Indicator Lamp Dimming?

[revolver1]

Hey guy's, I noticed recently that my indicator lamp was dimming ever so slightly when I hit the strings hard or if using a boost pedal. In the past few days it's become a little more noticeable. Then yesterday I had been playing for a couple of hours, I turned my tube screamer on and hit a note then the lamp went out completely and then came back on. It did it a couple of times, the sound was fine I stopped for a moment then went to play again and the volume faded away (a bit up and down) Then the sound came back. At this point I stopped and shut it down.

It's worth mentioning it's a 69 type build, the lamp is a 120v one wired between the 120v P.T tap at the voltage selector switch and the neutral at the mains power switch.
I have been mostly using it at home over the last year so it's been fairly heavily attenuated with a Marshall P.B 100. The valves are about a year old and haven't really been over used, I think they still sound fine. But I do hit the front end with quite a bit of level, fuzz, boost and drive. I took a look in the amp and the out put valves looked normal (no red plating that I could see) but I couldn't see the pre amp valves because of the covers.

I was going to pull it out this weekend for a proper inspection but thought I'd see if any one has any ideas for what I might be looking for.

Could it be a dodgy Valve??? Maybe Preamp I'm thinking/hoping???

[danman]

Have you checked the solder connections for the lamp? Might try lightly tapping on it while it is on to check for a fault. If something is truly pulling down the primary voltage enough to dim the light then I would think a fuse would have blown and the amps sound should definitely change. You could try connecting your meter across the lamp while playing to check for dropping voltage but i would suspect a bad lamp or connection.

[revolver1]

I'm going to have a look inside today, the thing is I seriously doubt that it would be a bad joint, I'm far too OCD for that to have happened Could be a bad internal connection in the lamp but I think your right about the primary voltage being lowered, troubble is what ever could be causing it goes way beyond my limited knowledge.

[danman]

I have the OCD myself so I know what you mean...lol! Clip your meter leads on both ends of the lamp leads and monitor the AC voltage as the lamp is dimming. If the voltage doesn't drop when the lamp is dimming then I would suspect a bad lamp. If it is dropping then you need to determine if it's the amp/PT that is causing the issue or is it the wall voltage that is fluctuating.

[revolver1]

Cheers danman, hopefully I think the neighbours are out tomorrow so I'll give it a go then when I can make some noise.
The more I think about it fluctuating wall voltage seems a likely culprit. We had a fairly cold day on Friday plus I live in London near a train workshop that sucks a lot of juice off the grid.
I get a little paranoid when something like this happens since I've had the misfortune of blowing a couple of P.T's back in the 90's from bad quality output valves. I couldn't bear the thought of having to replace the P.T on this amp.

[neikeel]

Cheers danman, hopefully I think the neighbours are out tomorrow so I'll give it a go then when I can make some noise.
The more I think about it fluctuating wall voltage seems a likely culprit. We had a fairly cold day on Friday plus I live in London near a train workshop that sucks a lot of juice off the grid.
I get a little paranoid when something like this happens since I've had the misfortune of blowing a couple of P.T's back in the 90's from bad quality output valves. I couldn't bear the thought of having to replace the P.T on this amp.

What is the amp? Correct value fuses should really protect the PT from dodgy output valves (less so the OT tho') they are not really in the same part of the circuit. Good condition filter caps, screen resistors of correct value are a good idea and if you are really paranoid some anti-flyback diodes on the output tube sockets will help protect the OT together with avoiding impedence mismatches.
Of course hammering the amp at full tilt into an attenuator is always a risk

[revolver1]

