Variable HT regulator circuit.

[jim_jobe]

The transformers in old shaver sockets are roughly 240 to 240 and, though they don't look it, they are usually rated at 100mA. There's no heater winding with these though.

[Restoration73]

http://livinginthepast-audioweb.co.u...p?p=xfrmrvt227

[David G4EBT]

Originally Posted by SurreyNick

Could someone recommend a suitable (cheapish) transformer to use with this circuit?

Easiest and cheapest way to do this is to use two mains transformers with low voltage secondaries of say 6, 9, 12 Volt or whatever, (actual voltage unimportant as long as they’re the same), which will provide the current you’re looking for, and to connect them back-to-back, as shown in the diagram below. Depending on the method of construction you are employing, could use either a PCB-mounted transformers, or wire-ended. As small transformers are so cheap new, given that you only want 100mA, you could use say two 300mA transformers to allow for any efficiency losses incurred by connecting them this way. For example, these wire-ended 6-0-6 300mA transformers from ESR are only £4.17 each plus £3.50 P&P plus VAT. Thy also sell PCB mounting transformers. Lots of other suppliers (Rapid etc) supply similar ones at similar prices:

Order code 311-306, 6-0-6- 300mA 3.6VA Chassis Transformer:

http://www.esr.co.uk/electronics/pro...ansformers.htm

Just ignore the centre taps of course, so in effect, you're connecting two 12V windings back to back. The same technique was used in the Radio Bygones Capacitor Reformer project that I built, which used two back-to-back transformers, then a diode voltage doubler to obtain the 500Volts maximum (adjustable) to reformer electrolytic caps. (As an aside, it's an excellent project which works very well and monitors the reforming progress).

Hope that helps.

Good luck with the project!

[Studio263]

Remember that 300mA is the low voltage secondary winding current, connecting two 300mA 6V transformers back to back would not give you anything like 100mA at 240V. Going back to the original spec, 300V at 100mA is 30W, so transformers rated at 30VA are a minimum requirement. 50VA would seem a sensible minimum for a practical design.

[Diabolical Artificer]

I've rewound several transformers using this program as a rough guide for turns.
http://www.electronicecircuits.com/e...ation-software
Obviously there is a lot more to winding transformers than winding a few turns of wire on a core. But for transformers needing only a few 100mA or so, I,ve found it a relatively simple way to get a transformer for a project. The beauty of this approach is that is very simple to add heater windings which need only a few turns of heavy gauge wire.

I,ve used the circuit using a IRF740 and found it easy to make and reliable.

Re- adjustable HT circuits, I'm pretty sure i have an Elektor article; I'll see if I can dig it out.

Andy.

[David G4EBT]

Originally Posted by Studio263

Remember that 300mA is the low voltage secondary winding current, connecting two 300mA 6V transformers back to back would not give you anything like 100mA at 240V. Going back to the original spec, 300V at 100mA is 30W, so transformers rated at 30VA are a minimum requirement. 50VA would seem a sensible minimum for a practical design.

Of course - thanks for pointing that out!

[Diabolical Artificer]

Found the article for a 0 - 300v stabilised PSU , hope you can make out the details ok as its a Jpeg. Andy.

[daviddeakin]

Originally Posted by Studio263

Remember that 300mA is the low voltage secondary winding current, connecting two 300mA 6V transformers back to back would not give you anything like 100mA at 240V. Going back to the original spec, 300V at 100mA is 30W, so transformers rated at 30VA are a minimum requirement. 50VA would seem a sensible minimum for a practical design.

If you increase R2 to ensure a suitable current limit, you can use a smaller transformer. For example, this one is rated for 76mA:
http://livinginthepast-audioweb.co.u...=xfrmrvt1581cl

A 10 ohm resistor would set a current limit of about 60mA.

I would recommend putting a 1k base resistor on Q2 though, as it provides some robustness when switch on with a short circuit load.

[Mr Moose]

Hello,
Any of these from cpc would be more than adequate and supposedly free carriage on web orders.

http://cpc.farnell.com/jsp/search/br...vNValues%3D411

The last one if connected in reverse would give you a higher voltage supply.

