Pushing a CW Multiplier

For Short Term Learning Discussions ONLY. This area is for CURSORY questions and connecting with other users ONLY. ALL technical contributions need to be made in the appropriate forums and NOT HERE. All posts are temporary and will be deleted within weeks or months. You should have already search the extensive FAQs in each of the forums before posting here as your question may already be answered.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Pushing a CW Multiplier

Post by Matt_Gibson »

I recently got the complete CW multiplier assembly from a 100kV at 1mA Glassman power supply.

While I see that it is rated for 1mA, the components look to be capable of a lot more. The capacitors are 15kV 0.015uF and the diodes seem to be rated for 30kV and around 100mA (Varo H1601-30).

It’s only a half wave multiplier, so my guess is that the 1mA rating is to keep ripple very low. Has anyone pushed these things well beyond the rating? I’m thinking that this could wind up being a future 100kV at 10mA supply :-)
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Pushing a CW Multiplier

Post by Joe Gayo »

How many stages and what frequency do plan on driving it at? That will, along with the capacitance, determine the voltage drop across the multiplier (effectively it's impedance).
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

It’s 10 stages. I’ll probably drive it at 100khz since that is what I think glassman does.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Pushing a CW Multiplier

Post by Joe Gayo »

My guess is Glassman drove it at 50kHz or less if it's older than 10 years.

So the multiplier has 20 diodes and 20 caps?

---


Quick math says that @ 10 stages, 40kHz, 15nF, the voltage drop is 12kV @ 10mA ... so the multiplier capacitors would dissipate 120Watts ... Yikes.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

I may have to see how high I can push the frequency then…120w seems excessive…At 100khz, it’s around 50w I think.
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Pushing a CW Multiplier

Post by Joe Gayo »

100kHz on the diodes might be rough ... the recovery loss might be prohibitive. There's a trick, but I don't imagine you would be willing to stomach the cost. The first stage diodes need to have 0 recovery time, so that means SiC Schottky.

Some people count stages differently ... that's why I asked about the number of capacitors and diodes.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

It’s 20 diodes and 20 capacitors.

What’s the trick? I’m always up for some fun :-)
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Pushing a CW Multiplier

Post by Joe Gayo »

As mentioned: "The first stage diodes need to have 0 recovery time, so that means SiC Schottky."
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Pushing a CW Multiplier

Post by John Futter »

with lots of stages you usually beef up the caps in the lower stages ie 3 of the same caps per stage for the first 2 stages then 2caps per stage for the next 3 the the rest at one cap
the VM website used to have a brilliant excel spreadsheet you could use that allowed complex stages both half wave and full wave.
the more stages the lower the freq so 50kHz max for low output currents and 22-30kHz for grunty output current stages

otherwise I see a shelf of sacrifices to the gods of high voltage in the form of dead diodes and punctured caps
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Pushing a CW Multiplier

Post by Richard Hull »

The 60kv 60ma Hitek large stack and Flyback I have, according to factory specs, needed an input frequency of 25khz of flaming hot HV out of the giant flyback.... That sort of tells the tale as John and Joe discuss.
This setup was driven by a large rack mount system that drove the flyback with a large H bridge. Input to the box was 208 - 3 phase power. Even with efficiency of a switcher the rated 3600 watt output demanded a bit over 4000 watts input.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Pushing a CW Multiplier

Post by Joe Gayo »

Here's the math for Half-Wave...

Code: Select all


loadCurrent = 0.01;(*Amperes*)
voltageCap = 11200;(*Volts*)
stageCap = 15*^-9;(*Farads*)
stageMaxHalf = 10;
freq = 40000;(*Hz*)

capNumHalf[ePk_] := (2*ePk)/voltageCap
stageCapAdj[ePk_] := stageCap/capNumHalf[ePk]

(*single stage, 2 diode and 2 cap*)
voltageDropHalf[n_, ePk_] := (loadCurrent/(6*freq* stageCapAdj[ePk]))*(4 n^3 + 3 n^2 - n)
voltageOutHalf[n_, ePk_] := 2*n*ePk - voltageDropHalf[n, ePk]
voltageRippleHalf[n_, ePk_] := (loadCurrent/(2*freq*stageCapAdj[ePk]))*(n^2 + n)

capNumTotalHalf[n_, ePk_] := 2*n*capNumHalf[ePk]

GraphicsGrid[{{
   Plot[voltageDropHalf[x, voltageCap/2], {x, 1, stageMaxHalf}, Frame -> True],
   Plot[voltageRippleHalf[x, voltageCap/2], {x, 1, stageMaxHalf}, Frame -> True]}, {
   Plot[voltageOutHalf[x, voltageCap/2], {x, 1, stageMaxHalf}, Frame -> True],
   Plot[capNumTotalHalf[x, voltageCap/2], {x, 1, stageMaxHalf},  Frame -> True]
   }}, ImageSize -> {800, 500}]

Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

So how is Glassman achieving their low ripple (0.1%), at 1mA (the multiplier speced for 100kV at 1mA) without going to much higher frequency?

Playing around with the numbers and a half wave calculator:
10 stages
0.015uF capacitors
100kV at 1mA
100khz

I get 477v of ripple.

