CT current phase leading by 60°

[Karl Fischer]

Hello all,

After building a few slayer exciters it was time to try a DRSSTC. I quickly realised that the parts would cost way more when bought separately than the assembled PCB's, so I bought the driver, GDT and inverter of the shelf and build the rest myself.

I build the secondary and checked the Fres with my oscilloscope. Next, I built the primary and roughly tuned it to the same frequency. The PCB's do what they should do and the primary coil runs smoothly on it's own. So far so good.

When I put the primary and secondary together, the frequency jumps by 30%. Coupling the coils somehow causes the CT current phase to start leading by 60° and pushing the frequency up resulting in no ZCS, and low efficiency. Tuning has no effect at all.

What is the cause of the leading phase and what can I do about it?

[Mads Barnkob]

So you are saying that it was working fine with a dummy load? Like a metal bucket of water in place instead of the secondary coil?

Did you try reversing primary phasing?

How is your ground connection to the secondary coil?

It sounds like something else is going on, could you please upload some pictures of your setup.

[Karl Fischer]

Thank you for your reply Mads.

The primary did not have any load but was running on just 15 volts for testing purposes. Wouldn't a load not decrease the frequency in stead of increasing it?

Reversing the primary phasing resulted in MHz noise. Probably some feedback bypassing the LC circuit.

No ground connection. It's a bipolar with a cylindrical primary. I will upload some pictures of it ASAP.

[Mads Barnkob]

Thank you for your reply Mads.

The primary did not have any load but was running on just 15 volts for testing purposes. Wouldn't a load not decrease the frequency in stead of increasing it?

Reversing the primary phasing resulted in MHz noise. Probably some feedback bypassing the LC circuit.

No ground connection. It's a bipolar with a cylindrical primary. I will upload some pictures of it ASAP.

Low voltage testing can show some very special artifacts due to high output capacity of IGBTs, below 100VDC across C-E junction.

Reversing primary phase should result in either working or not working, as it would be driven 180 degrees out of phase.

I got no experience with bipolar coils, so I am not sure on how to advise on that

[davekni]

Pictures of your coil and electronics would help.  So would scope traces.  Traces of bridge output, primary current, and feedback input to driver.  Also, what driver are you using.  Is it UD2.7 or similar?  If not, posting schematics would be of great use in helping solve your issues.

[Karl Fischer]

I think the problem is coupling. When coupling is high the inductance of the primary is affected by the secondary. When coupling is low this effect is small enough to tune for but my coil is at least 0,50. The measured inductance of the primary decreased significantly when the secondary was placed. The primary needs to be larger to reduce the coupling. Back to the drawing board. Some pics:

(too) high k primary



Driver (don't know what version it is) + GDT + inverter, powered bij 30VDC PSU.



Driver schematics (interupter and sound circuit excluded)



Scope traces: sine wave is secondary coil voltage, square wave is TC4422 output probed at TP3. I expect (correct me if i am wrong) to see the driver to switch when secondary voltage peaks (+ or -) when tuned correctly.

[davekni]

Phase lead for DRSSTCs is generally lead of H-Bridge output voltage relative to primary current.  In order to measure that, you need either a second CT or a low-ohm resistor in series with the ground lead of your existing CT secondary (between CT and ground side of driver CT input connector).

Based on your image, secondary Fres should be much higher than the ~40kHz shown on your scope traces, likely over 1MHz.  Also looks like wires much too close to secondary coil ends and no mounting to allow both secondary ends to generate arcs.  I suggest studying DRSSTCs and especially bipolar coils more before building one.

(too) high k primary

High coupling is generally fine until it is high enough to generate racing sparks or primary-to-secondary arcs.

[Chrisader]

I don't see anything wrong. Current leads voltage, because You have not phase lead. ZCS lag is mostly similar and depends from electronic components used in construction. You need to do phase lead. If Fres will lower then angle of lag will be lower also. (time of delay electronic components is same). If Fres will be higher then angle of lag will higher too.
Have You tried to do phase lead?