Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: Resonant choke HV PSU questions  (Read 7069 times)
2E0ILY
Member

Posts: 131




Ignore
« on: December 18, 2012, 06:08:14 AM »

I am building an HV power supply. The last one I built used a
capacitive input filter, and works fine. I want to build a smaller
one, and have a suitable HV transformer that gives 2650 volts at about
an amp. I have acquired a 4 Henry 2 ampere choke (conventional, not a
"swinging" choke), from a Hirschmann 3500V DC supply. I am wondering
if I might build this supply using a resonant choke input filter. I
know a little about this, such as a choke type allows more current
from the transformer for less energy into the primary, but gives less
voltage out. I believe it needs to be resonated and someone else has
done some maths suggesting a resonating cap of .68uF may be right for
100Hz. My mains is 50Hz, being in the UK. I am wondering if I could
use a VNWA to prove or disprove this before buying fairly expensive HV
caps? Choke type input filters usually use pretty low value bleeder
resistors to keep a fixed minimum load on the supply, do I need to
have these bleeder resistors across the set up whilst testing? I am
hoping by using a VWNA I can check for resonance with (cheap) low
voltage caps?


Thanks, and a very happy Christmas to all.
Logged

Best regards, Chris Wilson.
AA4PB
Member

Posts: 12843




Ignore
« Reply #1 on: December 18, 2012, 06:35:22 AM »

There is some info here: http://www.qsl.net/i0jx/supply.html
Logged
KA4POL
Member

Posts: 1979




Ignore
« Reply #2 on: December 18, 2012, 07:25:18 AM »

VNWA sounds like a good idea. If you got one that does the 50 Hz you should be fine. However, I'd suggest you use some audio software for the measurements.
Logged
G3RZP
Member

Posts: 4585




Ignore
« Reply #3 on: December 18, 2012, 09:02:49 AM »

First point is why a resonant filter? The answer is that a choke input needs an inductance of a critical  minimum value. Because of saturation in the core of a choke with no air gaps, the choke inductance decreases with current: this is a swinging choke. The critical inductance also decreases as the current goes up. So by tuning the choke to JUST above the ripple frequency, it looks like a much bigger inductance. If you hit resonance, you get some VERY big voltages, and it would look resistive.

So you need to know the actual inductance of the choke with the minimum amount of current that will be drawn. Then you can calculate the critical inductance. For 50 Hz, and a series resistance of the choke small compared to the load resistance, critical inductance is the load resistance/940.

So assuming the valve is biased off in receive, and the bleeder load is 50mA with a voltage of 2500 volts, the choke needs to be 50,000/940 or just over 50 Henries.

The procedure then is to figure what the reactance of 50 henries is at 100Hz, what the reactance of your choke (real measured value of inductance with 50mA flowing) and from that, what shunt capacitance you need to make the reactance of the parallel choke/capacitance combination equal to the reactance of the critical inductance.

Actually resonating the choke can be very nasty, as Tom, W8JI, has graphically told us.

Remember, the smaller the load current, the bigger the inductive reactance needs to be. Even then, you also need a reasonable amount of output capacitance to stop the volts swinging too much with transient variations in load current.

There are also advantages in negative lead filtering in that the insulation requirements on the choke are decreased. The tuning capacitor needs to be a high voltage type capable of carrying a fair amount of AC, too.

Changing the tuning capacitor to get the performance you want works as an empirical method, but it means you need a high power load for the PSU.

