Oscillator Grid Current, ECH81,ECH35 etc.

[ms660]

I've never tried measuring the oscillators grid current, just wondering if anyone on here has and did the result more or less tally up with the typical specs in the valve data?

Lawrence.

[John M0GLN]

I'd like to know how as well, if I go anywhere near an oscillator with any sort of test equipment it just stops.

John

[ukcol]

Originally Posted by John M0GLN

I'd like to know how as well.......

I would think if you connected your measuring instrument in series with the bottom end of the grid resistor that may be successful.

[G8HQP Dave]

You can measure the average grid current (which may be what the datasheet means by 'grid current') by measuring the average voltage across the grid resistor. To do this take a 10M input DMM and add a 1M resistor to the probe tip (or tack solder the resistor to the grid pin tag). This provides a low pass filter with stray capacitance. Measure the voltage, and add 10%. Use Ohm's Law.

[kalee20]

That's how I'd do it, too.

Grid current is a messy thing to measure directly - you'd have to unsolder the earthy end of the grid resistor. But with the measuring equipment commonly available in the era, it was the best indication of oscillation amplitude, which is the thing that really matters.

[ms660]

Yes, that's what I was meaning, unsolder earthy end of the grid resistor and measure the current flowing through it, wondering if it would more or less tally up with what's given in the valve data or is there more to it?

Lawrence.

[G6Tanuki]

I generally dislike 'doing anything' to a radio in order to measure the local-oscillator.

Wasn't it Heisenberg, or perhaps Schrodinger? who said "to Observe is to Interfere" ??

Removing shielding and hooking scope- or frequency-counter-probes up to the grid/anode of an oscillator can pull it significantly off-frequency, add enough loading to stall the oscillator entirely, or introduce enough external hum to make you think the LO is nastily-unstable.

My preferred approach is to use another receiver to listen for the local-oscillator of the receiver-under-test. This way you can get to know whether it's properly on-frequency in the circuit-as-it-will-be-expected-to-work-in [with all the shielding-cans fitted and the radio in its normal case] and also get a good feel for its purity/frequency-stability.

[G8HQP Dave]

The grid current in an oscillator is usually short pulses, so if you want to measure it you need to insert a small resistor below the grid resistor and display the resulting voltage on a scope. Or you can measure the average grid current, which is usually what the datasheet actually specifies. If it is within a factor of 2 either way then I would be happy.

[kalee20]

Yes you can use another radio as a 'sniffer' to check the LO is running, on-frequency, etc.

I've generally believed that oscillator grid current is indicated by valve manufacturers as a guide to optimum amplitude, from a frequency-changer design point of view. Conversion conductance rises as oscillator amplitude (grid current) rises, then levels off. Bigger amplitude has little effect, but increases the susceptibility to spurious responses.

It's worth noting that even if the oscillator is super-low-distortion sine wave, the effect on the mixer is to drive it into non-linearity, so harmonic mixing can easily take place if the RF selectivity is lacking from aerial up to signal grid.

So, from a service point of view, there's little benefit in measuring grid current - you don't know how competent the designer was; or if he was just pressurised to get the thing into production regardless of how optimum the front-end is. Sub-optimal grid current may not be a dying valve, or a lossy coil, it could be just inadequate design. Many domestic radios are rather lacking here, in fact!

[Skywave]

In years gone by, when Amateur Radio on 2m. A.M. was popular, it was common practice to fit a 1kΩ resistor between the 'cold end' of the grid leak and chassis of a class-C freq. multiplier, the 1kΩ being adequately by-passed with a capacitor. That way, a d.c. voltmeter could be used to measure the voltage across that resistor and thus calculate the grid current.

Al.

[stuarth]

Re post #4 - Heathkit were onto this 50 years ago, their V7-A valve voltmeter came with a probe with a 1Meg series resistor in the probe tip and a screened lead to the instrument.

Stuart

[Al (astral highway)]

Originally Posted by ms660

I've never tried measuring the oscillators grid current, just wondering if anyone on here has and did the result more or less tally up with the typical specs in the valve data?

Interesting Q, Lawrence. Just curious, why would you need to know?

I measured it in a Nuvistor mixer that I built once and it was around 12- 15uA, much to my astonishment!

[ms660]

Just curious to know if anyone has, in effect that's how the old style GDO's worked.

Lawrence.