Can I run Steve Ward's Mini SSTC without an interrupter? / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Can I run Steve Ward's Mini SSTC without an interrupter?

first previous 3 4 5 6 next

Move Thread

LAN_403

Cheezey

Tue Jan 28 2014, 12:55AM

Registered Member #28483 Joined: Fri Jul 12 2013, 09:00PM
Location:
Posts: 29

Sigurthr wrote ...

Thanks.

If you driver checks out fine then there's no reason to rebuild it really.

That is a N30 core, I've used worse, but there are certainly better. I recommend T38 cores as most applicable core sizes are >10,000AL. This is the core I use and recommend:

it is good up past 1MHz.

OK, I'll make sure it works. Thanks for the core suggestion, I'll buy that instead.

Sigurthr

Tue Jan 28 2014, 01:43AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

Dr. Dark Current wrote ...

Matt,
The rating of diodes and electrolytic capacitors are two different things. An electrolytic cap will absorb a voltage spike, as well as it will handle temporary overvoltage. A diode will explode within 1us of overvoltage. Plus the rectifier is a much cheaper component than electrolytic capacitor, so overrating it is a good practice in quality electronics (not china).

The standard here is to use 600-800 volt rectifiers and 500-600 volt switching transistors for 230 VAC mains, and 1200 volt components for three phase (560 VDC). I have even seen 1700 volt three-phase rectifiers used even in smaller motor controllers.

I see, thanks for that info and perspective. I sometimes forget how much engineering down to a price goes on over here/there. I recently repaired a VCR/DVD player combo for someone; all that it required was a new bridge rectifier, filter cap, and switching FET. The PCB was discolored from the heat near those components. This is quite common, as I've had similar repairs for many common household items. My UPS monitors line voltage and datalogs it all, never seen any spikes myself, but something is blowing the mains components for others. I always assumed it was insufficient current handling capacity.

I think I'll revise the BOMs for my mains powered circuits and up the rectifier ratings.

Cheezey

Tue Jan 28 2014, 02:42AM

Registered Member #28483 Joined: Fri Jul 12 2013, 09:00PM
Location:
Posts: 29

Thank you for all your help. Before I place my order, I have one more thing to ask: do I need anything special for the .68uf bus caps? I looked more closely at some photos and it looks like there are bigger caps than mine.

Sigurthr

Tue Jan 28 2014, 04:11AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

The ones Ward uses are CDE snubber film/foil PP caps, iirc rated for 600V. You can use any value larger than that as far as capacitance goes (my 3kW tested half-bridge uses two 2200uF electrolytics), but for the voltage rating you want it to have some headroom over your maximum expected bus voltage. This ties in to the sub-discussion I was commenting in earlier. If you go with large electrolytics (better for output) for the bus caps you will want to put an additional snubber cap directly across the bus rails to suppress spikes. You can use the same 680nF 600V CDE PP film/foil cap for this purpose (I do!). If you can afford it get some 2200uF 450V electrolytics (two of them) for the bus caps. 450V will let you even run a doubler circuit as in a half-bridge the two bus caps are in series yielding double the voltage rating (but half the capacitance!).

When choosing electrolytic capacitors there are a few values to consider:
1) capacitance and tolerance. 20%tol is normal but make sure the listed capacitance -20% is still enough to get the job done.
2) voltage rating. >2x Vpk (more is better)
3) RMS ripple current rating - don't exceed it for long durations. (more is better)
4) ESR (less is better)

Cheezey

Tue Jan 28 2014, 10:05PM

Registered Member #28483 Joined: Fri Jul 12 2013, 09:00PM
Location:
Posts: 29

Sigurthr wrote ...

The ones Ward uses are CDE snubber film/foil PP caps, iirc rated for 600V. You can use any value larger than that as far as capacitance goes (my 3kW tested half-bridge uses two 2200uF electrolytics), but for the voltage rating you want it to have some headroom over your maximum expected bus voltage. This ties in to the sub-discussion I was commenting in earlier. If you go with large electrolytics (better for output) for the bus caps you will want to put an additional snubber cap directly across the bus rails to suppress spikes. You can use the same 680nF 600V CDE PP film/foil cap for this purpose (I do!). If you can afford it get some 2200uF 450V electrolytics (two of them) for the bus caps. 450V will let you even run a doubler circuit as in a half-bridge the two bus caps are in series yielding double the voltage rating (but half the capacitance!).

When choosing electrolytic capacitors there are a few values to consider:
1) capacitance and tolerance. 20%tol is normal but make sure the listed capacitance -20% is still enough to get the job done.
2) voltage rating. >2x Vpk (more is better)
3) RMS ripple current rating - don't exceed it for long durations. (more is better)
4) ESR (less is better)

I'm definitely not using electrolytics because I'm cheap. I found some snubbers that are 1.2KV, .68uf and have low inductance, low ESR, and high current ratings. I'll just buy those and hope they work, it seems like they should.

Thanks!

Cheezey

Tue Feb 11 2014, 01:10AM

Registered Member #28483 Joined: Fri Jul 12 2013, 09:00PM
Location:
Posts: 29

OK, I rebuilt everything except the secondary and the driver. Still tripping the breaker. The driver tested OK, it briefly gave me a nice square wave when I put my finger on the antenna (then my power supply failed. I fixed that.)