Hey neikeel, it's a 69 build with lower filtering. Its pretty stock really. I finished it back in the beginning of 2012, so its not that old really. However It does have a lot of NOS bits in it.
I'm running the attenuator down around 60 to 65% I guess. The eq is not too mad pres 5, bass 3, mid & treb 6. I jack the channels together Input 1 on a bit over 2 and input 2 at 5. The only mad thing I do is run the ts808 level almost flat out with the gain set just over halfway.
I'm just wondering if anything springs to mind?
I've been sifting through previous posts and so far not much joy. The only thing that came close was a post referring to Ed turning his variac down till the lamp dimmed on heavy chords.
They said that was around the 90-95volt mark. So given that they are running on 120v that's a drop of around 25-30volts. I could easily see the wall voltage dropping that far around here especially on a cold day.
I usually have to get up before the tubes start if I want a decent steady voltage to bias with.
If I can't see anything obvious its a reasonable explination. Just thought I'd be wise to ask the more experienced guys like your self If you had any thought's.

[revolver1]

Oh sorry neikeel, to answer your first question. The amp I blew the pt's on was a 96 Marshall 100watt SLP head. I'm sure it was the pt's but but there's been alot of Stella under the bridge since the mid 90's, maybe it was the ot.

I had it back to Marshall for repair, it went again not long after so I asked for a new unit from where I bought it.
I blew that one not long after that, then took it to a tech I met around the time. He ended up fitting Marshalls new EU compliant pt and recommended GT E34ls as they were the most robust tubes I could get at the time and that was the last of the problems.

But it's always in my mind, I was young and didn't have much spare cash and it was costing me £300 each time it happened.

Think I was using jj tesla back then from watford... and every time it came back from Marshall it had their valves, Svetlana I think.

Suppose what I was getting at is I really don't want to have to go through ordering a new whistle: transformer then pay for shipping then import duty. But more than that I'd hate to have to pull apart all that lovely lacing and lead dress ya know...

[revolver1]

Well I haven't got a clue what's going on... I found one preamp valve that was a bit suspect, it was causing a very intermittent crackling and loss of volume on occasion. I replaced that and it's all good.
I checked the bias it was a tad cold so I sorted that out. I gave all the joints a once over and there's no problems there, I couldn't test the voltage on the lamp whilst playing because I was on my own (not enough hands). But I gave It a good wiggle around and there's nothing wrong with the lamp. Wall voltage is pretty good and steady.

So I put it all back together and gave it a play today. The sound is fine, no red plating But when you hit the strings heavily the lamp fades out and returns when the intensity of the signal fades and now it happens on it's own with out the tube screamer or fuzz. (Probably because I re biased). Something has got to be sucking the voltage down but I haven't got a clue what it is.

I played it for about 3hrs today, I can't imagine the lamp is going to last long at this rate but it's a fancy light show for now.
Guess I'll just keep playing it till the offending component fails. It's not a very technical approach to fault finding but it'll have to do.

[danman]

You can pick up some clips at Radio Shack that will slide over the tips of your meter probes. They make it much easier to take voltage readings while you play.

[revolver1]

Cheers, I'm going to pick some up today, see if I can get to the bottom of this.

[revolver1]

Right, sooooooooo.
I pulled all the valves out and cleaned all the pins. Reinstalled V1, V2, V3, then tested all voltages as per build instructions. Nothing out of the ordinary... I installed the out put valves and noticed the lamp dimmed when I flicked the stand-by on. So that initial power up caused the voltage to sag enough to dim the lamp. I went ahead and biased it up to 65%.

One thing... I get the highest wall voltages before the tube trains start say before 4:30am then as you might expect it drops off a bit then comes up a little as people get off to work. It then drops off again as the kids get home from school around 4:30pm ish. This is all normal I know. I bias for worst case scenario so I get up real early to do It but as I monitor things through out the day things don't follow suit, not sure if this is just my amp.
I know people have a lot of arguments over biasing methods but as I see it there is no one perfect method and anyway I don't have a scope. I can adjust it perfectly with the plate voltages up around the 473vdc mark then as the wall voltage drops off the plate voltage can go down to as low as 450vdc. But instead of the inverse relation ship between plate voltage and the current draw that you would expect to see, well it all goes a bit way ward.
Today at 450vdc on the plates I was seeing a current draw reading that should have had around 483vdc on the plates???? But at 473vdc it was perfectly biased at 65%. What's up with that? Is it down to the type of valve? I'm using the E34ls??? I only mention this because it seems a little odd and may be an indication of other problems??? It follows suit to a point then goes to hell and picks it up again, I don't know...

