|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
7th Sep 2020, 12:47 pm | #1 |
Heptode
Join Date: May 2015
Location: Bradford, West Yorkshire, UK.
Posts: 510
|
A choke in a CT supply.
Pardon my ignorance, but while reverse engineering an HF linear power supply, drawing out the diagrams of the various supplies. On the main HT, the transformer is 612-0-612 650mA, the CT is connected to ground via a large choke marked 1.2H @ 350mA (DC resistance 52 ohms). Is it connected to ground so as to have a low PD across it for insulation purposes ?
Thanks
__________________
worried about the electrons entering the circuit and the smoke leaving Andrew |
7th Sep 2020, 1:41 pm | #2 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: A choke in a CT supply.
I think that that's quite likely- it makes sense, whether choke-input or capacitor-input (reservoir between centre-tap and rectifier output). It's something I do when building bespoke HT supplies- I can't see any purpose in fitting the first choke in the positive rail other than being a hang-over from when multi-section HT electrolytic capacitors were widely used for economic rather than technical reasons. Sometimes, the voltage drop across a negative-line choke can be used for biassing purposes but it needs filtering.
The Murphy B40 is a good example of something with a negative-side HT choke (separate paper-block reservoir and smoother), I'm sure that there are plenty of others- the AR88 would have benefited internal heat-wise from this way of doing things. |
7th Sep 2020, 2:10 pm | #3 |
Octode
Join Date: Jan 2004
Location: Southport, Merseyside, UK.
Posts: 1,156
|
Re: A choke in a CT supply.
I've seen this suggested for high voltage power supplies in editions of the ARRL 'Radio Amateurs Handbook'.
John |
10th Sep 2020, 6:51 am | #4 | |
Heptode
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
|
Re: A choke in a CT supply.
Quote:
https://dalmura.com.au/static/Note%2...its%201934.pdf
__________________
Robert |
|
10th Sep 2020, 9:14 am | #5 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
|
Re: A choke in a CT supply.
By Fred Terman, no less!
If you have appreciable ripple across the choke, then although placing the choke in the -ve side of the supply reduces the insulation requirements on the choke, it also increases the insulation requirements of the power transformer. Also be aware that choke-input supplies need a minimum amount of load current to work as mean-value rectifiers.... designed to store enough energy in a large enough choke that the choke discharge provides output current while the transformer voltage is too low. Take less current than the calculated minimum and the output DC voltage goes up. At low current the circuit turns into a peak-value rectifier. If testing things downstream, or just pulling audio output valves while chasing a fault, the capacitors in the thing can get a high-voltage work-out! So you need to spec working voltages to survive. If you use solid-state diodes, or a fast-heating directly heated rectifier valve, this overvoltage will happen on every switch-on. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
10th Sep 2020, 2:18 pm | #6 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: A choke in a CT supply.
I've chewed over the implications of shifted transformer secondary potentials quite carefully before wiring things this way- my take on things is that the centre-tap connection in bi-phase full-wave circuits is typically physically between the two winding layers of its half-windings, (hence the oft-noted difference in resistance of the two sections) so well separated from other secondaries/screens/primary, and elevating to several units/few tens of negative volts wouldn't be a problem with a half-decent transformer. The absorption of that voltage with a negative-side choke means that the two winding ends (physically closer to other electrically-separate windings) are at a correspondingly lower running peak potential to adjacent windings/core, which is surely a good thing.
I can see Terman's take on the capacitance-to-ground problem- a good illustration on how the most important components in a circuit can sometimes be the ones that aren't shown on the circuit diagram!- but presumably this problem is worst with choke-input, with capacitor-input and centre-tap choke connection the reservoir and smoother are effectively in series to 0V from the centre tap's viewpoint- still much larger and lower-impedance than winding-to-core capacitance. Sometimes with choke-input HT supplies, the presence of screen-grid/oscillator supply gas-filled shunt stabilisers can help by firing more-or-less instantaneously with initial overvoltage from DH or semiconductor rectifiers and drawing current to limit the voltage-rise problem- they're still something that can age and lose effectiveness, so not an excuse to under-rate capacitors! I've certainly heard and can understand that using a transformer in a much-changed manner to the designer's original brief and consequent physical layout can give rise to problems- e.g using a 1000-0-1000V winding for a conventional full-wave rectifier as a half-wave rectified 2000V winding would be risky, Colin |
10th Sep 2020, 2:55 pm | #7 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
|
Re: A choke in a CT supply.