At lower power levels, negative lead filtering and rectifying the volts across the choke to get negative bias works well, too. But because that is effectively a shunt resistance, you need to be careful of the effects as it will, if taken to extreme, just force the PSU to be a badly regulated capacitor input one.
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #4 on: December 18, 2012, 09:22:16 AM »

Thanks for the detailed reply Peter, so are you basically saying stick with a capacitive input filter, as resonant choke ones are fickle and not worth the headache? Smiley I have no need or desire to overcomplicate things, but would have given it a shot if the likelihood of a working supply without too much hassle was strong. Thanks for your time, much appreciated, all the best for Christmas and the New Year.
Logged

Best regards, Chris Wilson.
G3RZP
Member

Posts: 4585




Ignore
« Reply #5 on: December 19, 2012, 01:21:12 AM »

I use a tuned choke input, but you have to appreciate the possible problems. 4 Henry is rather small, and I don't use less than 10 Henry. Plus the tuning caps I use are BIG - metal cased ones, 0.1 mFd at 5kV, from a WW2 radar, rated for pulse applications.

So unless you can get a bigger choke and suitable tuning caps and the ability to measure the choke inductance with DC flowing, yes, stick to capacitor input. Use step start to limit inrush current.
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #6 on: December 19, 2012, 04:06:54 AM »

I use a tuned choke input, but you have to appreciate the possible problems. 4 Henry is rather small, and I don't use less than 10 Henry. Plus the tuning caps I use are BIG - metal cased ones, 0.1 mFd at 5kV, from a WW2 radar, rated for pulse applications.

So unless you can get a bigger choke and suitable tuning caps and the ability to measure the choke inductance with DC flowing, yes, stick to capacitor input. Use step start to limit inrush current.

OK, this sounds like it's a job for the more experienced builder. But I would like to know how you should test a high voltage choke to get a handle on its inductance, with DC flowing. I am curious as to how many Henrys the, err, Henry choke is, that is shown as the same part number in both the 3 phase and single phase supplies. It's a big beggar, and extremely heavy. The schematics do not give a value for its inductance though. Thanks you Peter. Ohh, and by the way, my AIM4170 won't go below 0.5MHz, so I can't use that to check inductance around 100Hz. I am curious as to how these things are assessed in real time.
« Last Edit: December 19, 2012, 04:10:04 AM by BASIL » Logged

Best regards, Chris Wilson.
G3RZP
Member

Posts: 4585




Ignore
« Reply #7 on: December 19, 2012, 07:56:56 AM »

There are various bridges which allow you to superimpose DC on the choke while measuring the inductance. The Owen or the Hay bridges are examples, where you can use a transformer to apply AC to the bridge, with a DC bias applied from a power supply through the transformer secondary. The amount of DC then depends on the voltage and the series resistance in that arm of the bridge.

When you use a polyphase rectifier, you get a smaller value of critical inductance, which is why the Henry choke should be smaller for 3 phase. It will also be smaller for 60 Hz, too. Also, didn't that come from a RF heater? If so, they may well have switched the HV off in the standby case where there wouldn't be any load: unless in an amplifier, you want to switch the HV off  in receive, you have to make sure that the inductance stays above the critical value. Otherwise, when you go transmit, there's a sudden drop in the HV that can send the choke into ringing.
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #8 on: December 19, 2012, 09:00:38 AM »

There are various bridges which allow you to superimpose DC on the choke while measuring the inductance. The Owen or the Hay bridges are examples, where you can use a transformer to apply AC to the bridge, with a DC bias applied from a power supply through the transformer secondary. The amount of DC then depends on the voltage and the series resistance in that arm of the bridge.

OK, thanks peter, I'll do some reaearch now I know what to look for.

When you use a polyphase rectifier, you get a smaller value of critical inductance, which is why the Henry choke should be smaller for 3 phase. It will also be smaller for 60 Hz, too. Also, didn't that come from a RF heater? If so, they may well have switched the HV off in the standby case where there wouldn't be any load: unless in an amplifier, you want to switch the HV off  in receive, you have to make sure that the inductance stays above the critical value. Otherwise, when you go transmit, there's a sudden drop in the HV that can send the choke into ringing.

