Passing low freq. through CW bridge?

[Bob Jennings]

As part of my research project, I need to build a 5KV 10Hz signal
source. An audio step-up transformer and CW multiplier bridge seems
the most practical solution.

However, the bridge must pass the 10 Hz sinewave that will be AM
modulating the transformer drive signal of a few KHz.

I understand multipliers are not normally suited to AC out. But, given
the low frequency, is it possible to do this, perhaps by selection of
larger cap values? Any recommendations?

To fulfill the second design consideration, is it possible to
passively filter out the KHz carrier at some stage?

A possible solution may be to FM a carrier with the 10 Hz using a wide
frequency deviation, so the multiplier HV out follows the step-up
efficiency of the transformer.

Any insights on this application would be most appreciated.

Bob Jennings

 

[Winfield Hill]

Bob Jennings wrote...

As part of my research project, I need to build a 5KV 10Hz signal
source. An audio step-up transformer and CW multiplier bridge seems
the most practical solution.

How about getting your hands on a 10kV piezo speaker transformer?
5kV at 10Hz is relatively easy, there are several good ways to do it.
10kV at 600kHz is a bit harder, this I know from personal experience.

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)

 

[Tom Bruhns]

You didn't mention the power you need from that 10Hz signal. If it's
low power, it might be possible to do it as you suggest...but only if
you don't mind a large DC signal in addition to the AC, because your
multiplier will only have one polarity output. I suppose you would
use smaller, not larger, capacitors, so that the output would track
the input amplitude more quickly. You might try modelling it (e.g. in
Spice) to get an idea of what's going to happen. The large ratio
between the excitation frequency and the desired output frequency
should let you keep the excitation out of the output...an LC filter is
possible, but may not be practical because of the large inductance
required.

I've designed and built moderately high voltage audio-range DC-coupled
amplifiers by stacking multiple active devices, but 5kV (is that peak,
p-p or RMS?) wouldn't be a whole lot of fun with low-power solid state
devices.

Have you considered using a transformer directly? It may be feasible
if you have a low-impedance driving source and don't need much
power--an inexpensive audio power amplifier may do that for you.
Finding an appropriate transformer would likely be the tough part.

Another way to step voltage up to a high value when you are dealing
with a single frequency is to use a resonant tank circuit. If you can
come up with a 1000 henry inductor with a fairly high Q at 10Hz, that
will resonate with 0.25uF at 10Hz, and a 75mA circulating current
would give you 5kV. But that's going to be one big choke, to have
that much inductance and be able to store 10 or so joules. (Reminds
me of the modulation choke from a plate-modulated AM broadcast
transmitter.)

If you do try the FM trick, you can use resonance to narrow the
required FM range, though I suspect you'll have trouble getting that
to work properly in a reasonable time.

Cheers,
Tom


Bob Jennings <bjen2003@aol.com> wrote in message news:<g3nt80pf97pkblcuv4b4kj5ph2q4k1e9kg@4ax.com>...

As part of my research project, I need to build a 5KV 10Hz signal
source. An audio step-up transformer and CW multiplier bridge seems
the most practical solution.

However, the bridge must pass the 10 Hz sinewave that will be AM
modulating the transformer drive signal of a few KHz.

I understand multipliers are not normally suited to AC out. But, given
the low frequency, is it possible to do this, perhaps by selection of
larger cap values? Any recommendations?

To fulfill the second design consideration, is it possible to
passively filter out the KHz carrier at some stage?

A possible solution may be to FM a carrier with the 10 Hz using a wide
frequency deviation, so the multiplier HV out follows the step-up
efficiency of the transformer.

Any insights on this application would be most appreciated.

Bob Jennings

 

[Bill Sloman]

k7itm@aol.com (Tom Bruhns) wrote in message news:<3200347.0404272246.13dc7171@posting.google.com>...

<snip>

Finding an appropriate transformer would likely be the tough part.

Another way to step voltage up to a high value when you are dealing
with a single frequency is to use a resonant tank circuit. If you can
come up with a 1000 henry inductor with a fairly high Q at 10Hz, that
will resonate with 0.25uF at 10Hz, and a 75mA circulating current
would give you 5kV. But that's going to be one big choke, to have
that much inductance and be able to store 10 or so joules. (Reminds
me of the modulation choke from a plate-modulated AM broadcast
transmitter.)

Hmm. I wonder if a big choke wound with high temperature
super-conductor wire might do the trick? People were making noises
about power-sized components a few years ago but I've not seen
anything recently.

------
Bill Sloman, Nijmegen

 

[Winfield Hill]

Tom Bruhns wrote...

I've designed and built moderately high voltage audio-range DC-coupled
amplifiers by stacking multiple active devices, but 5kV (is that peak,
p-p or RMS?) wouldn't be a whole lot of fun with low-power solid state
devices.

Or Bob Jennings could simply buy a Trek amplifier,
http://www.trekinc.com/Amplifier_Selection_Chart.htm
e.g., a 615, http://www.trekinc.com/615/615_series.htm
like this one, http://www.trekinc.com/615/615aspec.htm

Thanks,
- Win

whill_at_picovolt-dot-com (use hill_at_rowland-dot-org for now)