High Voltage Forum
Tesla coils => Dual Resonant Solid State Tesla coils (DRSSTC) => Topic started by: TCWilliamson on July 05, 2017, 07:33:11 PM
-
Right'o!
First off, I've been loitering in tesla-coil related forums, soaking up information for... years. goodness knows how long I've been at 4HV...
Procrastinator's DRSSTC? Where did that come from?
Well.... I've been collecting parts and putting stuff off for as long as I've been aware of what a tesla coil was.
So, let's try doing something about that!
Last week I assembled 3 UD2.7C boards with components (untested)
... So, what have I got for this in terms of components?
- Driver boards assembled.
- A drawer full of 942C20P1K-F .1MFD 2000VDC Capacitors (like... 20-30 I'd imagine, possibly more)
- Spare components galore.
Power stage options:
- I have a good pile of CM300's not sure how many, they are buried under many boxes... 4 or so bricks maybe? Idk.
- 6pcs FGH40N60SMDF
- 6pcs FGH40N60SMD
- 6pcs FGH60N60UFD
- 5pcs FGH60N60SFD
- 6pcs FGH60N60SMD
(Crikey I was taking the piss out of samples back when I was young.)
I also have DC Bus Capacitors Galore.
Secondary / Primary coilforms:
Primary: 120mm x 200mm x 3-4mm acrylic transparent tube
Secondary: 110mmx 545mm x 3-4mm acrylic transparent tube
not sure what wire to wind it with however, I also have 1kg of epoxy for the secondary.
So... where to next?
-
You got the driver, you got capacitors for the MMC, primary and secondary coils are also pretty much decided upon.
You have a gap between the TO-247 IGBTs and CM300 bricks, I fear that the TO-247 is not powerful enough, my DRSSTC1 used 60N60 minibricks and they are really not powerful enough to drive a 160mm diameter secondary coil. So unless you source up some SKM75/150/200 or such, I think you just use your CM300 bricks, your coil will properly be around 100 kHz and that is no problem for CM300s.
Maybe something in the range of 0.25 mm diameter wire, 1800 turns and you should get in around 120 kHz with topload of same small diameter as coil diameter and topload large diameter as coil length.
For MMC capacitance, check this discussion out: https://highvoltageforum.net/index.php?topic=90.msg462#msg462
-
You got the driver, you got capacitors for the MMC, primary and secondary coils are also pretty much decided upon.
You have a gap between the TO-247 IGBTs and CM300 bricks, I fear that the TO-247 is not powerful enough, my DRSSTC1 used 60N60 minibricks and they are really not powerful enough to drive a 160mm diameter secondary coil. So unless you source up some SKM75/150/200 or such, I think you just use your CM300 bricks, your coil will properly be around 100 kHz and that is no problem for CM300s.
Maybe something in the range of 0.25 mm diameter wire, 1800 turns and you should get in around 120 kHz with topload of same small diameter as coil diameter and topload large diameter as coil length.
For MMC capacitance, check this discussion out: https://highvoltageforum.net/index.php?topic=90.msg462#msg462
Okay, I popped over to the local machine shop, he's making up some white plastic end caps for my coilform. I'm going to attach it to a motor and make a small jig and get winding!
What kind of prep should I do on the acrylic tube? Do I want to sand it down where I'm putting the wires, or leave it all nice and smooth?
Also, tad confused, the secondary is only 110mm diameter... does the rest of your statement still hold true regarding bridge?
-
I never made a secondary coil on acrylic tubing, but sanding it would properly help against it being too slippery, you could also give it a very thin layer of varnish just before where you wind, to glue it good into a sanded area.
Even for a 110 mm diameter secondary coil, I would take the CM300 over the TO-247 IGBTs, simply to make sure its not the bridge that will limit your spark output :)
-
Righto, Just been handheld through JavaTC by Mr Barnkob,
My current status:
(https://highvoltageforum.net/proxy.php?request=http%3A%2F%2Fi.imgur.com%2FhTjpGsD.png&hash=50d75b9796d3b6fb78f5cb15de4014b45e2d7e23)
So.... MMC, primary coil, distance to secondary.
*puppy-dog eyes for support*
-
I would assume that something in the range of 0.3 to 0.6 uF MMC would fit a coil of this size, but lets try to look at the numbers.
