Valve volt meter v Digital meter

[hillmanie]

When I first started reading Practical Wireless 1950s vintage I often read that analogue voltmeters drew too much current in certain measurement situations and that a valve voltmeter was required. Quite outside my budget of course. Then digital meters came along cheaply. How does the digital meter compare in accuracy with the valve voltmeter? Is the digital meter of inherently very high resistance and does this apply only to quality meters like Fluke or also to budget models from my local electrical shop? I'd appreciate a recommendation on any particular brand of the latter that folks have found satisfactory.
Thanks
Tony

[Radio Wrangler]

That's a bit like asking which is longer, a piece of wire or a piece of string?

Valve voltmeters were more expensive than plain ones, so they tended to have good quality meter movements and so missed out on the crappy level that could be found in some plain passive voltmeters.

Digital voltmeters have been made in all sorts of grades from pretty much junk to the current state of the art.

If we trim off outlying exceptions, most DVMs offer finer resolution and better linearity (well, agreement between actual voltage and reading) up the scale.

Analogue meters are still the bee's knees when you're trying to peak things or null things.

There are special FETs offering spectacularly high insulation resistance, and these are matched by special 'electrometer' valves. Once you have a switched range attenuator on the front of even ordinary FETs and valves you are OK.

Of course, those FETs could presage a d'Arsonval moving coil meter (E.G. AVO EA113) or a digital meter.

Where the solid state jobs win is at low voltages. Valve voltmeters tend to drift a bit and so usually had most sensitive ranges of a volt or few FSD.
DVMs tend to have 200mV fsd as a common value for their most sensitive range.

That said, My AVO EA113 analogue meter goes down to 10mV, but my Datron DVM can read microvolts DC.

Oddly, prices for EA113s are a bit jaw dropping. It seems AVO call them up in the calibration procedures for some models of valve characteristic meter and that sets the nutters who pay crazy prices for them out looking for EA113s.

I picked up the Datron for a fiver and fixed it.

The real value of Flukes comes down to personal safety if what you probe isn't what you thought it was. Still plenty of people trust the products of the Shonky Voltmeter Company without thinking, and some of them survive.

So it all depends on what you want to do with it.

There is no one answer.

Keep your eyes open and you'll find bargains. I have a Fluke hand-held in the car. That EA113 in the radio shack along with HP and Fluke bench grade DVMs and a pair of Datron standards lab jobs. None were expensive. Look around and you'll eventually find one of every type you could need.

David

[joebog1]

Not to mention the ease of reading an analogue meter!!! you have selected the 50 volt range and the needle moves halfway, 25 volts in a flash.
Digital, it first revs up digits flying everywhere, then the auto scaling kicks in. You then have to read the ONE pixel number JUST below the level of the display surround is that MV or mV or Mv a pain !!!!

I have both of course and many of each. My standard bench meter is an Avo Model 8, if I need greater sensitivity I use my Kyoritsu VTVM, If I need really sensitive measure I use my HP410B.
Avo was a gift as was the Kyoritsu. HP cost me $100 and just needed cleaning and a lube of the switches. If I need measure 10 amps I use digital. It costs $30 to replace. I do have a Fluke 8050A which has a black ruined display, but can still be read if I need 4 1/2 digit accuracy.

Just my take

Joe

[Bazz4CQJ]

In the context of vintage radio repair, good service sheets typically show voltages at specific test points and then also note the 'ohms per volt' rating of the meter used to make the measurements. In the case of vintage equipment, the kinds of meters in common use in typical repair-shops, and even more so by field engineers, may not have been particularly high.

The point here is that ideally, when doing a repair job it's a good thing to be using a meter with something like the same resistance as recommended by the data sheet, or accept that some difference in reading is to be expected. Some tests might have needed a voltage voltmeter, but it would be wrong to generally think 'high resistance good, higher resistance better'.

Just before Maplin went bust, they were selling an analogue test meter which had a rating of (only) 20kohm/V, and lots of people on this Forum bought them because they recognised that was akin to the sort of ratings often found with the vintage test meters originally used on vintage radios. The very first meter I even owned as a school boy in the 1960's was only 1kohm/V.

Compared with the old analogue meters, even the budget-priced modern digital meters often have remarkably high input impedance and high accuracy.

B

[hillmanie]

Knew the expertise was out there! Thanks David for taking the trouble to explain so fully the story on meters valve and otherwise. Looks like the safest bet is to make sure what one buys has a quality pedigree, by brand and performance. This is a good forum for a recommendation from actual experience
Tony

[G6Tanuki]

While I've used a Valve Voltmeter (we always called them VTVMs - Vacuum Tube Volt Meters) extensively in the past, my biggest issue with using one today would be calibration accuracy.

Components (valves, capacitors, resistors) drift with age - and this can easily introduce significant errors.

Really, you need a known-good meter (or voltage-source) to calibrate your VTVM against.

