I need an HV fast sqaure wave generator. / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

I need an HV fast sqaure wave generator.

first previous 3 4 5 6 next

Move Thread

LAN_403

Pinky's Brain

Fri Oct 28 2011, 01:24AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Transformers just won't do ... flyback doesn't really work, and using normal step up transformer you run into the problem that any transformer with a fast rise time can't be used at 100 kHz (magnetizing current becomes too large).

There is simply no alternative to a half bridge to get a square wave with symmetric behaviour in on/off switching ... and the only ways to stack MOSFETs in that configuration are hard ways. You can use optical insulation, but then you need a floating power supply for MOSFET drivers. You can use HF transformers, but then you need separate transformers to turn the MOSFETs on and off (for the same reason you can't use a normal transformer to just step up the square wave).

Either way you need a break before make delay so you don't get shoot through.

BTW, I found a paper describing a half bridge HV square wave generator with a MOSFET stack ... they used optocouplers and drivers with floating power supplies.

"Improved highâ€voltage, highâ€frequency squareâ€wave generator"

Patrick

Fri Oct 28 2011, 01:30AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

ok ill look that up.

Patrick

Thu Nov 03 2011, 03:10AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Ok im gonna try to build this one:

]rsi03066.pdf[/file]

In the above pic i have no idea what im looking at? is the coax shield soldered to a ground plane or a 50pF cap?

this schematic seems to include everything important, as i assume the transformer is a gate drive/signal one. yet with the base tied to emitter through no resistor, it seems the transistor should not work?

are there smt resistors here? they dont appear on the schematic.

Im thinking this article isnt well written, they dont list the components or their values, and the schematic may not be accurate.

Mattski

Thu Nov 03 2011, 04:56AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

wrote ...
In the above pic i have no idea what im looking at? is the coax shield soldered to a ground plane or a 50pF cap?

The shield is soldered to the ground plane and the center conductor is soldered to the capacitor.

wrote ...
this schematic seems to include everything important, as i assume the transformer is a gate drive/signal one. yet with the base tied to emitter through no resistor, it seems the transistor should not work?

The current is probably limited on the primary side of the transformer. They mention in the paper that it's a pulse transformer, so probably a small capacitor discharging through the transformer, current is limited by the inductance rather than a resistor.

wrote ...
are there smt resistors here? they dont appear on the schematic.

They certainly look like 100 ohm resistors. I'd guess that they are there to ensure that the current flows out of the emitter instead of the base, though I'm not 100% sure why that matters.

wrote ...
Im thinking this article isnt well written, they dont list the components or their values, and the schematic may not be accurate.

It's a journal paper, not a popular media article or a how-to. Besides, if they told you all of their little tricks for actually getting a project to work then why would anybody continue to give them money after they published the paper?

Patrick

Thu Nov 03 2011, 05:10AM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Mattski wrote ...

It's a journal paper, not a popular media article or a how-to. Besides, if they told you all of their little tricks for actually getting a project to work then why would anybody continue to give them money after they published the paper?

I dont expect a how-to, from a physics journal, but it would be nice to have a well noted schematic... this looks more like high school or community college kind of work, maybe ill find a better source.

Pinky's Brain

Sat Nov 05 2011, 01:38PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

wrote ...

They rely on internal resistances to forward bias the junction during avalanche, which is why transistor selection is so finnicky I imagine.

R.J. Baker has a paper describing why using an external resistor is generally not a good idea :

PS. I still think a MOSFET half bridge stack with galvanically isolated gate drivers is the way to go :)

Patrick

Sat Nov 05 2011, 03:10PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

Pinky's Brain wrote ...

PS. I still think a MOSFET half bridge stack with galvanically isolated gate drivers is the way to go :)

Hey pinky, since ive already ordered and since this 2N5551 idea is so cheap. if youll help me figure it out, then maybe we can build and compare the MOSFET bridge and this avalanche thingy. My hv probe should be done this weekend.

Pinky's Brain

Sat Nov 05 2011, 08:13PM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

Dunno if being a negative nancy is helpful, but I'll sum up a couple of the problems you have to overcome ... with some solutions you probably won't like (because with the solutions the circuit stops being simple).

- Pulse generators are generally intended to put energy into a dissipative load in a short amount of time, not generate well defined signals for calibration.

You could put the probe in parallel with some avalanche diodes, that in series with a ballast resistor and all that in parallel with a HV capacitor ... to use as the load for the pulse generator. The avalanche diodes would regulate the voltage on the probe.

- To get to a couple kV with 2N5551 you probably need a Marx configuration, resistor based Marx generators tend to have rep rates much much lower than 100 kHz.

You can use common mode chokes instead of resistors for higher rep rates, but 100 kHz is still pushing it.

- If you want to generate a flat top pulse with pulse generators you have to get tricky ...

With a MOSFET based half bridge I can see the complexity, but I can also see that with enough engineering it's workable ... with avalanche transistors I still see a lot of complexity once you get beyond a simple low rep rate pulser and try to generate the square wave, but I'm not sure it's even workable ...

Patrick

Sat Nov 05 2011, 08:29PM

Registered Member #2431 Joined: Tue Oct 13 2009, 09:47PM
Location: Chico, CA. USA
Posts: 5639

well Hmmm.... is your earlier schematic still what you suggest?

Im willing to use some of these.

STW13NK100Z:

Look at the avalanche rating, not sure if this is useful or not.

teravolt

Sun Nov 06 2011, 03:31AM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

hi Patrick I have made a cuple of changes to your 2n5551 shematic what do you think? the pulse width is detirmined by the lenth of cable. I dont know if the grading resistors are necisary with the 2n5551 but I would use them. I have two pausibilities.

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