Marshall DSL 50 channel switch issue

GazLion

Hi all, after a bit of advice please. I've just picked up a used Marshall DSL 50, but I'm getting very loud popping noises when switching between the 2 channels. 

It isn't the footswich as it does it when switching via the button on the amp also. Interestingly however, it doesn't do it when the 'tone shift' button is engaged. With that pressed in, it switches perfectly. 

The guy who I bought it off said that he hasn't used the amp for a few years, and it hasn't been serviced in a while. 

I thought I'd ask for any suggestions on here before booking it in to an amp tech.  

Cheers 

ICBM

Try changing the V3 preamp valve for a known good one, preferably not Russian-made (Sovtek, EH, or Tung-Sol) or JJ. V3 is the tone stack driver and is a cathode-follower - these are prone to failure with Russian-made valves and also sometimes JJs. (V1 is under the metal cover, then V2 and V3.)

If you don’t have a spare valve then just swap V2 and V3.

If that doesn’t help then it could be a faulty cap in the tone stack - which will be bypassed by the tone shift - this could cause a DC voltage offset which will definitely make a loud pop. I’ll have a look at the schematic later and see if I can spot which one it’s likely to be.

It’s going to be a tech job if it’s not a valve anyway though.

ecc83

Does the amp show any evidence of having been in a damp garage? If so you might have a very slight "leak" across a PCB (I assume the amp have one) .

I had this problem with a Mersey Super 15, EL84s would slowly go to redplate. Fixed it by soaking the PCB in most of a can of WD-40 and leaving it to drip dry for a week. You have to get the PCB out of the chassis so as IC says, tech job.

Dave.

ICBM

A leak across a PCB is a definite possibility with a Marshall DSL! They're notorious for it...

But in this case, I would definitely NOT do that with WD40 - the PCB is so poorly designed and prone to arcing in various places anyway that I would not risk leaving any potentially conductive residue on it.

Since it's stopped by using the Tone Shift switch, I think it's more likely to be a cap. Now I've looked at the schematic, it should be C20 (the treble cap in the tone stack) since engaging the Shift puts C19 in series with it, which would block any DC leak.

GazLion

Hi guys, thanks for the replies. The amp was apparently stored in the house, inside a flight case (which came with the amp!), and for what it's worth doesn't smell damp or anything. 

I've just pulled the back off the amp. @ICBM going from left to right, do you suggest I swap round the 2 middle valves then test? 


https://imgur.com/mV4eYrK

https://imgur.com/95pS7Ov



Hope the pictures work? Looks like I have a mix of JJ's and also one that just says made in China? 

ICBM

GazLion said:

I've just pulled the back off the amp. @ICBM going from left to right, do you suggest I swap round the 2 middle valves then test?

Yes, but going by the schematic I doubt that it will do anything, I think the leaking cap is far more likely to be it. This cap is fairly prone to failure since it's under the full DC voltage from the cathode follower. Older types were very prone to moisture absorbtion which would cause it, but modern ones can do too.

GazLion

Thanks @ICBM , sorry just one more question! I don't know if you saw that I edited the above post to link to a couple of pics. One of the valves seems to have a non branded 'China' stamp, and the others are JJ's. Ideally should these valves all be the same brand? Will this cause an issue? 

I appreciate the help - as you can probably tell i'm not too hot on this stuff! 

Thanks 

ecc83

ICBM said:

A leak across a PCB is a definite possibility with a Marshall DSL! They're notorious for it...

But in this case, I would definitely NOT do that with WD40 - the PCB is so poorly designed and prone to arcing in various places anyway that I would not risk leaving any potentially conductive residue on it.

Since it's stopped by using the Tone Shift switch, I think it's more likely to be a cap. Now I've looked at the schematic, it should be C20 (the treble cap in the tone stack) since engaging the Shift puts C19 in series with it, which would block any DC leak.

Ah! Bit of misinformation about WD-40 methinks? Since the product has been used on car electrics for decades, especially distributor caps,  there is obviously no "conductive residue" left behind otherwise the 25kV+ would surely track?  I grew up with crappy British engine electrics and WD was a constant winter companion and before that its predecessor "CRC Rocket fuel".

