Voltage comparator with valves?

[Dekatron]

I need help in finding a voltage comparator that is built with valves, preferrably just one valve (if possible) like any of the ECC81, ECC82 or ECC83 valves or the E90CC, E92CC valves. I need this voltage comparator to be able to compare a voltage up to 400V on the measuring input to a set voltage on the other input and at the same time draw extremely low current on the measuring input. With extremely low I mean in the region of a few microamperes or lower if that is possible.

I need the circuit to be quite cheap to build so rare and expensive valves is not an option unfortunately.

I have searched the Internet and looked through all books at the local library but I have so far not found anything at all.

So I hope someone here either has and idea where I can find such a circuit or how to build one.

[glowinganode]

Something with a top cap grid, like a 6BS7 see.
Rob.

[woodchips]

Sounds like the dual sweep position pickoff as used in scopes like the Tek 545? Bob

A long tailed pair would work, when you say up to 400V, what is the low point? How much HT have you to power this circuit and what output do you require?

[glowinganode]

Or the ME1400/1/2 electrometer tube designed especially for this purpose.
A Dynamco Digital Volt Meter I have uses a pair of these in the input stage. Unfortunately I don't have a schematic.
Rob.

[Dekatron]

Thanks for all input, after looking at the valves suggested and also trying to find schematics where they are used I reailised I had to give a better explanation on what I need.

Merlinmaxwell, I have up to 500-600V HT to use and the voltage I need to measure is from 14V, stepped at 14V steps, up to atleast 140V but it can be biased up to 156V for the lowest point resulting in 170V as the lowest point and 310V as the highest point (140V plus 170V). The lowest point is also allowed to be from 20V, stepped in 20V steps, which means that the highest point can be up to 356V (200V plus 156V).

The tube I need to measure the voltage on is an E1T scaler/counter tube, you can read more about this fascinating tube here : http://www.dos4ever.com/E1T/E1T.html . On one side there is a deflection plate D´ held at up to +300V with a 1M resistor, this plate will automatically step the voltage in ten steps when you pulse the other plate D with 14V-20V signals of the correct shape. Since this plate will automatically hold the voltage it has been stepped to, until another input pulse is put on the other plate, by how the valve is constructed internally not much current can be drawn from it. If you draw to much current the valve will change steps as the plate voltage falls below a stable point.

Philips made a valve named E80T (6218), which is very rare, that can be used for this purpose. It is a beam deflection valve which is almost like an E1T but without the screen with numbers and instead of ten slots it has only one slot which when the voltages on D and D´ is equal lets the electron beam through to a plate which then outputs a signal - like a voltage comparator. Burroughs made a similar valve named 7360, but that valve is equally rare and I have not seen any application examples with it as a voltage comparator, only been told that it could be used as that. The 7360 is mostly used as a mixer according to the datasheets.

And I need atleast 20 of these for my project, that's why I am looking for a simpler solution. So anything with a double-triode would be most welcome.

[AlanBeckett]

The Airmec 314 Valve Voltmeter uses a 6BS7 and measures up to +/- 300 volts DC. Actually it's good for 30kV with the optional HV Probe.
Why comparator as opposed to straight measurement?
Alan

[Ed_Dinning]

Hi Martin, try Googling or check your theory books for a Schmidt trigger. This is what it used to be called in valve days.

There may also be some data in books on trigger tubes or valved counting ciruits.

PM me if you have no success and I'll see what can be found (but it may take a while)

HTH Ed

[Herald1360]

Following on from Alan- why not just use a 10:1 divider on a 10meg input DVM- nine 10meg resistors in series with the input should do- you can check their actual values with the DVM if you need better than their tolerance for accuracy.

Finally, loading the 1meg source with 100megs would give a reading 1% low.....

Cathode follower from the output, fairly small cathode resistor.
A number of triodes with the cathodes biased to the voltage you want to compare (will need tweaking)
Feed each grid via a few MOhms to stop grid current being a problem.
Each anode will take current when the grid goes up past the cut off point.

One thing is watch out for cathode to heater voltage.

[G8HQP Dave]

Sounds like a single range Vacuum Tube Voltmeter! An ECC82 should do the trick.

[Dekatron]

Thank you all for your inputs, now I'll have to check all different types and see what I can find.

If anyone has references to books which contain examples with schematics for these kinds of solutions I'd be very happy if you could write those here in the thread, or even if you could include a picture or two.

My mind works best when I have a picture as a reference for what it is that I should be looking for - especially now when I am on totally unfamiliar grounds.

[Dekatron]

Now I have been trying to find more information, but it is not easy to find schematic drawings with tubes and an explanation at the same time.

However it looks like I will be using a schmitt-trigger and by using a switch with ten defined steps I can raise the "ground" level of the schmitt-trigger to set the voltage range around which it will have its reference level from which it will trigger.

Please correct me if I am wrong, but that is how I understood that you "bias" a schmitt-trigger to set the voltage level around which it will trigger with respect to the high & low trigger points (hysteresis?) decided by the calculation of the resistor values in the circuit.

I need the circuit to work like I have tried to describe above with ten different voltage levels set by a switch as a "bias" level and then use the schmitt-trigger to "capture" the variation in the voltage that is checked.