Maybe that's something else, when an experienced engineer looks at a diagram, they not only see what is shown, they also see the strays.
It's like listening to an interview with a politician, you have to pay a lot of attention to what they DON'T say! David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
10th Sep 2020, 3:09 pm | #8 | |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
|
Re: A choke in a CT supply.
Quote:
However, when going negative, the voltage will be rather more extreme! I did some quick calculations. Assuming perfect diode rectifier valve, and zero resistance transformer and chokes, 250-0-250V RMS secondary: Choke in series with rectifier cathode, CT to chassis. HT voltage: 225V DC. Peak positive secondary voltage with respect to chassis: 354V. Peak negative secondary voltage with respect to chassis: 354V. Peak positive CT voltage with respect to chassis: 0V. Peak negative CT voltage with respect to chassis: 0V. Choke in series CT to chassis, rectifier cathode direct to smoothing capacitor. HT voltage: 225V DC. Peak positive secondary voltage with respect to chassis: 225V. Peak negative secondary voltage with respect to chassis: 482V. Peak positive CT voltage with respect to chassis: 225V. Peak negative CT voltage with respect to chassis: 129V. So the highest voltage does actually appear when the choke is in the centre-tap. Last edited by kalee20; 10th Sep 2020 at 3:13 pm. Reason: Correction |
|
10th Sep 2020, 5:05 pm | #9 |
Dekatron
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
|
Re: A choke in a CT supply.
I've just looked at a Murphy B40 schematic I have, there's two chokes, one in the -ve line and one in the +ve line, the one in the -ve line is parallel tuned, copy of said manual is available here, then scroll down for the power supply schematic:
http://www.portabletubes.co.uk/boats/murphy3.htm Lawrence. |
10th Sep 2020, 7:18 pm | #10 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: A choke in a CT supply.
I see what you mean (kalee20), I managed to overlook the negative-going excursion which is kinda dumb! I have to admit I was thinking of the more familiar-to-me capacitor-input case where there's less cyclic voltage swing at the choke input (there would still be a higher negative peak though), whilst the OP I think was about choke-input. Choke-input comes under the heading of "things that might really hurt" for me, like big AF and RF PAs that might have mercury vapour rectifiers, not my comfort zone.
|
10th Sep 2020, 7:34 pm | #11 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
|
Re: A choke in a CT supply.
Had to do a multioutput power supply with choke input filters (required by the design manual for the system feeding 24v 40kHz sine power to the transformer. Small ferrite E core transformer, not got the winding space to get everything in with heavy formvar, needed to use ordinary solder-through varnish, so I had to watch the voltage ratings. A filter choke needed to run in continuous current mode certainly does gymnastics in the voltage domain.
Thanks, Peter. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
11th Sep 2020, 6:23 pm | #12 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,951
|
Re: A choke in a CT supply.
I'm sure I've seen a circuit for a 'beefy' HF linear-amp which placed the smoothing-choke in the centre-tap-to-ground path, but then used the ripple-voltage developed across it to feed a circuit where the ripple-component was rectified and provided 100V or so of 'free' -ve grid-bias.
|
11th Sep 2020, 10:36 pm | #13 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: A choke in a CT supply.
The Marconi 559 here has its loudspeaker field coil connected between transformer HT centre tap and chassis in its additional role as smoothing choke, generating approximately -100V across its 1600 ohms, there's a potentiometer across it to pick off approximately -15V for the output valve grid and about -2V for AGC delay.
|