As from https://highvoltageforum.net/index.php?topic=90.msg462#msg462 we have the equation: Vmax/Imax < k / (2*pi*f*C)
If you are going to use CM300, I guess that 1000Apeak is more than enough for this small coil, properly way more than it can handle, so lets say 600A and assume that you use single phase 230VAC, simply because 3 phase for 564VDC, the higher voltage will not match the secondary
325VDC/600 = 0.15 / (2*pi*110khz*0.3uF)
0,69 < 0,72
This would be a high impedance circuit from the low capacitance, which can be seen from the low peak current, so this would require long on-times
If we really want to party and build a low impedance coil (which is generally more expensive)
325VDC/1500 = 0.15 / (2*pi*110khz*0.6uF)
0,27 < 0,36
But here you will properly run into more fun issues like heating secondary coil :)
For the choice of primary coil I would go with a flat spiral coil, start it 2-3cm away from the secondary, keep a coupling at 0.15 or up to 0.2 if you feel adventurous, depending on inductance needed you could vary the turn spacing, but be sure to leave 1-2 turns extra than the tuning point, so you can detune the primary circuit in order to run the coil in lower pole mode, explained very well here: https://highvoltageforum.net/index.php?topic=78.msg624#msg624
-
I would assume that something in the range of 0.3 to 0.6 uF MMC would fit a coil of this size, but lets try to look at the numbers.
As from https://highvoltageforum.net/index.php?topic=90.msg462#msg462 we have the equation: Vmax/Imax < k / (2*pi*f*C)
If you are going to use CM300, I guess that 1000Apeak is more than enough for this small coil, properly way more than it can handle, so lets say 600A and assume that you use single phase 230VAC, simply because 3 phase for 564VDC, the higher voltage will not match the secondary
325VDC/600 = 0.15 / (2*pi*110khz*0.3uF)
0,69 < 0,72
This would be a high impedance circuit from the low capacitance, which can be seen from the low peak current, so this would require long on-times
If we really want to party and build a low impedance coil (which is generally more expensive)
325VDC/1500 = 0.15 / (2*pi*110khz*0.6uF)
0,27 < 0,36
But here you will properly run into more fun issues like heating secondary coil :)
For the choice of primary coil I would go with a flat spiral coil, start it 2-3cm away from the secondary, keep a coupling at 0.15 or up to 0.2 if you feel adventurous, depending on inductance needed you could vary the turn spacing, but be sure to leave 1-2 turns extra than the tuning point, so you can detune the primary circuit in order to run the coil in lower pole mode, explained very well here: https://highvoltageforum.net/index.php?topic=78.msg624#msg624
Thanks for the information.
I figure I'll start off with 0.3-0.4uF and move on towards more in the future if I feel playful.
The capacitors I have are
942C20P1K-F, 2000V @ 0.1uF 288Apk, 12.1Arms
I'm not sure how many Volts I need for my primary capacitance, I've always been curious why we have one higher then the silicon really...
for 2000V @ 0.3uF and 864Apk, 36.3Arms
just 3 in parallel
for 2000V @ 0.6 uF and 1728Apk, 72.6Arms
just 6 in parallel
if I need higher voltage:
for 4000V @ 0.3uF and 1728Apk, 72.6Arms
2 series of 6 in parallel
for 4000V @ 0.6uF and 3456Apk, 145.2Arms
2 series of 12 in parallel
seems like a lot for 4kV, I hope 2kV will do?
Also, how should I be choosing the primary Inductance?
-
As you can see it quickly takes a lot of capacitors to get a higher capacitance, with higher capacitance comes lower primary inductance (to maintain the same resonant frequency), then peak current goes up, which results in higher voltage rating needed, as the voltage across the MMC is equal to the impedance Zc of the MMC multiplied with the peak current your controller is set to limit at. Add some head room for transients etc.
I used my own spiral coil calculator here: http://kaizerpowerelectronics.dk/calculators/spiral-coil-calculator/
As it looks like you have the capacitors for a 0.3 uF MMC, lets try to work out a solution for that. Here I went with a primary coil that can be ~10% detuned compared to the secondary resonant frequency.
[ Invalid Attachment ]
Using the MMC calculator with the capacitor ratings and the calculated primary inductance from above: http://kaizerpowerelectronics.dk/calculators/mmc-calculator/
[ Invalid Attachment ]
[ Invalid Attachment ]
4000V @ 0.3uF and 1728Apk, 72.6Arms, 2 series of 6 in parallel looks like a fine solution as you have some head room for the voltage rating, peak current is grossly overrated, but that is general problem when using CDE942 capacitors, as they have a relatively high peak current rating to their small capacitance, so many is needed in parallel.
If we want to be sure that we never have more than 3kV on the MMC, lets check what peak current is then the limit
Ipeak = VmaxMMCvoltage / (2* Pi * Lprimary * Fresonance)
583A = 3000 / (2 * pi * 8,18uH * 100kHz)
but due to the low primary inductance, we might hit 600A real fast,
Number of half-cycle to hit voltage limit of MMC = (V[MMClimit] / V[DCbus]) / 2
So maybe it is desirable to build a MMC that can withstand a higher voltage, bu then also start over on the calculations as the peak current will rise too with longer on-time, or just run with 600Apeak and longer on-time, but this will properly result in long damped energy transfers. Longer on-times results in higher losses, but its also cheaper to build, it is all one big pool of choices in regard to money, performance and reliability :)
-
To avoid hijacking the topic, the disucssion on inverter vs. secondary impedance have been moved to a separate topic: https://highvoltageforum.net/index.php?topic=113.msg657#msg657