So I took a voltage reading at the voltage selector, where the indicator lamp is joined to the 110v P.T tap and I was seeing about 100vac. This was around the lowest expected wall voltage time of the day. I plugged in a guitar and took a reading while striking a chord and the voltage dipped about 5 volts and the lamp dimmed heavily.
I checked and it didn't do it on low volume settings, it needs to be up around 5 but it does it on each channel and on both channels jacked together as long as one channel is up around 5 ish.
I also tried it with the attenuator and with out and it was maybe slightly worse with the attenuator but not a lot in it.

So it's not the lamp. It seems the more signal that hits the front end or when the amp is working hard, something is drawing so much current off the secondary side of the P.T that the overall voltage on the secondary side of the P.T is sagging. Although its not enough to blow a fuse.

The sound seems ok but I just wonder if this is the power caps smoothing things out?
Really at the end of the day I'm an Electrician so electrical principals aren't so hard to grasp but It's all a little bit out of my realm. I built the amp yeah, but I really haven't got a clue what is doing this. Logic is telling me its gotta be a valve, they are on the brink for a change as I like to play it safe. Or maybe one of the mustard caps I bought wasn't up to much but I wouldn't even know how to test for it or know if those components are capable of these symptoms. These are the most possible weak links to my mind at this point in the amp's life.
Every thing is telling me to seek out a professional to sort it but to be honest most of the so called pro's I've come into contact with don't seem to be much better than me.... so whadda ya do? And as I'm outa work at the moment I'd rather do it my self and at least learn a bit...

[revolver1]

Just thought of something else, may be unrelated but... I originally had F&T filter caps on it and then I started hearing good things about ARS. I thought to my self I'd change to ARS when the time comes. Then I noticed (I think it was the power cap not the screen closest to the valve) had a blister on top. I thought it was just a reaction to the heat coming off the valve. So anyway I changed them all to ARS. I'm just wondering this problem may have been there for a while and its been stressing the power caps but not bad enough for me to see the lamp dim??? Again I don't know...

[danman]

It's not unusual to see the plastic covering deformed slightly on the end of the filter cap. The vent is usually located on the other end of the cap where the terminals are attached. This is the end that will usually blow if a problem arises. The coupling caps (the mustards) are also unlikely to be causing excessive current draw on their own. There are instances where one may be bad, and DC voltage will leak through to the next stage. This will create problems with the amps performance and biasing but from what you mentioned I don't believe that to be the case. If you are concerned about one of the electrolytics drawing excessive current, you could try subbing a new one in temporarily at each location and test the amp each time. This may help you to weed out a bad one.

I am wondering why you show such a low AC voltage reading (100vac) on the primary side of the PT. Is this the same voltage reading that you are getting from your outlet? It's not uncommon for mine to dip down around 115v at certain times of the day but I don't believe that I have ever seen mine go that low.

[revolver1]

Thanks danman, some good info there, at least I'm learning something.
I'm in England so our wall voltage is 230v the lamp is wired off the 110v tap wich I believe is for Japan. I agree though it's a bit odd that the voltage was as low as100v it was at the worst time of the day but still. The readings dipped to 95v when the lamp went out so that's not surprising as it's supposed to run off 110v.
There's barely enough voltage there to run the lamp.

Even as I'm writing this it's dawned on me even though that tap was ment to be 110v on a British amp I'll bet that tap is actually wound for 100v. It is a hand wound American transformer. But its still strainge its never dipped before.

I had a chance to find few a good articles on the web about fault finding and a few possible faults that could be going on so I'm going to have another look at it over the weekend.

I'm thinking I have to look at everything that runs off the P.T. Now I've got my head around it I'm kinda getting to grips with it. One things for sure I bet its not going to be straight forward.