Selmer Treble n Bass 50, making strange noise when turning on

[Bman]

Hi,

I'm having problem with my Selmer Treble N Bass mkiii.

I have just replaced the filter cap cans (3 x 32uf+32uf), and the bias cap, and I have re biased with new resistors. Other than this the amp is completely as original bar tubes.

Since doing these caps replacements when I turn on the power from cold the amp makes this rising oscillation noise that gets higher and louder for a few seconds - quite scary sounding! It wasn't there before I did the work.

-It does it only when turning on from cold, it's ok when its warmed up
-It does it still with the first preamp tube of each channel is removed
-It doesn't do it when the next tube is removed - the one after the Echo send/return
-I have checked the solder joints around this tube and solders to the corresponding caps and resistors,
-I have swapped out the corresponding cathode bypass cap
-I have swapped out all preamp tubes
-I have since gone back and swapped out the new filter caps and bias cap one by one, to see if I bought a bad one or made an error in wiring them - same problem

-I don't have any spare EL34s to swap in and out, but wondering if this would be the next thing to try, apart from that it doesn't make the sound when that 3rd preamp tube is removed, so I'm thinking the problem is earlier on in the circuit, but running out of ideas.

Any ideas?

Video here

[pdf64]

…
I have just replaced the filter cap cans (3 x 32uf+32uf), and the bias cap, and I have re biased with new resistors…

Has the HT voltage at the reservoir cap changed much, before and after replacing the caps?

[sluckey]

Since doing these caps replacements when I turn on the power from cold the amp makes this rising oscillation noise that gets higher and louder for a few seconds - quite scary sounding! It wasn't there before I did the work.

The most likely suspect is your cap job. Maybe a bad cap. Maybe a bad connection or wiring error.

[Bman]

Has the HT voltage at the reservoir cap changed much, before and after replacing the caps?

It's very similar voltage to before, it has only changed a few volts

[Bman]

Since doing these caps replacements when I turn on the power from cold the amp makes this rising oscillation noise that gets higher and louder for a few seconds - quite scary sounding! It wasn't there before I did the work.

The most likely suspect is your cap job. Maybe a bad cap. Maybe a bad connection or wiring error.

I initially thought this was the case, but I've now switched out all the new caps one by one, including the bias cap, and it still does the same thing. I have also now changed those old 27k resistors, and it's still the same.

I'm increasingly thinking it could be one of the el34s, I just don't have any to swap in to check without buying them. I have checked for shorts between the pins, and it all looks normal, anything else I can do to test them manually?

Or any other ideas of things to try?

[bmccowan]

I suggest you post a high res pic of the cap job - someone here might spot something.
And, I do not think there is a downside to having a spare pair of power tubes - well except going without beer and pizza for a bit. You will need them someday.
What prompted the cap change? Just age?
Choke wiring still proper?

[dwinstonwood]

Disconnect the choke from the circuit, measure its resistance, and solder in a new resistor in its place that's close to the choke's resistance. At least you'll know if the choke is the problem. 

[sluckey]

You should also consider the possibility of collateral damage, ie, something that unknowingly happened while replacing caps. I still think this new problem is somehow related to your recent work on the amp.

[pdf64]

Maybe that old bias supply rectifier diode is getting dodgy? It may be a lower spec type than those for the main HT.

Whatever, my understanding is that rectifier diode quality has much improved since the early days they’re from.

I suggest to try UF4007 in there.

[Bman]

I suggest you post a high res pic of the cap job - someone here might spot something.
And, I do not think there is a downside to having a spare pair of power tubes - well except going without beer and pizza for a bit. You will need them someday.
What prompted the cap change? Just age?
Choke wiring still proper?

*Since the photo I have changed the old 27k dropping resistors, still the same issue

[Bman]

You should also consider the possibility of collateral damage, ie, something that unknowingly happened while replacing caps. I still think this new problem is somehow related to your recent work on the amp.

