Continuing my accrual of test gear applicable to components used in receivers up to 1935 it occurs to me that some means of plotting the characteristic curves of simple valves would be useful, particularly for homebrew designs. I know manufacturers were in the habit of providing this information and that it’s an easy matter to get hold of, but my understanding is that valves do suffer degradation, particularly the older they are, the more use they have had and the conditions under which they have been stored. So being able to plot the curves for the valves I possess might be an advantage. At the very least I need some way of telling how well they are working.
All the receiver circuits I have considered use triodes, with just the odd one or two having pentodes too. With this in mind I was going to start by constructing a very simple triode valve
static curve plotter along the lines of the attached circuit. Of course it would be nicer to have a tester that would accommodate pentodes too but I fear to build one would be a lot more complex and taking me towards the realms of a proper valve tester. At present I have neither the skills to build one nor the funds to purchase one. And the fact is I don't want to buy one. I'd much rather make something as it contributes to the learning curve.
The circuit I have designed uses batteries for HT, LT and GB. The LT supply is fixed and uses 2v Cyclon SLA cells to supply 2v, 4v or 6v as required. The HT and GB supplies are variable by means of potentiometers with accompanying voltmeters to select the desired voltage. A milliameter in the anode circuit supplies the data for plotting, which will be done on an Excel spreadsheet with a graphical output (as illustrated in the attached picture).
I could begin building this circuit immediately as I have all the parts, but I recognise it has a number of shortcomings and I am consequently very open to critique of my proposed approach and to thoughts and suggestions for a better solution, keeping in mind of course my limited knowledge and aptitude
Nick.
Nick, there's potentially a grid bias battery shorting problem with the bias pot arrangement, the switch (6) might be problematic in the position shown, a D.P.S.T switch would be better to switch the HT + and the LT +.
That aside you should be able to plot anode/plate characteristics with it, once you have a series of plots then other things can be worked out, Ra, ma/v etc.
You might already have this but if not here's a link to Langford Smiths article on the subject:
http://www.paleoelectronics.com/RDH4/CHAPTR02.PDF
Here's a link to The Radiotron Designers Handbook (the source of the above article) From this link you can get to various other chapters easy enough:
http://www.paleoelectronics.com/RDH4/
This post might or might not cross with someone's
Lawrence. EDIT: You could use a higher value for the grid bias pot (10) and possibly for the anode supply pot (5)
Nick,
I think you may have a problem with power dissipation in the anode supply pot. If I've got my sums right the standing current in the pot will be 66mA (198/3000 = 0.066A), which means that it'll be dissipating just over 13 watts even before you take into account the triode. That's quite a big pot!
As Lawrence has suggested you could use a higher value and that would reduce the standing current. Whatever value you go for though this will be dissipating power and it'll probably always need to be a wirewound pot rather than a carbon film one.
Hugh
The other option is to go for monitoring cathode currant with a separate grid bias supply and not the common supply used for transistors in the case of my late 1970s home design/build when I was a teenager and relied on bags of unknown transistors for all my experiments.
It can also be seen with a bench power supply as a range extender allowing power transistors to be tested in anger.
It is still fully working apart from some of the old resistors in there having drifted a bit.
Thanks for the article Lawrence. In fact I have a copy of the Handbook but until now it is unread. I have been referring to the subject in one of my FJ Camm books until now but the chapter you provided is much more comprehensive so I'll study it and expand my knowledge further. I will also do as you suggest guys and will use wire wound higher value pots. I will also use a DPST switch.
If I wanted to test pentodes too, perhaps switching from one circuit to another, do you know of a source of information from which I could derive a simple circuit?
Thanks
Nick
(15-07-2015, 12:19 PM)Refugee Wrote: [ -> ]The other option is to go for monitoring cathode currant with a separate grid bias supply and not the common supply used for transistors in the case of my late 1970s home design/build when I was a teenager and relied on bags of unknown transistors for all my experiments.
It can also be seen with a bench power supply as a range extender allowing power transistors to be tested in anger....
I'm a little embarrassed to confess I didn't understand any of this!
Nick
What it does is to set the currant by monitoring across the fixed resistors so that the two LEDs switch over when the voltage across the resistor is in my case 5 volts so that there is no need for a second meter as these were expensive when I made it. The base currant is then read off and the gain calculated. Now I have a bench PSU I can set it to constant currant and feed it between collector and ground in order to get more currant ranges.
A valve or FET version could be created by fitting the range selector resistors between cathode and ground and with FETs take the gate from a potential divider across the source resistor and HT supply. The gate voltage on FETS is positive. With valves the grid needs to be negative so it would most likely need an isolated grid bias supply and pot as your circuit has. With a 3 position switch you could do valves FETs and transistors with ease.
If I built a new one there would be an analogue self adjusting circuit and no pots.
I'm a great believer in hooking up ad-hoc measurement rigs with a couple of power supplies, odd pots, meters and croc clip leads so I like your basic approach. You can learn a lot that way too. You can also get think badly wrong but that's all part of the learning experience.
Or you could do it in style:
http://www.dos4ever.com/uTracer3/uTracer3.html About £160 for the kit at current Euro exchange rates. I'm seriously considering getting one.
Ah.
I came across that uTracer a while back Jeffrey. Looked like a very interesting project but far too sophisticated for my needs. For the stuff I'm interested in I suspect I will only ever need to plot basic curves for triodes and pentodes. Not even sure I will need to do dynamic measurements, static should suffice.
Nick
(15-07-2015, 01:40 PM)ppppenguin Wrote: [ -> ]I'm a great believer in hooking up ad-hoc measurement rigs with a couple of power supplies, odd pots, meters and croc clip leads so I like your basic approach. You can learn a lot that way too. You can also get think badly wrong but that's all part of the learning experience.
Or you could do it in style: http://www.dos4ever.com/uTracer3/uTracer3.html About £160 for the kit at current Euro exchange rates. I'm seriously considering getting one.
I agree, but then there's the matter of having the wherewithal. I can summon up an HV PSU, a couple of LV ones, Meters, Bases etc, but if you haven't got any of those it's probably cheaper to buy or build something to fit your needs.
I keep pondering on the utracer, but haven't bitten the Bullet yet. Anyway, I haven't got a Workshop PC.
Alan