I know I’m missing something here, just don’t know what it is :-)

-Matt
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Pushing a CW Multiplier

Post by Joe Gayo »

You are ignoring the output capacitor that is integrated with the voltage divider for feedback.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

Here’s the schematic of the Hv section. I see some 2000pF capacitors. Are those contributing much to the regulation?
423FC916-0AB4-40FA-AF75-9DC09564C520.jpeg
User avatar
Joe Gayo
Posts: 404
Joined: Sun Jan 06, 2019 9:34 pm
Real name: Joe Gayo
Location: USA

Re: Pushing a CW Multiplier

Post by Joe Gayo »

In more ways than one ... but don't take my word for it ... figure it out and test
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

Once I get a good driver for my HF transformer, I’ll try it out, testing frequency up to 100khz to see how the diodes and capacitors heat…

Insulation for 100kV is probably the harder part of this anyways.

This will be a back burner project since I already have a 40kV at 15mA supply.
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Pushing a CW Multiplier

Post by John Futter »

Don't forget that glassman do the ripple spec with the standard output coax connected a fairly big smoothing capacitor
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

Hi John,

Where is this smoothing capacitor located? Externally? I don’t understand how they can give a ripple spec while using “extra” capacitance that the end user may or may not have…Either I’m just misunderstanding, or they are “fudging” things.
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Pushing a CW Multiplier

Post by John Futter »

It is the standard output cable ie coax 30pf per foot.
Now you work out the impedance at the operating freq at the top of the multiplier and shunt it with 300 pf of extra smoothing
ie not much ripple due to the high impedance and the cable capacitance --nothing said about regulation which is not that good but that is another whole lot of calculations
Rex Allers
Posts: 570
Joined: Sun Dec 30, 2012 3:39 am
Real name:
Location: San Jose CA

Re: Pushing a CW Multiplier

Post by Rex Allers »

Hey Matt,

I was looking at that ebay listing. I thought it ended before the expected date. Did you make a "buy it now" that was accepted? Just curious.

Interesting to follow the discussions about the electronic analysis of the multiplier circuit.

I'm wondering where you found the schematic you shared. I've followed a lot of Glassman stuff but don't think I saw that one before.

Do you know, was this multiplier, like most Glassman stuff, intended for air operation (not oil imersed)?

Unrelated, I recently made a bid on an EK supply (600 W out) and surprised myself when it was accepted. I haven't tried to test it yet but I opened it up and was surprised to find the multiplier was also half wave. I have a few Glassman supplies and the others all had full wave multipliers.
Rex Allers
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

Hi Rex,

I made an offer that was accepted. The seller had a link to a university website with the manual that this multiplier went in. It doesn’t look like it was designed for oil. I’ll have to take it apart so you guys can see the insides, they are pretty beefy!

My 40kV 15mA glassman was also a low bid surprise. I had to replace the mA scale, reverse the diodes and jumpers, but it works well other than that. 600W ought to be plenty, no?

I was sorry to see that 75kV negative Glassman go so high…

-Matt
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

Here are some pictures of my 100kV @ 1mA Glassman multiplier, opened up. Pretty beefy looking!
54B74646-0D3A-4B44-A2BF-DBA415567568.jpeg
29AA8D18-5BCB-4F1A-9E1F-9517A774691F.jpeg
C5FC63EF-0A07-4339-8B0A-1E77C71F44E2.jpeg
ED9454D8-E002-41D2-8CF1-5BBEFD4B2974.jpeg
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Pushing a CW Multiplier

Post by Dennis P Brown »

I am familiar with simple voltage multiplier circuits but don't recognize some of those components. It appears to be full wave but the purpose of the resisters isn't clear to me. Could you explain a bit about each picture and for others (besides me) a bit on the function of the entire system, maybe? What frequency does it need and current/voltage?
In any case, thanks for sharing the images.
Matt_Gibson
Posts: 505
Joined: Thu Sep 23, 2021 10:36 am
Real name: Matt Gibson

Re: Pushing a CW Multiplier

Post by Matt_Gibson »

While I am by no means an expert, I’m pretty sure it’s just a half wave multiplier…I wish it was full wave, more power!

I don’t know, for sure, what the series resistors (in series with the 0.015uF capacitors) are for, but would guess for current limit/protection.

The other resistors/capacitors are for voltage and current metering.

Don’t know the intended frequency either.

Once I have more time, and been successful with my 40kV supply, I’ll design/build a proper switching supply for this 100kV multiplier so that I can see how hard I can push it. I’m guessing that it is more capable than 1mA.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Pushing a CW Multiplier

Post by Dennis P Brown »

Thanks - guess the frequency matches most their std. commercial units.

I built a rather powerful 120 kV voltage multiplier (used very beefy door knob caps and high current diodes) and charged it a quarter way (I thought but either it was over charged or had a far higher upper limit - my calculations always errored on the under values.) When my 45 kV probe went full scale and I was still 5 cm from the unit - the field effects by the small discharge sphere (under oil!) convinced me it was far too dangerous to be near this unit. It is in an attic unused. Maybe someday (I tell myself) I'll have the nerve to try it again and with a better (safer) methodology to charge and test it.

So be careful at these very high voltages - as I learned (and can tell the tale) they get rather squirrelly and their behavior can be unpredictable.
Post Reply

Return to “New User Chat Area”