Yes this is what I'm thinking is most likely, something I've accidentally nudged in the process of doing replacing the caps

[Bman]

Maybe that old bias supply rectifier diode is getting dodgy? It may be a lower spec type than those for the main HT.

Whatever, my understanding is that rectifier diode quality has much improved since the early days they’re from.

I suggest to try UF4007 in there.

I don't have any UF4007 but I do have some 1N4007

[pdf64]

Yes a 1N4007 would probably be an improvement on the old bias rectifier.
Old coupling caps can get leaky too, it’s worth checking for that every time a vintage amp is on the bench. At power up there may be 500V across the caps between the LTP anodes and output valve control grids.
Lift the grid end legs of the caps and check for any V DC on them after powering up. More than a few mV eg 50 then consider replacing them.

[Bman]

Yes a 1N4007 would probably be an improvement on the old bias rectifier.
Old coupling caps can get leaky too, it’s worth checking for that every time a vintage amp is on the bench. At power up there may be 500V across the caps between the LTP anodes and output valve control grids.
Lift the grid legs and check for any V DC on them after powering up.

Thanks

[Bman]

So I seem to have stopped the oscillation!

It seems to have been an issue with the main grounding point for the power cable which is also the grounding point for one of the cap cans. When putting in that can I must have tweaked the connection for that ground point. So maybe there was some resistance in the ground path and could that have created the oscillation? Can anyone make sense of that?

I then still had a rather large buzz in the amp, but to my huge relief that turned out to be the dimmer light in the next room.

The amp now sounds great. It has some background white noise, but not a huge amount, and I may be over analysing it having spent so much time looking at potential issues.

Thanks everyone for all your help and suggestions.

[bmccowan]

the main grounding point for the power cable which is also the grounding point for one of the cap cans

I have always considered that a bad idea. Power cord ground should be its own solid connection to chassis. How it caused your issue - others here who studied electronics will know better than me.

[pdf64]

The integrity of your chassis safety earth connection seems suspect to me?
A toothed star washer between lug and chassis is required.

[Bman]

The integrity of your chassis safety earth connection seems  to me?
A toothed star washer between lug and chassis is required.

It's soldered directly to the brass control plate, I have now re-flowed that solder. I've also added an additional solder connection between the chassis and brass plate just next to it by use of a star washer/lug on one of the chassis bolts bent down to connect and solder on to the brass plate.

[bmccowan]

I think its preferable to connect that safety ground directly to the chassis. As pdf said with a bolt and toothed washer - nothing else connected at that some point. I would not trust soldering it directly to the brass plate. But others may disagree.

[Bman]

I think its preferable to connect that safety ground directly to the chassis. As pdf said with a bolt and toothed washer - nothing else connected at that some point. I would not trust soldering it directly to the brass plate. But others may disagree.

Yes it's a good point, I think attaching to the main chassis would be a good idea. I suppose I was assuming that's how selmer designed it and if I move it then I may encounter other grounding problems. But then again they weren't designed for 3 prong plugs so that must have been a new addition.

Interestingly looking at other 60s Treble n Basses mkii and mk iii, they're nearly all connected to either this point or the ground lug of the first filter cap can. All the other preamp ground attach to the brass plate too, so maybe that is the best place for eliminating noise.

[pdf64]

The amp would have been fitted with a 3 wire power cable when new, as with all such UK gear, dunno why that schematic doesn’t show it. The mkI does https://el34world.com/charts/Schematics/files/Selmer/Selmer_treblenbass_mki.pdf

With regard to how the chassis safety earth connection is implemented, consider that’s what’s acceptable practice has changed over the decades. Part of the process is the industry experts analysing what’s gone wrong when, eg, someone gets zapped.

In the long term, a brass plate in contact with a steel plates will corrode, what starts as a good connection degrades to being insulated by a layer of yuck. So using it for the safety earth ends up being a terrible idea.

There’s only a single safety critical connection in the whole amp chassis, so do it right.