A Simple, Valve Curve Plotter

[pwdrive]

With the pot divider don't forget the two currents, pot and load, work out the highest dissipation likely to be encountered and what the pot resistance (top arm) is at that setting, divide the top arm resistance into the total pot resistance to obtain a ratio, check that against the wattage rating for the pot for a derated value for that pots setting...Have I got that right? if not the guilty as charged.

Lawrence.

[SurreyNick]

I'll be careful to do the calcs Lawrence. I have the pot handbook you gave me to refer to too.
Nick

[pwdrive]

Nick (surrey) I've just been looking at the LM317 data sheet, apparently it's a floating device so that means it can be used for HT regulation, but only over a certain voltage range, that range is governed by it's maximum voltage differential rating (voltage difference between Vin & Vout) for a bog standard LM317 it's 40 volts, there is a higher differential voltage version the LM317HV which has a maximum voltage differential rating of 60 volts.

A wider range of HT voltage can be obtained by tying the LM317 up with a pre regulator, TI (Texas Instruments) do a notes sheet for it in the link below, it shows a variable HT supply with an output range of 1.2 volts to 160 volts at 25 ma, if the minimum voltage setting could be increased then that might give a voltage range more suited for plotting valve characteristics if a voltage of more than 160 volts is required, the output current can be increased above 25 ma if needed as per the TI notes.

http://www.ti.com/lit/an/snva583/snva583.pdf

Food for thought.

Lawrence.

[SurreyNick]

Oooh, now that's interesting. I'll have a look at that Lawrence. Thanks
Nick

[pwdrive]

A TL783 will do 120 volts as is.

Lawrence.

[SurreyNick]

Well that TL783 sorts out the screen grid circuit and the LM317 sorts out the control grid circuit. Just the pre-regulator method to sort out for the plate circuit.
Although it may be unnecessary to use ICs to completely smooth out the rectified curent it's so simple to drop them in I can see no argument against doing so
Nick

[pwdrive]

Even more food for thought, use the anode and screen grid supply circuits as used in the Sussex valve tester, looks complex but it's reasonably straight forward, schematic can be found here:

http://www.vintage-radio.net/attachments...sex-v1.zip

Lawrence.

[ppppenguin]

What happens when the HT suffers a momentary short? Yes, you've guessed, new LM317 time. Floating a regulator is really only for very controlled environments, not the relative rough and tumble of valve testing.

[pwdrive]

What happens when the HT suffers a momentary short? Yes, you've guessed, new LM317 time. Floating a regulator is really only for very controlled environments, not the relative rough and tumble of valve testing.

That's why I like valves, they recover better, the variable HT PSU that I have is a valve one I cobbled up with various bits, it's had a few wars but still works ok.

Lawrence.

[SurreyNick]

After all the debate I have, in fact, backed off a bit and opted for something a bit 'rough and ready'. The design has no real safeguards for the valve under test, but because the unit is intended for my personal use and I am the designer I ought to know what procedure to follow when using it so I deemed it a reasonable risk. I also went away from using regulators to get perfectly smooth supplies. I thought I would initially have a go without them and see how the unit performs. They can always be added in if necessary. I have also persisted with using wire-wound potentiometers to adjust the various circuits. It's an experimental unit and I believe it's worth seeing how it performs with this rough and ready approach.

I've drawn a circuit and made an illustration of the enclosure front panel, both of which are attached. I'm not sure I have got the circuit right though. I was going 'line blind' by the end and fatigue had well and truly set in. If you spot any mistakes do please shout

I'll return to it in a day or two to make any corrections.

Nick