Developing HV DC Pulses

Seems I need to drop the idea of grounding one side of the secondary. I'll be purchasing plexiglass and solvent to make a physical obstacle to prevent getting fingers where they shouldn't be. Thanks, Mikek

I had placed a bid on a B&K 3011B and did not get it. I can still get a 3011B, if that is satisfactory. Is there a Function Generator anyone can recommend that I can use to Drive the tube? I'll be buying used, probably Ebay to keep the price under $100. I'll probably lend it for the experiments and then keep it for me.

Mikek

Dmitry's latest Push-Pull Tube Pulse circuit.

Thanks, Mikek

Ground the transformer case and drive the sample cell differentially.

That doesn't make much sense either.

Why do some people insist on putting resistors, and even bypass caps, across voltage sources in Spice?

Is that in the subckt, but not shown in the schematic? I don't see extra resistors or caps in the schematic. Mikek

Sure, bipolar grid drive is easy and ideal in Spice. It's more work in real life. Commiting to +-5 may be risky; see the 6BK4 data sheet.

What are R4 and R5 for?

His comments are wrong. For some reason, people get sloppy when they use Spice. I treat a Spice design like any other engineering document.

R4. R5. In plain sight.

That way the LTSpice circuit diagram can look exactly the same as the one that goes into your documentation.

It's a waste of time from any other point of view, but designing circuits, and getting them built and documented so they come out exactly as intended can take quite a bit of effort.

And you can put a series resistance into your voltage source. The default seems to be one milliohm.

On 5/16/2023 9:01 PM, Lamont Cranston wrote: <snip>

Can you put the aluminum tubes inside PVC pipe? Or otherwise make it impossible to touch them while power is on? For example, the whole mechanism is inside a box with an interlock, such that when the box is open power *120VAC) is automatically removed?

Ed

All this is crazy. A neon transformer is not designed to work even remotely this way. Something bad will happen, and the investigation will notice that this very much violates the National Electrical Code. And make sure that your affairs are in order.

Joe Gwinn

I’m a little disappointed in the way the voltage divider is working. Using 1.5MΩ, 2 Watt 5% resistors, 8 in a string. Setting the input at 8.04kV, we get 7.02Kv, 6.06kV, 5.16kV, 4.28kV, 3.45kV, 2.7kV, and 2.08kV. When I expected 8KV, 7kV, 6kV, 5kV, 4kV, 3kV, 2kV and 1kV. Kind of messed up at the bottom end. Here’s what my voltage divider string looks like on the banana jacks. Do the think this is caused by leakage across the resistors? I wiped them down with alcohol and it didn’t get any better. Using 100MΩ probe.

Puzzling. Have you measured all eight are the same value?

What is the max voltage rating of those resistors, even 2W parts can have max rating only 500 or 750V - you could try doubling with two identical in series between each banana jack?

piglet

I tried a couple of things this morning, I applied the Kapton tape that was bought for this purpose. That seemed to have a few 10s of volt difference at the low end. under 3%. But then I noticed where I'm standing and how the meter leads are dressed makes a difference in the Voltage reading. I wound the long HV probe cable around the probe and rerouted the low end meter lead. That improved the last two low end measurement from 2700V to 2130, the expected is 2000V and 2080 down to 1280V, expected 1000V. So, where getting closer to what I expect. My son is taking the unit back to work to do some experiments today, I will suggest that he shorten the meter leads to the minimum required to take the measurements. Thanks, Mikek

Those don't look like 2W resistors, but it's hard to tell.

The observation that the tap voltages depend on the position of nearby conductors is pretty diagnostic of a corona problem.

Cheers

Phil Hobbs (Who is doubtful of this whole proceeding)

Quoting error. I did not say the above, or anything similar.

Joe Gwinn

Is this setup AC or DC, if AC then I wonder if stray capacative coupling is your problem? Either that or corona?

piglet