The metal 6J7 tube requires no shield, it grounds via pin 1. Always a pin to watch I have a set here where HMV put B+ on pin ! for a KT66 and some Goose put a 6L6 metal in there putting 250VDC on the body.

6J7-g is internally shielded & GT is back to pin 1 shield. Philips & AWA Also Franks electron tube pages can be handy.

Message always check pin one on an octal & refrain from using it as a tag strip. NB some metals Pin one may not ne its body. and that applies to Metalised Philips as well. Beware with them a loose envelope to base can cause the earthing wire to break.


Marc

I suppose you could call it a fart but I couldn't maintain the duration.
Audio feedback is a possibility because the aerial terminal is between the 6J7 and 6F6G, and the wire from the back of the terminal runs across the back of the 6F6 socket about an inch away, on its way to the wave change switch about 4 inches away. I haven't changed any of that wiring under the chassis though. Just cleaned up the 6J7 top cap arrangement. Nevertheless will review the circuit to see if some such cap as you suggest is already there and maybe add one if it's not.

The radio came to me with a 6J7G in place, un-shielded, despite what the valve line-up label says. So continued with them as I didn't have a strong 6J7. The 6J7 was only there to make sure that the new top-cap feed would work OK with the shorter metal valve. Haven't put the 6J7G back in yet. Might do that today as I want to recheck the insulation of the shielded wire so will need to take it off the volume control.

Will check if pin one goes to the metal body of the 6J7.

If I have time - good weather means cleaning up the mess of sticks and leaves in front of the house following the windy weather.

The biggest issue in modern times is lead dress. The worst offending area is the area around the Detector / 1st AF and that includes the lack of shielded wire on volume & tone controls. Often moving & using modern caps exacerbates stability issues..

Do not overlook riveted earthing. That can become corrosive and loose and other dry joints are possible likely in the HT as well as flashovers.
If the output transformer secondary side goes open that can generate EHT & flashover. Mica caps with HT on them will create Snaps Crackles & Pops (do not pour milk on it).

My biggest surprise was a hacked STC chassis 59, where in the process of finding out where the signal got lost, with Signal Generator & Oscilloscope. There was a factory Lead Dress issue causing a secondary oscillation in the mixer? I ended up actually shielding & moving the screen wire. I also had to adjust the bias on the IF (6D6 later 6U7 nasty piece of work unshielded) to stop it overloading the plate detector.

The hacker had destroyed the bias train & lost the volume control trying to put the volume control in the grid of the output tube. An original circuit could not be found; but in that era it was never there.

I am currently creating my own sticks to pick up, albeit nature is helping. Pruning season, out with the secateurs, chainsaw, axe, tractor & trailer & into it.

I've sometimes had to move wires and components around to prevent unwanted feedback. Sometimes just looking around you can see the possibilities. Other times it can be a bit trickier.

And yes, this is the time for weeding and pruning, before it all explodes in spring.

I notice a lot of the available modern replacement capacitors are not showing the shielded end.
A real trap for those that venture into the habit of throwing out all caps before proper appraisal and repair of radio first.
Best is to find and mark these new capacitor ends before creating more man made faults.
JJ

I notice a lot of the available modern replacement capacitors are not showing the shielded end.

I haven't seen that marked on caps for many years. I'm reminded of that any time I come across old ones that are marked, such as in the tone generator of my 1963 Hammond organ.

It would appear that outside foil was not an accident and was marked that way as they had found out that perhaps ensuring that the foil to be on the outside, totally lapped the outside & did work as a shield. Not so with modern ones.

In a radio with RF and two IFs as well as mixer and separate detector and two audio plenty of room for lead dress problems. There are wires going all over the place. But overall have not had oscillation problems.

The big changes to the wiring have been to the screen bias arrangement, but haven't created a problem.

Will check out riveted and screwed earthing - 5 valve shields altogether with bases riveted to the chassis. Would have to buy a rivet set to make a proper job of it if a fix needed. The chassis is in four parts - main body, two ends self-tap screwed on (couldn't remove two of the twelve screws for the paint job), and a grommet isolated plate carrying the tuning gang, RF, mixer and detector valves, and aerial, RF and oscillator transformers. This plate is connected to the main body with metal braid in four places - two were broken. Have replaced one but the other hard to get to. Have been considering using braid to connect the end plates as well for electrical safety.

