RF valve wiring

[mpegjohn]

I am building a SW valve receiver.
I have realized how critical component placement is with respect to grounding decoupling caps.

I have put solder tags on the mounting bolts for each valve. However, if I try to use these then invariably there will be a resistor or cap that crosses over from one side of the valve socket to the common input ground tag.
This looks ugly.

Does anyone have a few photos of good layout with regards IF stages, etc, that I can get a few ideas on layout from.

many thanks,
John

[Sideband]

For a really neat job, you should use the valveholder centre spigot. That is what it is for!

To give you an idea, here is a picture of an IF amplifier valve (in this case an EF80) that I wired last year when I was involved in a beginners project (a simple superhet).

Full use has been made of the centre spigot to provide neat direct and short chassis returns.

Or use a solderable ground plane, 'tin' from biscuit tins is good, make it as a sub sheet to the chassis and screw through both, copper clad PCB does the same job (and can be used as the chassis too, simply solder on some sides of the same stuff).

[Skywave]

The lower the frequency of operation, the less critical the component layout is and the accompanying lead length of most components. In my experience, at 465 kHz, it is often surprising what liberties one can get away with when it comes to component placement. Component layout usually becomes 'ugly' at frequencies above 20 ~ 25 MHz or so, and especially at VHF and up, where even a half-inch of component lead can introduce significant reactance. However, in all such builds, the two features that really matter are the use of a common ground-point for that stage and plenty of inter-stage screening, especially taking care that the 'output' and 'input' of any stage cannot 'see' each other.

Al.

[G6Tanuki]

As Skywave has said - unless you're dealing with above-30MHz stuff it's a fool's errand to get uptight about lead-lengths.

Once you're into the 30+ MHz range things are different - you may for example find a specific brand/pattern of decoupling capacitor specified in a circuit - because the designer has matched the natural self-resonant series-characteristic of that particular brand of capacitor to provide optimal decoupling of an otherwise-unwanted feedback path in her original design.

Swap-in an identical-capacitance capacitor of a different brand, whose natural self-resonance is different - and you can have problems!

[ms660]

Sometimes a poorly laid out relatively low RF frequency amplifier stage can have the propensity to turn into a HF/VHF oscillator.

Lawrence.

[G6Tanuki]

Originally Posted by pwdrive

Sometimes a poorly laid out relatively low RF frequency amplifier stage can have the propensity to turn into a HF/VHF oscillator.

And so can an audio-amplifier !!

I have a philosophy that "Any amplifier is just an oscillator that's not yet managed to achieve it's full potential".

With 'high-slope' tetrodes and pentodes you need to 'think RF' and provide really-good decoupling - include 'parasitic-stopper' resistors/chokes in control-grid and anode connections. Otherwise you run the risk of your 10-Watt push-pull EL84 audio amplifier 'going ultrasonic' and dedicating 9 of its 10 Watts at destroying the insulation of the output transformer.

[kalee20]

Never mind audio amplifiers, even stabilised power supplies - which you only want to deal with zero frequency DC - can oscillate at anything from AF to VHF.

Basically, the rule is, adopt a layout consistent with the frequency you don't want it to oscillate at.

[Skywave]

A design engineer with whom I used to work with many years ago was fond of one motto:
"You can never have too much decoupling".

With the design of any item of electronic equipment, the inexperienced easily fall into the trap of thinking that it is only the components that are physically present on the circuit diagram which will determine the performance of that item. Experience teaches us to look for the hidden and invisible components - R, L and C (and occasionally temperature) - and their effective impact on that design. Experience is a hard teacher: it gives the exam first; the tuition follows later.

Al.

[Radio Wrangler]

The standard of construction has to handle all the frequencies you want it to work at AND all the frequencies it could get into trouble at.

If you are using VHF capable devices, then you need that sort of care to ensure they stay stable even if you're using them at DC or audio.

Spurious oscillation which never gets seen on a multimeter or on a low frequency oscilloscope is probably the cause of a lot of things which went bump in the night and circuits which seemed disappointing.

This isn't too bad with valves unless you start trying UHF double triodes in audiophile stuff, but it's a major risk now all small signal transistors have Ft's offering gain to uhf or low microwave frequencies.

David

Ferrite beads are very useful, put on most leads to a valve and they behave.