed to prevent heating. In a DRSSTC it is typically very easy to meet this requirement with an MMC, because DRSSTC run very low duty cycles.

I would imagine for a coil the size you are thinking of it would not need more than a couple caps.

StormInABottle

Tue Dec 17 2013, 07:37PM

Registered Member #9252 Joined: Fri Jan 04 2013, 06:27AM
Location: Andromeda
Posts: 253

Â£10.80 is the cost for the 942C20P15K-F

This one is half the price with peak current of 211 amps
940C20P22KF
Can i buy two and put them in series.. And that's it?
Is the schematic good for my low power application of not asking for more than say. 12 inches of sparks?(Though more is better XD) ? I think the igbt is well suited for the job.
Can can you please explain to me how would i exactly use the RCA cable to connect the interrupter's signal

1 2 next

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