Sigurthr

Tue Feb 11 2014, 01:22AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

Pictures please. At this point we're going to need to actually see what is going on. Also, updated coil parameters would help greatly (number of turns, diameters, wire gauges, etc). Also, are you trying to run the coil on a mains circuit with a GFCI?

Cheezey

Wed Feb 12 2014, 01:44AM

Registered Member #28483 Joined: Fri Jul 12 2013, 09:00PM
Location:
Posts: 29

No GFCI.

Here's the JavaTC data, look at the inputs.

J A V A T C version 13.2 - CONSOLIDATED OUTPUT
Tuesday, February 11, 2014 6:07:56 PM

Units = Inches
Ambient Temp = 45Â°F

----------------------------------------------------
Surrounding Inputs:
----------------------------------------------------
5 = Ground Plane Radius
0 = Wall Radius
0 = Ceiling Height

----------------------------------------------------
Secondary Coil Inputs:
----------------------------------------------------
Current Profile = G.PROFILE_BARE
2 = Radius 1
2 = Radius 2
0.5 = Height 1
11.5 = Height 2
932 = Turns
30 = Wire Awg

----------------------------------------------------
Primary Coil Inputs:
----------------------------------------------------
Round Primary Conductor
2.75 = Radius 1
2.75 = Radius 2
1 = Height 1
2.4 = Height 2
11.5 = Turns
10 = Wire Awg
0 = Ribbon Width 
0 = Ribbon Thickness 
0 = Primary Cap (uF)
0 = Total Lead Length
0 = Lead Diameter

----------------------------------------------------
Top Load Inputs:
----------------------------------------------------

----------------------------------------------------
Secondary Outputs:
----------------------------------------------------
396.39 kHz = Secondary Resonant Frequency
90 degÂ° = Angle of Secondary
11 inch = Length of Winding
84.7 inch = Turns Per Unit
0.00178 inch = Space Between Turns (edge to edge)
976 ft = Length of Wire
2.75:1 = H/D Aspect Ratio
94.8662 Ohms = DC Resistance
48745 Ohms = Reactance at Resonance
0.3 lbs = Weight of Wire
19.572 mH = Les-Effective Series Inductance
25.565 mH = Lee-Equivalent Energy Inductance
27.369 mH = Ldc-Low Frequency Inductance
8.237 pF = Ces-Effective Shunt Capacitance
6.306 pF = Cee-Equivalent Energy Capacitance
23.47 pF = Cdc-Low Frequency Capacitance
4.76 mils = Skin Depth
0 pF = Topload Effective Capacitance
273.6159 Ohms = Effective AC Resistance
178 = Q

----------------------------------------------------
Primary Outputs:
----------------------------------------------------
0 kHz = Primary Resonant Frequency
100 % high = Percent Detuned
90 degÂ° = Angle of Primary
16.56 ft = Length of Wire
15.71 mOhms = DC Resistance
0.02 inch = Average spacing between turns (edge to edge)
0.694 inch = Proximity between coils
0 inch = Recommended minimum proximity between coils
26.296 ÂµH = Ldc-Low Frequency Inductance
0.00613 ÂµF = Cap size needed with Primary L (reference)
0 ÂµH = Lead Length Inductance
276.576 ÂµH = Lm-Mutual Inductance
0.326 k = Coupling Coefficient
0.129 k = Recommended Coupling Coefficient
0  = Number of half cycles for energy transfer at K
0 Âµs = Time for total energy transfer (ideal quench time)

I couldn't get attachments to work, so here's an imgur album:

Sigurthr

Wed Feb 12 2014, 04:27AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

26.3uH primary at 396KHz should only draw 1.3A in a half bridge with 120Vac mains input or 2.6A at 240Vac mains input.

At this point I'd check that your GDT secondaries are phased properly (one must be the opposite phase of the other) because if both secondaries are in phase the half bridge presents a direct DC short to the DC Bus. I'd also check the bridge rectifier and DC bus filter caps to make sure they haven't failed.

As far as construction goes, we've all seen high power coils run just fine with this style of protoboard and free wire construction, so I won't knock you for that, but I will say it looks damned dangerous. Especially with the frayed wiring at various connections. Take a DMM and make sure there are no shorts (or very low resistance readings!) between points that are not supposed to be direct connections, such as across the DC bus rails when no power is applied (be sure to unplug the whole thing).

I can't tell from the picture but make sure you know the pinout on MOSFETs is GDS left to right if you are holding the pins in your hand with the printed face facing so you can see it. It almost looks like the GDT leads are connected between Gate and Drain, and not Gate and Source as they should be.

EDIT: it just struck me! Did you use silpad insulators under the mosfets? The backs are connected to the Drains and the Heatsink will short them out without insulators.

Cheezey

Wed Feb 12 2014, 11:02PM

Registered Member #28483 Joined: Fri Jul 12 2013, 09:00PM
Location:
Posts: 29

It magically stopped tripping the breaker after I insulated all the close connections with hot glue, but now the power supply light isn't coming on.

I can't believe that's been the (main) problem this whole time.

I'll look into the power supply issue. Thanks again for all your help.

Edit: there's a short somewhere in my driver now... power supply works fine alone, but not with the driver (and the resistance between the +12v and GND is 2.8 ohms on my meter, which is just above what it reads when I touch the leads together)

first previous 3 4 5 6 next

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