The Henry schematics show the same part number choke for both single and three phase supplies. The only difference is the single phase supply has a pair of parallel 0.1uF caps across the choke terminals. The unit is a Henry 2000D RF Generator, rated for continuous duty at 2kW, on either supply. It was supplied new, direct from Henry Electronics in the US, to Aberystwyth University, here in the UK, for plasma generation, and was specced for our 50Hz supply, although the choke is the same part number as that for the US 60Hz supply.... The three phase HV transformer is a different part number to the US ones though, so presumably wound with 50Hz in mind. Both single and 3 phase supplies use the same bleeder resistor values (20k, 100W), and the same 4uF smoothing cap.

 I just built a capacitive input filter supply using a single phase transformer, from a different source, and an electrolytic array, but again rated for continuous duty. It uses a 3CX3000A7 valve. I'll stick with capacitive filtering on the next supply, too, although this is all very interesting. Thanks again, much appreciated. Schematics link for the two Henry supplies, single and 3 phase, below.

http://www.gatesgarth.com/resonance.jpg
Logged

Best regards, Chris Wilson.
G3RZP
Member

Posts: 4585




Ignore
« Reply #9 on: December 19, 2012, 12:08:28 PM »

That suggests that there is enough inductance to be above critical on a 3 phase supply but not enough on single phase, so requiring some tuning on a single phase supply.
Logged
N2EY
Member

Posts: 3879




Ignore
« Reply #10 on: December 19, 2012, 05:37:19 PM »

That suggests that there is enough inductance to be above critical on a 3 phase supply but not enough on single phase, so requiring some tuning on a single phase supply.

That's because the ripple frequency is higher with a 3 phase supply.

With a single-phase full-wave rectifier, the ripple frequency is twice the supply frequency.

With a three-phase full-wave rectifier, the ripple frequency is six times the supply frequency. Much less inductance needed.

73 de Jim, N2EY
Logged
G3RZP
Member

Posts: 4585




Ignore
« Reply #11 on: December 20, 2012, 02:47:46 AM »

Jim,

Not only is the ripple frequency higher from a polyphase rectifier, but the ripple amplitude is less, too.

73

Peter G3RZP
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #12 on: December 20, 2012, 03:17:24 AM »

I would like 3 phase connected at home here, we had it at the old house, but now we are out in the sticks I have to use a phase converter for the car ramp, milling machine, surface grinder and lathe etcetera. Having a 3 phase supply connected here was going to be hugely expensive. TA phase converter is a poor substitute, but cheaper than converting everything to single phse motors and control gear. Thanks for the replies, learning a lot!
Logged

Best regards, Chris Wilson.
G3RZP
Member

Posts: 4585




Ignore
« Reply #13 on: December 20, 2012, 03:46:35 AM »

What sort of phase converter are you using? I have a variable frequency inverter for giving me 3 phase variable speed for the lathe and mill, but it sounds as if you need a lot more power than that. I wonder if there are any cheap multi kW motor generators around these days?
Logged
2E0ILY
Member

Posts: 131




Ignore
« Reply #14 on: December 20, 2012, 04:51:18 AM »

I have had this rotary converter for as long as we have been in Shropshire, so about 12 or so years. I can't recall the maker, but I do remember I picked it up from them, and they had a huge range of sizes. I'll see if the maker's plate is visible, but it's buried under a lot of stuff in a corner. I think it's a 10HP motor in it. 3 phase motors run on it are a tiny bit "growly", but it has worked perfectly, even on things like a two post 3.5 tonne vehicle ramp. These are notorious for very high start currents. It was at the expensive end of the market, I remember I did try the one they first suggested, exactly half the size of the one I ended up with, and took it back the next day, it just wasn't man enough, nothing like man enough. One tip I have heard subsequently is if someone says you need "X" size, buy one twice as big. I have a friend who needed something similar, but decided to go solid state. His ramp goes up and down, but not with a vehicle on it, as it won't kick the loaded motor into life.... I have found rotary converters far better for things with a big start up draw. I  am considering a diesel generator actually. Mainly for bigger workshop equipment, but would be interested how something like a 3 phase  HV PSU would fare, driven off one.
Logged

Best regards, Chris Wilson.
Pages: [1] 2 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!