In the case of the MS 15 the fluid washed off any grease and crap and soaked into the paxolin and drove out any moisture. I was then very careful to let it dry very, very well.

Dave.

ICBM

GazLion said:

One of the valves seems to have a non branded 'China' stamp, and the others are JJ's. Ideally should these valves all be the same brand? Will this cause an issue?

No - preamp valves operate independently and there's no advantage to having a 'set' really. In fact, the Chinese valve should be in V3 anyway even if it's not a valve fault causing this problem since JJs have a higher than average failure rate in cathode-followers in my experience, whereas Chinese ones don't.

Swap these two first and then leave them like that even if it doesn't fix it.

ICBM

ecc83 said:

Ah! Bit of misinformation about WD-40 methinks? Since the product has been used on car electrics for decades, especially distributor caps,  there is obviously no "conductive residue" left behind otherwise the 25kV+ would surely track?  I grew up with crappy British engine electrics and WD was a constant winter companion and before that its predecessor "CRC Rocket fuel".

Car HT electrics are properly insulated. When you see some of the common arc points on a DSL circuit board - caused by unbelievably stupid trace design - you wouldn't risk it. Even if it traps a speck of dust it could make all the difference.

Have a look at this - admittedly this is a MkI board and they did later change it to increase the separation slightly here, but they left a couple of other similar ones...



If you look right in the middle of the picture you can see the point on the standoff leg of the 1K 5W resistor is less than half a millimetre from the large PCB trace. The resistor is a screen resistor at +470V and the trace is the filament supply at 0V. Guess what happens!

It's quite unbelievable when you see that the shape of the trace is totally unnecessary, if they'd just done it at a 45º angle like the bit under the fuseholder the separation would have been about 5mm.

ecc83

ICBM said:

ecc83 said:

Ah! Bit of misinformation about WD-40 methinks? Since the product has been used on car electrics for decades, especially distributor caps,  there is obviously no "conductive residue" left behind otherwise the 25kV+ would surely track?  I grew up with crappy British engine electrics and WD was a constant winter companion and before that its predecessor "CRC Rocket fuel".

Car HT electrics are properly insulated. When you see some of the common arc points on a DSL circuit board - caused by unbelievably stupid trace design - you wouldn't risk it. Even if it traps a speck of dust it could make all the difference.

Have a look at this - admittedly this is a MkI board and they did later change it to increase the separation slightly here, but they left a couple of other similar ones...



If you look right in the middle of the picture you can see the point on the standoff leg of the 1K 5W resistor is less than half a millimetre from the large PCB trace. The resistor is a screen resistor at +470V and the trace is the filament supply at 0V. Guess what happens!

It's quite unbelievable when you see that the shape of the trace is totally unnecessary, if they'd just done it at a 45º angle like the bit under the fuseholder the separation would have been about 5mm.

"Gaps"  are not PCB substrate IC. I thought you meant WD-40 would lower the resistance of the actual board? Plainly it does not do that else dis caps would not work after spraying!

Yes, anything in that gap save PTFE could cause an arc which is why I was at great pains to point out the importance of drying the PCB really well.

BTW,  PCBs should have an insulating  coating that prevent such arcs but, as we both know it can be dangerous to rely on the coating alone! Or indeed a similar insulating coating on resistor bodies.

Dave.

ICBM

ecc83 said:

"Gaps"  are not PCB substrate IC. I thought you meant WD-40 would lower the resistance of the actual board? Plainly it does not do that else dis caps would not work after spraying!

No, nothing to do with it getting into the board. But I'm sure you can see now why I would very much not recommend just spraying WD-40 into one of these amps... even if the slightly oily coating it leaves couldn't be conductive by itself, which I'm not convinced about under that sort of voltage/distance condition.

Terrible piece of design and I assume you can now see why I detest them! Presumably so do Marshall now, they must have replaced hundreds of the boards under warranty. This is far from the only design fault, or even board layout fault, in these amps too - some of them are really shocking for such a major manufacturer with a long history of valve amp production, it's just down to basic ignorance in my opinion.

GazLion

Swapped the valves and unfortunately the issue still persists. Can anyone recommend any amp repairers in the Bath/Bristol area? Cheers