• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

choke input power supply

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Ive been pondering changing one of my amps from gz34 clc to ss choke input for some time. The change to ss was because its finally going to get a case and I don't have room for a valve.

I bought a 300-0-300 hammond, intending to use it end to end into a bridge and ignore the ct. So I figured on using the existing 10h choke and a large cap, 220 or 470uf. Then I was reminded about free lunches, no matter what I did with psud it showed 500-800mv of ripple. Then I tried with 2 chokes and bingo, down to 1 mv according to psud....sounds good, but I ditched the valve to save space, only to end up having to find space for another choke.....whatsthatallabout!

It has been suggested to me that ct transformers are not right for a bridge rectifier, can anybody comment on this.

anyway I'm here at the moment, wanting 500-550v(110ma) and 300v(13ma). The only addition would be the drain/choke load resistor. Has anybody got any comments, which would be appreciated because I've never done choke input before.
 

Attachments

  • sv572-ss-chokeinput-s.jpg
    sv572-ss-chokeinput-s.jpg
    103.5 KB · Views: 1,688
One thing to keep in mind is that the supply is unloaded during valve warmup or some unusual condition (open valve, etc.) and voltage will soar to full cap-input value, in this case about 750 volts DC. You must be prepared for this, or be scrupulous about start-up sequencing (and lucky!).

All good fortune,
Chris
 
For starters, which 300-0-300 Hammond power trafo is being used? The VA rating of the entire thing has to be examined. A good deal more B+ current can be drawn, when choke I/P filtration is employed, as opposed to cap. I/P filtration. If the power trafo has the requisite VA "stones", a 10 KOhm bleeder resistor of appropriate wattage rating can be placed in parallel with the 1st filter cap. and a guarantee of the critical current being drawn from the moment of turn on will be established.

The 1st choke in a LCLC filter takes a pounding. Unless the part being used is specifically rated for choke I/P service, current handling capability has to be substantially derated. If a 0.22 μF./1500 WVDC film cap. is placed in front of the 1st inductor, you reduce strain on the inductor and protect the SS diodes against nasty kick back spikes.

DF96's advice about taking a lower voltage rail with a 5th diode from the CT, ala Pete Millett, is GOOD. There are no worries about adequate decoupling. FWIW, I suggest you use a CRCRC filter on the low voltage rail and keep the 1st cap. on the small side. The 1st cap. should be only large enough to keep the rail voltage up. Start with as little as 3.3 μF. and see what you get. Perhaps PSUD simulations will reduce the amount of cut and try you have to do.

BTW, get rid of the 1N4007s and install UF4007s. At a cost of pennies, the amount of nasty SS diode switching noise generated will be reduced. Along the same line of thought, use a 600 PIV Schottky as the "5th" diode and block the crud from entering the low B+ rail outright.
 
Last edited:
Thanks for the input guys

anyway I'm here at the moment, wanting 500-550v(110ma) and 300v(13ma). The only addition would be the drain/choke load resistor. Has anybody got any comments, which would be appreciated because I've never done choke input before.

the choke load resistor I referred to in the first post was to cater for the min load to make the choke operate...from Morgan Jones I'd calculated I wanted 52ma, which was 10k @ 25 watts or so...I couldn't factor it in on psud, which is why I mentioned it.

The Hammond is a 372fx which is 175ma, I know I needed more than my 110-120ma but I wasn't sure how much more and I found the section on Tx current rating a bit taxing.....Have I got enough headroom??

thanks for the heads-up on the diode types Eli...as you can see I'm floundering a bit with the SS stuff.

Ed
 
As an alternative to a fat resistor for establishing the minimum current you could use a string of fat zener diodes. The advantage is that once you take external power they have less to do and so less power is wasted. The sum of the zener voltages should be a little more than the usual HT voltage, so they only operate when needed. If you do this you must ensure that the smoothing caps can take full voltage, in case a zener goes o/c.
 
A 10 KOhm bleeder will dissipate 30.25 W., when the rail voltage is 550 V. Good practice requires a 100 W. rated part for the job. Such a resistor is big and expensive. DF96's Zener diode stack seems to be the best option. Use a 50 KOhm/3 W. metal oxide part in series with Zeners that total to "600" V. Use series connected pairs of 450 WVDC 'lytics in combination with 150 KOhm voltage equalizers.

I make the 150 VA rated 372FX as being adequate, when the 5 VAC winding is unused and a power wasting bleeder resistor is not employed. A nice thing about choke I/P filtration is that you can access the full VA capability of a power trafo. OTOH, cap. I/P filters are limited to approx. 1/2 of the RMS rated current the rectifier winding can deliver. The difference in the way I2R heating losses occur is the reason for the previously mentioned facts.
 
Thanks Eli

It's very comforting to hear the Tx will be adequate. I am completely in the dark about zeners so as this is my first foray into choke input I thought I'd keep it as simple as possible. The intention was to parallel an array of high wattage resistors to burn off the 25 or 30 watts for the choke load.

Any chance of a reference for the zener use, I will read up in the meantime.

Ed
 
When a choke transformer is used in a choke i/p filter this choke takes a lot of "pounding", and the choke radiates a (much?) lager and stronger magnetic field than if the same choke was used in a cap.i/p filter. Does anyone have any practical experience from using a choke input filter and finding / not finding influence from this magnetic field upon other components in the surrounding circuits?

The reason I ask is because I know of two commercial amps that have mounted choke i/p-filter in rather packed single chassis solutions, while here on diyaudio one mostly find postings that say it is an absolute necessity to have a separate chassis for the powersupply if you build a choke i/p-filter. Both these commercial amps gets stunning reviews.

When you plan an amp -build the chassis is often the most troublesome part and building two of them is double trouble, not to mention the cables inbetween. Maybe the most interesting question is how you approach the question if a two chassis solution is necessary and which method could one use to find data of the influence from the magnetic field upon the other components.
 
I guess that the biggest problem is if other transformers pick up this magnetic field. Here are a table that a friend made for his amp build. It shows how the powertrannys magnetic field induces current in the choke windings at different distances between pwr and choke.
 

Attachments

  • pwr+choke.jpg
    pwr+choke.jpg
    156.3 KB · Views: 606
  • plot pwr+choke.png
    plot pwr+choke.png
    19.9 KB · Views: 602
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.