MikeyDee's Sussex build diary

[Radio Wrangler]

The lack of beads will not be a problem with some valves, but it will with others, so having a couple of valves OK doesn't show whether you're going to need them.

You are going to be biasing valves up into the optimum part of their transfer function, where the gain is highest. Some of them have gain to hundreds of MHz and will regard wiring as jolly nice resonant circuits.

The time to stick the beads in is before you go any further. Oscillation affects readings and usually gets blamed on duff valves. You might not recognise it when it bites. I recommend a preemptive strike while the enemy isn't looking.

David

[mikeydee]

Thanks Radio Wrangle - let me just to check that I have the understanding right:

I have one wire coming from each pin selector and going to the A08 connector and then daisy chaining to all the valve sockets.

would it be best practice to put one bead on next to each valve socket pin so that would mean that the daisy chain wires will have two beads one on each end for example.

Ok, no problem it's just a wiring change I wonder if there is there any way I can see if I have a problem before and after? Measure the improvement for example.

[Radio Wrangler]

Easy!

Put a bead on each wire from each pin of each valveholder. You're trying to stop the valve seeing the wiring.

Putting beads only after the selector switches leaves all the daisy-chaining wires and the switchess for a frisky valve to play with.

Cheers
David

[mikeydee]

Ok, I'll give it a shot.

Thanks, Mike

[ColinTheAmpMan1]

Originally Posted by mikeydee

Ok, I'll give it a shot.

Thanks, Mike

Mike,

You might take a look at the conversations between myself and "Dekatron" (Martin) in the original Sussex Homebrew thread. Martin is a really good source of info on valve testers and his advice on putting the valve-holders in a loop of wiring, keeping these loops as equal in length as possible and putting ferrites on the wires going to the pin-selector switches as well as one on each wire going to the valve-bases is what I will use when I get to making my "Sussex VT1".

Don't forget that many valves used in audio circuits (eg ECC83, EF86) have high amplification factors and can turn into oscillators all too easily.

Colin.

[mikeydee]

Colin, I have been off the sussex chain for a bit but back on it now.
I have studied those AVO patents now and it's interesting stuff. I think I will do this in a number of stages just to save messing up my wiring and hopefully learn something . . .

1. Add the equal loop and a series 100R resistor - hopefully observe at least some improvement.
2. If there is still an issue then perhaps a ferrite bead from the switch to
and if that fails....
2. start adding a beads close to each socket, and insulated with heat shrink.

[mikeydee]

Sussex debugging and testing.

I had some problems with my Sussex which some suggested could be caused by parasitic oscillations. I have added ferrite beads to the selector switches and the valve socket pins or at least the first two sockets.

Below is a scope trace of the anode and you can see a slight flicker on the waveform.

Also you can see a picture of the modified circuit board - please refer to the main Sussex thread for discussion and details on this.

[mikeydee]

My Sussex build is complete here are the final pictures.

For me this was my biggest homebrew project to date and I have learnt such a lot from the forum and Mike Rowe's original fantastic design.

MY personal lessons were:

• Print the panel before cutting. I did it the other way round which made it very difficult to align things.
• The case that I used was ok in the end. Initially I regretted buying it as it's a bit flimsy but now that the base is lined with Aluminium actually it's pretty solid.
• Colour code all the wiring without it debugging is tough.
• The D connectors worked very well
• Don't use maplin choc a bloc the screws are made of very soft metal.
• Stop faffing about and get on with it . . .