Shaver sockets are often only rated at 20VA

A cheaper alternative would be to use a mains transformer from scrap valve radio or radiogram and use full wave solid state rectification instead of the bridge rectifier. An average radio would only give about 60 - 80mA but a stereo radiogram or many valved radio should be good for 100mA and the transformers are rated for the pulsed nature of rectified current.

Yours, Richard

[jim_jobe]

Thanks for that info Andy

[jim_jobe]

Originally Posted by daviddeakin

If you increase R2 to ensure a suitable current limit
I would recommend putting a 1k base resistor on Q2 though, as it provides some robustness when switch on with a short circuit load.

Good advice that 1k resistor David. Yes about R2 as well which is easy to calculate:

R2, in kohm =0.6/ max required output current in mA (where 0.6V is the base-emitter voltage needed to switch the bipolar transistor on.)

[SurreyNick]

Swell, thanks, that deals with the transformer.
One final (I think) question... what heatsink should I get for this IRF740 MOSFET?
Thanks
Nick.

[Diabolical Artificer]

They don't get very hot Nick; they can be bolted on a metal chassis (make sure to use a mica washer to isolate from any other devices/ - shared earth.

If you need a single heatsink, I have one spare, Nick. Drop me a PM.

Andy.

[daviddeakin]

Originally Posted by SurreyNick

Swell, thanks, that deals with the transformer.
One final (I think) question... what heatsink should I get for this IRF740 MOSFET?

That depends on how robust you want the supply to be. If you have the output voltage turned down to almost zero (or the output is shorted) then virtually the whole supply voltage will be dropped across the MOSFET. That might be 350V or something. If you're current limit is 100mA, then it's going to have to dissipate 35 watts. That's going to require a finned aluminium heatsink about the size of a half brick, or so.

[Al (astral highway)]

Originally Posted by SurreyNick

Could someone recommend a suitable (cheapish) transformer to use with this circuit

Hi Nick, I see you have plans for this but if cost is an issue, you can buy a ready-made boost converter for less than £10. About 200V 150 mA output as the front end and then make that variable...

Just a thought if you need alternatives as transformers ain't cheap. Let me know and I can send you a link

[SurreyNick]

Hi Al
I thought this variable voltage circuit would be a good option to provide the HT supply to my simple valve curve plotter. I will be plotting curves at increments from 60v to 200v. Cost is an issue yes, so the most cost effective option is very desirable.
Thanks
Nick

[daviddeakin]

Originally Posted by SurreyNick

I thought this variable voltage circuit would be a good option to provide the HT supply to my simple valve curve plotter.

Are you sure that's what you want for a curve plotter? It would be rather easy to crank up the voltage accidentally to the point where you're over dissipating the valve. Curve tracers normally sweep the anode voltage up and down continuously, to minimise dissipation (e.g. unsmoothed rectified AC).

EDIT: Just saw your thread about it. I'm guessing you don't have an dual-trace oscilloscope then?

[Al (astral highway)]

Originally Posted by SurreyNick

Cost is an issue yes, so the most cost effective option is very desirable.

Here you go, Nickhttp://www.ebay.co.uk/itm/POWER-SUPP...item3f37a8f31e

You have to factor in a few £ shipping but we're talking probably £15 including the board and shipping. You can't really beat that, it's less than the cost of just one pretty average transformer of the kind you're looking for.

Handle with care. It's only a little board but can sink up to 30mA easily at 200V.

You don't need more power output than this to plot your curves, and maybe it's a safety-wise to design accordingly.

But if you believe you do at any future point, it's easily arranged at little extra inconvenience. Two limiting factors are the MOSFET (The little square up there in the top ... It has four leads but two are connected to the Source) and the current-handling capacitor of the inductor. (Bottom right, labelled 101 = 100uH.)

You'll need some filtering, not hard to do.

Post back on a new thread if you want info on how this works, don't want to go OT.

Cheers,

Al

[SurreyNick]

Thanks.
Nick

[Al (astral highway)]

Meant to change my OP but ran out of time. The filtering on board is in fact very good, so ignore that comment. The power supply is designed to operate nixie tube clocks with no flicker.

If you replace the on-board pot with one with a longer spindle, you can easily control the output. The whole thing can fit into a small plastic box complete with your various voltmeters/ mA-meters and a couple of RCA jacks for the output.