A curiosity is that the 6A8 second grid (oscillator plate) is supposed to be fed from the rectifier side of the HT choke according to the AORSM circuit, but is from the other, fully filtered, HT end. The supposed arrangement is to keep the oscillator plate voltage isolated from output valve load effect on HT voltage, and keep oscillator frequency constant. Radiotron Handbook has a piece on this subject. Haven't tried to change it as the 831, a somewhat similar circuit but with 6K8, also feeds from HT.

Re caps marked with end connected to outside foil, what is the best way of determining this in unmarked caps?

Re pruning, in my case it is nature's work as have a 50 foot lilil-pili tree on the street which sheds live and dead twigs and leaves in strong winds. Could block up drains and flood the garage. Worse when its is blossoming (sticky flowers), and fruiting (can be a green bin full in a good season). No good for jam unfortunately as they are green. Nothing like a good lili-pili jam or jelly.

It would appear that outside foil was not an accident

It was deliberate.

Re caps marked with end connected to outside foil, what is the best way of determining this in unmarked caps?

There's a lot about that on the web, such as this: https://www.youtube.com/watch?v=tFa6XJccvTE

Some people have gone to the trouble of creating testers.

There's also a lot of discussion as to whether it's necessary with modern caps.

Take a couple of turns of hookup wire around cap, feed to oscilloscope.
Inject RF of a suitable frequency(experiment for best positive results), into each leg of capacitor.
Highest amplitude on the scope relates to outside foil.
JJ

Pity its not something easy like have the writing the right way up and it is the left hand end, or right hand end.

I wonder at some of that all six Pentagrids shown in the Radiotron Handbook I have their power sourced form the main B+ that is understandable as any hum in the system can be modulated by the Pentagrids. However 6A8 is actually the most unstable of the Pentagrids and the only time you would change it is if you get fluttering.

The only sets I have seen anything like that were HMV 661,666, and a few others. They fed DC to the screens from the filaments and to the 6V6. You can feed dirtier DC to 6V6 Plate as the plate does not amplify.

Leads should be kept close to the chassis as possible and leads in free space can be antennas & radiators. To really get an idea of rivet conductivity you need a bonding meter; However, there should be no voltage between the actual node & chassis. Don't pop rivet with valves in situ. Astor on Aluminium chasses ran a dedicated ground wire to riveted tag strips & earthed to a common node.

I would not stress too much about end caps its the transformer & what it and the mains are mounted too that will become alive.

OK thanks for that Marcc. As I said the radio has not given oscillation problems generally, only while fiddling under the chassis, so am not inclined to fix things that ain't broke.

Re the 6A8 second grid feed; I have Vol 4 of RDH. The comment is in Ch 35.3 (vi) B on page 1245 para 5 of my sixth impression (1963). This recommendation was remarkable in that it was included in the official circuit, but not implemented in my radio - so made a note in the radio's notebook.

I'll resistance check all valve shield base rivets and wriggle them around and see if there are any problems. But the main issue there will probably be the connection between the aluminium shields and their bases. The shield bases are made either of tinplate gone brown or chromated or phosphated steel and got paint on them despite my best masking efforts. Will resistance check these connections too.

Re the end caps; while I have a megga and the tranny checks out OK, in the end I will ask an electrician to put a tag on its power cord so trying to eliminate all causes for a fail. For the same reason might put a back on the cabinet as an attempt at double insulation.

Re caps and which end has the outside foil connected, a solution could be to make your own shielded caps eg
https://www.antiqueradios.com/forums/viewtopic.php?f=6&t=324967

Haven't done anything on the radio - the cat reported in sick or injured. It is a street cat that has various servants up and down the street and nominally belongs to a disadvantaged school. Turns out it has congestive heart failure. Has gone off in high dudgeon for being taken to the vet.

In 50+ years I've never had an issue with outside foil orientation. And I must have replaced many thousands of caps in radios and TVs and supervised apprentices replacing just as many.

That being said, Philips Mustard caps have a kink in one lead next to the body of the cap. Anyone know what that means?

I would suspect that early caps had higher ESRs and of course they were larger, hence they had a larger outer foil area. Hence it mattered.

Yep, I have to say that in replacing these old paper caps with the black line, I have never had a problem either.