Simple PSU for a few valves

[AlanBeckett]

Al,
Ah, so.
I've never been a fan of mounting the choke under the chassis.
Ala

[Skywave]

Alan,
Choke under the chassis: in this case, it was simply a question of squeezing all the bits into the available space. However, generally speaking, apart from reasons of heat dissipation, why do you say that? I'm not questioning your judgement or knowledge of things technical & design, etc. - it's just that you might be aware of something here that I'm not.

Cheers,
Al.

[AlanBeckett]

Al,
I haven't a clue. Habit I suppose. It probably goes back to the days when such things were mounted on the top, and care was taken to keep the orientation of the cores at right angles to cut down on hum. That's where 'drop-through' transformers scored 'cause it gave you another orientation.
Alan

[Skywave]

Thanks for the reply, Alan. I kinda suspected that was the reason - "because that's the way I remember seeing it done / that's the way I was taught it should be done" when we were in our 'electronic formative' years.

If I may just be allowed to go off on a tangent from that for half a mo., it illustrates what I believe to be a weakness in the human psyche. It's all very well and good saying "Because it's always been done that way & we can guarantee success by doing it that way, that's The Way of Doing It", but that philosophy impedes discovery & invention. One of my favourite quotes from the general field of Science is "All great discoveries were made by mistake". I think that the phrase 'by mistake' in that quote is not quite the right - "when an unanticipated result developed " seems more appropriate - but I think we all appreciate the point of that quote (which, by whom, I cannot now recall).

End of tangential off-shoot.

Al.

[Skywave]

That's where 'drop-through' transformers scored 'cause it gave you another orientation.
Alan

And the simple fact that that enabled any high voltage tags on the transformer to now be located beneath the chassis - safely away from wandering fingers! :s

Al.

[Mark Hennessy]

HT power supplies seem to be something of a recurring theme and to me there is an answer so obvious that there must be a good reason for not doing it of which I'm unaware, namely using a thyristor.

Hi Joe,

I've seen thyristors used in lots of bench powers supplies. But these examples have been as pre-regulators; in other words they were followed by a conventional linear regulator. The control circuit was arranged such that the thyristor provided, if you'll forgive the over-precise maths(!), Vout plus a bit, thus reducing the DC drop across the linear regulator, hence reducing the heat generated, hence reducing the size of the heatsink needed, hence reducing the overall cost. Yes, always follow the dollar

I've never delved into these in huge detail, but would imagine that the raw DC at the smoothing capacitor might be a little rough - perhaps contaminated with switching spikes?

These days, switched-mode supplies have replaced thyristors, but many still have a linear post-regulator to clean up the output.

(Power supplies are a special interest area for me. I don't get out much :s)

Mark

[Brianc]

Hi all
I've got a home brew PSU, 300VDC & 6.3VAC which is going spare. If anyone would like it - either to use or to use the components from, they are welcome to it. I can bring it to NVCF on Sunday - first come first served.

[Joe]

Thanks for that, Mark.

I was curious as if somebody had just said to me that they needed to convert 240vac to 90vdc a tryristor would have been the obvious solution, though I couldn't have come up with a practical circuit, so I've always wondered (not that I've been into valve stuff for long) why there seems to be a myriad of other, apparently more complicated and expensive, solutions knocking about but not this one. From the data I've been able to find in recent days thyristors are easily able to handle the currents and voltages involved so perhaps it is just that the output would take so much smoothing and filtering that there'd be no advantage over a transformer.

Power supplies are something I tend to build as well, chiefly out of meanness regarding buying batteries. Mine have never been more sophisticated than transformer, rectifier, smoothing cap and an LM317T though. It's a pity that the latter and their brethren have a maximum output voltage around forty.

Switched-mode supplies are things I've never dealt with as I haven't had a need to as yet.

I look forward to reading you contributions on PSUs and getting juice into anything is the first step.

All the best
- Joe

[tiv48]

here is the circuit of a typical tv (phillips g8) tryristor power suppy

[Skywave]

Generally speaking, thyristor power supplies, conventional switch-mode PSUs and the line O/P stage in a T.V. have a lot in common. They all work on the principle of energy being stored in one reactive component being switched into another reactive component with the thyristor or transistor acting as the switch. It's the on/off ratio of that switch that determines the ratio of Vin to Vout. The diode that always features in such circuits simply completes the current path when the switch goes off. Consequently, with all that switching going on, to avoid the use of large reactive components and to reduce switching losses, the switch frequency is made high - way above 50 Hz (in the vast majority of cases). Since the switch - by definition - must be either on or off, the losses in it (compared to a linear regulating device, which, in effect, acts as a variable power resistor) are a lot less. The price to be paid for this dramatically improved efficiency is, of course, greater circuit complexity compared to a linear regulator PSU. The major problem when a thyristor is used - as opposed to a transistor - is that unlike a transistor, the only way to switch a thyristor off is either to totally break the anode - cathode current or by making the anode negative with respect to the cathode: often not easily arranged. Once the thyristor is on, the gate has no further control over the anode-cathode current. This unique feature of a thyrsistor in a PSU configuration is, I suspect, one of the main reasons as to why thyristors are infrequently used in switch-mode PSUs: it's far easier to do the 'switching off' with a conventional bi-polar transistor; other comparable advantages are minimal.

Al.