In setting about replacing the hollow rivet securing a 6A8 shield base and valve socket with a screw to fix poor shield earthing, I discovered that there is a piece of wire running from the tag to the rivet and soldered to the rivet. Puzzling since No 1 pin is not connected internally. The same for 6U7 and 6H6.
Any ideas?

Maybe wired to accept a metal envelope if fitted?
Fred.

A logical solution to my question. The only metal valve specified was the 6J7 1st audio. All the rest are G. But it would make sense to make metal valves usable in the radio.

Am presently having trouble with oscillation after opening up the IF transformers to fix loose internals and tight trimmer adjustment screws. If fixing this poor shielding doesn't fix the problem may start another thread. Have already started to write it up but thought I should fix the bleeding obvious first.

The circuit diagram is a guide to what should be in the hole. 6A8 with no suffix would to me indicate metal & there would be no shields & normally (not always) pin 1 wired to ground. There is often an earth symbol on the tube body in a diagram, to signify grounded body.

Oscillation is common with new wiring in the area of the second detector. Most modern caps are not "outside foil" and in the area around the second detector along with rework can cause havoc. That also applies to 6U7 which, if un-shielded will also cause havoc, manifesting as oscillation.

Sorry, it is actually a 6A8G specified and fitted. The 6A8G is very microphonic when oscillating so this is getting the first attention to shielding which I hadn't noticed was kilo-ohms to earth after repainting the chassis. Paint has got under the rivets. Other shields are a few to 10 or so ohms to earth. No new wiring in this area since 1989.

The 0.25μF cathode bypass was replaced by a bluecap in 1989 and these have no indication of outside foil so this is a candidate for investigation. At least shorten the lead length.

Will replace at least one rivet with nut screw and star washers and see what happens. Not keen to use a pop rivet in case it crushes the paxolin valve base, though it would allow the shield to fit right up against the chassis top and so look neater.

Pin 1 was often used with a metal strip that came up the side of the base and grounded (hopefully!) the Goat shield.

To be effective pin1 has to be grounded for the tab that earths the Goat shield: It and the entire Goat shield is subject to corrosion. It is often a case of treating them and painting them after adding a a slack wire to each half & soldering the tab to one plate.

You can solder a wire from the top of pin one to a plate of the Goat shield, or Goat shield to an earthing tap on the chassis: If the tab is missing, or you cannot make a new one? Goat shields can be an issue with 6D6

It has never been unusual here for wire to be ripped out & replaced with shielded. Especially with tone & volume wires run long distances & the odd wire laden with RF going through the audio radiating; or if away from the chassis either radiating or acting as an antenna. Often the det audio tube needs shielding.

Microphonics are common if there is a fault. Do check jointing. Often if there is feedback oscillation a chopstick used to probe, can change the frequency of the oscillation in the offending area.

As noted 6D6 / 6U7 socket different are the worst tube for picking up anything and using it to destabilise & cause havoc. If its radiation just a hand near it, will change the oscillation.

All caps are not equal. The yellow axials, as noted, are not outside foil and can be subject to capacitive induction as are parallel wires. Some disk ceramics are also prone to induction. Often the only answer is to form a jacket around it. I often use brass sheet /strip from a hobby shop.

Beware some octals do not use pin one as the body / shield.

The shields are pressed aluminium cans mounted on a base. Have put a 1/8inch brass screw and nut with star washer in place of the rivet opposite pin 1 to test if the shielding can be fixed - it can. 0.1ohm resistance now from aluminium can to earth. But no change in oscillation, so that shield is not the cause.

The rivet at pin 1 still has a high resistance to earth so that will have to be another screw as well, though it is much harder to get to. Will have to wait until the oscillation is fixed as I have a 6A8g in place.

Will now look at other imperfect shielding of 6U7s. If no-go will write up the full symptoms in another thread.

I am currently reading No 23 "How to fix a receiver that squeals and motorboats", Radio Servicing Methods.

https://archive.org/details/NRI_Radio_Servicing_Course_1949/No.%2023%20-%20How%20To%20Fix%20a%20Receiver%20That%20Squeals%20and%20Motorboats

A good mixture of theory and practical method. But the four basic symptoms given do not fit my case.

Re your comments Marcc, poking around The 6A8G wiring while oscillating, with a knitting needle, just having the needle adjacent the wiring in this area changes the oscillation. No need to touch or poke anything, but if you do any wire moved changes the whistle tone. So the 6A8G has been rendered extremely sensitive.

Quite happy to use shielded wire especially for tone control which reaches right across the chassis.

Don't give it any hard thought yet until and unless I start another thread.

It is likely there is a supersonic oscillation, that can happen with shielding that is not grounding and also with a resonant circuit where one of the elements of the resonant circuit has departed company with the other & you can end up with a "transitron" oscillator, but basically one with no control.

I would check jointing and the oscillator gang for either a short miswired, or wrong component.

How did you mark the IF cans to ensure that they were rewired as they were?

The 2nd IF transformer had two frayed internal leads. I fixed one then the other, so no error there I think, especially no unmade connections. All continuous.

The tuning gang was taken out for restoration including running 1000 grit wet and dry in the three shaft connections, cleaned and lubricated with a smidgen electronic lubricant. When refitted, everything good, tracking perfect. Despite that I have had a poke around with the knitting needle and no joy.

I am still attending to the shield earthing of the RF, mixer, IF and detector valves. Bottom of the drilled out rivet holes difficult to get to and especially difficult to fit anti-shake washers and nuts to the 3mm screws. Will work through each valve base and remake all soldered connections.

You can still use the toothed washers, but with that degree of awkward I take out the tubes & use blind rivets. You can get Aluminium, Steel, Stainless Steel & Copper: Copper solders well.

Bluetack on the end of a chopstick well get the nut through the maze & onto the metal thread.

Thanks for the bluetack tip - have been using a magnetised screwdriver for the washers, and screw nuts onto the end of a knitting needle if a tight spot, or if plenty of room, a nut driver stuffed with tissue to bring the nut to front of the socket.

I had thought of blind rivets but the fragility of paxolin put me off. The thought of pulling the nail of a pop-rivet through paxolin seemed a bit risky.

I use blind rivets with a washer slipped over the end.

Yes “pop” rivets along with heat shrink tubing, roof and gutter silicon and cable ties, just to mention some are on my banned list for vintage radio.