A progress report on this speaker after intermittent work on it.

The first problem was a resonance arrising from the outer corrugations. This was tracked down by pressing gently around the corrugations. It seems that in places the PVA used did not prevent resonance. So I lifted the corrugations (soak the glued area in alcohol) and reglued with rubber glue. This gave an improvement but not a complete solution. So I forced the cone outward a fair amount and ran a bead of rubber glue from a syringe between the cone suspensionand the basket. With this dried and the spacers forcing the cone out removed, the resonance was gone.

The second problem was that after a while another resonance developed audible as a buzz in time with bass notes, and most audible with tone turned to the bass end. A feature of this was that it developed after about a minute after turning the radio on from cold. My worst fear was that this was due to the voice coil heating up and expanding a bit, explaining the delay. There was no poling though, and it was not effected by manipulating the cone in any way.

By feeding in a signal at the pick-up the problem was isolated to the audio end. So looked for poor connections that might cause a thermal delay. Resoldered the valve connections for the preamplifier and output valves to no effect. Then changed out the output 6F6G - success!

This particulat 6F6G has a tiny piece of glass rattling about inside it indicating misadventure at some time in its life, but despite that has worked OK. I resoldered one suspicious looking valve pin, then all of the others without success.

The next step would be to completely remove the valve base and look for a cause there. But before doing that I thought I will put it to the forum - has anyone else had a similar problem with a 6F6G, characterised by a buzz with a time delay, or with any other output valve for that matter.

Edit: On second thoughts a better term than buzz, would be fizz. That is, higher frequency than 50Hz line noise. An odd thing was that this noise was barely visible on a CRO - I would have thought it should be a distinct "grass" on top of the audio, but no, just a barely discernible "wriggle".

There is always the possibility that one of the tubes is microphonic. 6AV6 is one that does that and is also psychic. If there is a fault on the chassis, it will often exhibit microphonics.

Tap the tubes & see if it rings, if it does try a rubber band around it unless its a hot tube.

Tapping the valve with the usual chopstick did make the fizz go away (sorry for this highly technical term). But what goeth away cometh back again on starting up from cold. But it does confirm something going on there.

There was no ring or other effects through the speaker.

Given the delay on starting from cold, my guesses are some conduction between heater and cathode on warming up, or the screen warming up and doing something. Whatever, the valve is effectively a dud.

Usual reasons are a loose grid wire, a bad weld or a mica flaw.

Microphonic output bottles? The 1Q5 battery OP bottle can be so bad that you get acoustic feedback from the speaker.

Now this is a point where it must be appreciated that Buzz and Hum are two very distinct animals.

Buzz can be a Sawtooth waveform generated by a charge discharge situation, which could be internal. Buzz on the other hand can be external.

Switch mode power supplies are by far, the worst offenders and I have had a FAX 20m away get into a 175kHz IF. And with CFL,s & such it can get in via the mains supply.


I often resort to an Oscilloscope to find the origin, if its in a set, Do note Plate resistors and those on the grid of power output tubes, have an attrition rate. Plate bypass caps on output tubes also have an attrition rate. Underrated voltage components do fail.

Ian: "Usual reasons are a loose grid wire, a bad weld or a mica flaw."

These faults are inside the envelope. I can only fix someting inside the base, which can only be a crook solder joint. I have redone all of these and that leaves taking the base off which I haven't felt like doing since I have a spare 6F6 and 6F6G.
Re microphonics, there is no sign of it when tapping.

Marrc: Re visuallising the waveform of the fizz, I will try a single frequency from a generator and see what is on top of it. Can't really see using broadcast waveforms, just a bit of a wriggle that can't be resolved very well.

The time delay eliminates external causes I think.

The plate resistors are original so worth a look given the time delay indicating a fault when something warms up. But I doubt this cause given the fault disappears with a 6F6G change.

Now that I have a speaker that works well I will look at writing up the rebuild of this speaker.

There may have been a corrosion, or contact issue with the 6F6 socket?

There is risk with swapping glass tubes and metal and I have had a live metal jacket on a 6L6 metal, as the KT66 is glass. The factory wiring put B+ on pin one. It was not designed for metals and that is a trap. Pin one is generally (check) the body of metal tubes and Philips metalized octal tubes. In this situation the body pin is grounded.

I haven't got the metal 6F6 so haven't had to check that out.
The only original metal valve is the 6J7 which is OK. Run a metal 5Z4 with the chassis upended as the rectifier is vulnerable on the corner of chassis. B hot though.

No fizz with another 6F6G or with a 6V6GT so I don't think that it is a socket issue. Worked them all over early on as part of the restoration anyway.

Have you tried resoldering all the pins of the 6F6.
Fairly common dry joint occurrence on octal valves.
JJ

Yes I have resoldered them all; all I can do now is take the base off in case the resoldering is stuffed up somehow, or there is some sought of fault in the wires comeing out of the glass.

If the balsam looses its grip, then you can get corrosion of the wires in both base and top caps. Top caps being the easier to repair.

Funniest effort seen was turning a set over & having the envelope fall out of its base (was posted at the time). One of those killer faults staring you in the face.