This is a good example of the 'vintage test-gear paradox' which applies to things like scopes and signal-generators too: you probably need a modern one as a service-tool in order to keep the vintage one working properly!

[Station X]

Originally Posted by G6Tanuki

This is a good example of the 'vintage test-gear paradox' which applies to things like scopes and signal-generators too: you probably need a modern one as a service-tool in order to keep the vintage one working properly!

Well said It's annoying when you have to fix the test gear before you can fix the radio!

[wd40addict]

Remember also that classic VTVMs normally have a 1M resistor in the probe used during DC voltage measurements. This allows you to probe a circuit for DC levels without much disturbance of its AC operation.

This was originally brought home to me when working on my Volksempfänger where measuring anode voltages etc using a DVM caused huge bangs from the LS due to the particularly high impedances in use. I dug out my Heathkit VTVM and all was good. The Heathkit remains on my testbench to this day for live DC poking, even with transistor gear. Obviously its accuracy isn't fantastic, but that's not necessary for 'why isn't this stage working?' type fault finding.

[Peter.N.]

I used analogue meters until about 30-40 years ago when digital meters first came out. I bought a couple of early digital meters but they didn't last long, very easy to blow up. I eventually spent a lot of money (then) on a fluke 11, I still use it now and its fine except for the display degrading. It even survives measuring the resistance of the mains!

Peter

[Bazz4CQJ]

Originally Posted by G6Tanuki

This is a good example of the 'vintage test-gear paradox' which applies to things like scopes and signal-generators too: you probably need a modern one as a service-tool in order to keep the vintage one working properly!

And if proof was needed, and with an EA113 too, see https://www.vintage-radio.net/forum/...d.php?t=170236

B

[avocollector]

I understand the common input resistance for DMM's is 10 meg or more whereas the analogue ones I have are mostly 20k per volt (although my first one as a schoolkid in the 60's was 1k per volt like Bazz4CQJ's). The discussions of the older voltmeters often talked in terms of how this input resistance affected the DC values of any circuit (obviously the lower the resistance the more chance of lowering the Voltage readings). But the used VTVM's here seem to be very cheap used - agreed the values can be off although most have easily adjustable and accessible calibration pots unlike most DMM's plus they were originally designed to handle the valve voltages on the input side. However it's all swings and roundabouts - what works for one person etc. One plus factor for the old analogue Avo's (as if they needed any - LOL) is they were often the instruments noted to be used to measure the tests voltages in circuit diagrams.

[Radio Wrangler]

You should always have an idea of the impedances of the places you are prodding so you know how susceptible they are to loading. You should always have an idea of the load your meter or scope probe presents.

There are places where the combination doesn't matter much, there are places where it matters a lot.

The voltages on many service sheets were taken with a 20k Ohm/volt meter because the AVO 8 dominated the market and rivals were of similar impedance because it was starting to get very difficult to make meter movements more sensitive than 50uA (with calibration adjusters). VTVMs weren't common in service departments and DVMs hadn't gone into mass production, they were very expensive propositions.

So the voltages on service sheets were distorted accounting for test equipment loading.

But wait a minute... 20k Ohms/Volt isn't a fixed number. It relates to the full scale voltage of the range you're on.

Sometimes, with a classic passive, analogue meter you might choose to use a higher voltage range than would seem obvious for the voltage being measured. The meter will load the circuit less, but the pointer will not move as far, and errors reading it from the scale will be magnified. You need to know what you're doing to know whether the loading or the reading would have been the greater source of error. Service sheets didn't mention which range had been used, but it was generally a safe bet to assume it would have been the lowest one for that voltage.

We can't see electricity directly in normal circumstances, so test equipment is our only window into the electrical world. No one window shows you everything, so you have to think about how much is enough.

Classic VTVMs were usually not terribly accurate, but their value lay in their light loading. They were more expensive so they were only brought into play in special circumstances. The expense of oscilloscopes did the same.

But nowdays, test gear is cheaper than it's ever been and there's tons of it on the surplus market. As already pointed out, with classic test equipment, there is the matter of fixing it when necessary. But if you bought an affordable new digital scope or DVM from the orient, you're likely to find it very hard to fix should it go wrong. Circuit diagrams are non-existent. At least with the older HP and Tek gear there is information around. If you go the new equipment, even from those once great brands, it will often become unfixable landfill if anything goes wrong.

David

[Craig Sawyers]

Then there is the pre-DVM Fluke differential voltmeters. These are null-reading instruments, with the null setting done via a six digit Kelvin-Varley divider. They use a zener voltage reference.

When at null, no current flows, so the input impedance is in principle infinite. In practice it is limited by leakage paths to a few tens of G-ohms. And the highest sensitivity range is 1V FSD - but with a 100uV FSD null range.

So for the best in high input impedance the not very much money for one of these is quite a good bargain.

Craig

[avocollector]

The nice thing is though modern stuff even if unfixable is fairly affordable and will serve to diagnose/fix the older test gear/check it from time to time.