Help! intermitent problem, very low volume!

[Silvio55]

Hello! I´m having a problem with my Vox Night Train, sometimes, playing at low volume and low gain, the volume drops, it doesn´t die but it goes very quiet, turning up the volume knob does make the volume go louder but at full volume it´s still very low and very thin sounding. If I move the gain knob up the the problem desappears, even if I turn it down again. As I said it´s an intermitent problem, not always happens, and it´s very rare to happen with the gain up. Other thing I notice is that sometimes when I turn the amp on is a noise like AM radio static that goes away after a few seconds.
Thanks in advance!!

[Martin Pedder]

This sounds like a classic dirty contact somewhere. I don't know your amp, an perhaps someone else who does can chip in, but i would be looking for any molex or edge connectors link leads from one board to another. Both the level drop and the radio static sound fit this symptom. It could also be a mucky or failing level control. I am not a great advocate of switch cleaners and lubricants as problems they cure rarely stay cured for long. They do have there uses occasionally, and if you eliminate other contact problems, exercising or squirting the level pot could be a logical next move.

[Enzo]

Oh the classic durty contact for that symptom would be in an FX loop return jack, or a powr amp in jack. Don;t know if your amp has one of those.

I don't know this amp, but it appears there is more than one Night Train Vox, so what is the model number from the serial number plate?

[Silvio55]

Yes, the dirty contact was the first I thought, but this amp doesn´t have an FX loop. It´s a Vox NT 15 watt head. I has some mods made by me, but I checked all the solders but until now I found nothing, they all seem right.

[Martin Pedder]

Presumably these two pictures I found on the web look like yours, with the PCB living the other way up:


I see what you mean, there isn't much to leap on as the likely culprit. I assume the input socket is fitted on the PCB, but out of picture. Definitely clean the socket.

Then, I would run a soldering iron over any joints that are looking dry, white, cracked or in any way not shiny. Definitely re-solder the jack sockets and posts. Most electronic faults are actually not electronic but mechanical faults. Anything that moves with use like sockets, pots and switches, or heavy components like coils will stress their solder joints and are the most likely source of problems. Next, re-solder the leads of any components that get hot. I can see 3 black wire-wound resistors top left for example. Give the valves a wiggle to dislodge any muck on their pins.

If none of that works, someone with better valve knowledge than me should take you through the next steps.
Martin

[Enzo]

The next step beyond that stuff would be get a schematic and follow the signal through the amp to see where it disappears. Alternatively we can also inject signal at the power stage then work our way back until we get to where injected signal no longer passes.

And even in the absence of schematics, we might first go down the two 12AX7s and check for B+ on the plates - pins 1 and 6. And also for a few volts on the cathodes - pins 3 and 8, to see if the tubes are conducting.

And we know which pins are grids on the tubes, we can inject some noise into each of them and see which ones do and do not come out the speaker.


Other not rare failures might be cracked volume control wafers.

[Silvio55]

Well, I opened the amp the other day and checked for loose contacts, I switched it on, and when I got the failure I did the "wood stick method" , but nothing except turning the gain up seemed to cure the symptom, then closed the chassis and played it. For my surprise it seems to work OK now, but that worries me cause I didn´t do anything, and I don´t know when it´s gonna fail again. I opening it again these days, remove the pcb and check all the solders.

[Martin Pedder]

Faults that go away are always a worry. You know they will probably reappear but you just don't know when.

I had a studio customer who swore that his equipment could smell my tool case as I drove into his road. Faults that had been there for weeks would disappear just because I was in the building. He reckoned that the machines were thinking "I better behave myself or that Pedder will get his screwdriver out - again". Needless to say, the faults would often come back a few days later.

I have had this signal related low-level problem so many times over the years, where turning up the level, or just giving it a burst of high level signal makes it work. It is mechanical somewhere - a dry joint, a hairline fracture in a track on the PCB, a mucky contact in a socket, a poor crimp in a Molex, a badly made IDC contact, a crack or contact issue in a level pot or even in the lead of a passive component. I had this problem on a resistor whose leads had been bent to sharply. Tantalum and ceramic capacitors can also do this. You are doing all the right things like tapping around with a wooden or plastic prodder.

Often, once the level has jumped back to normal, it is fine until you turn it off and the fault re-appears tomorrow. If a hairline fracture in a track or solder joint is causing this, a freezer spray and a hair dryer can be handy tools to pinpoint the problem. If when faulty from cold, you can "cure" it just with hot air, without touching it, use the freezer spray is short bursts to cool areas of the board a bit at a time to see if the fault comes back. Allow time for the cold to soak through.

Enzo's suggestion of a pot problem is a good one and worth pursuing. I have had pots with cracked wafers, damaged or scraped carbon tracks where grit has got in, and poor wiper contacts. If the pot is making scratchy noises, check for DC across the pot caused by a leaky cap. The carbon tracks or conductive plastic tracks in pots and faders cannot handle anything much in the way of current.

Do you get the impression that it might be moving the pot that cures it or just the increase in signal level? If injecting a louder signal at the input brings the gain back to normal in the same way as turning the level up on the amp, then at least you know the circuitry before the level pot is OK.

It is not a huge board, so "wetting" every solder joint on the board with fresh solder maybe a quick way to eliminate the problem. You will never know exactly where the fault was, but if the problem goes away and doesn't come back, that is more important.

Martin

[Silvio55]

Thanks Martin! I think that increasing the gain and not moving the pot is what cures the problem, I haven´t tried injecting a louder signal yet. I took out the pcb today so I will try to wet all the solder joints and re check anything else.

[olddawg]

I would at least resolder that gain pot and spray some Caig in it. I had a Marshall Lead MOSFET 100 drive me nuts one time. I finally figured put that one connection on the gain pot was wrapped but never soldered from the factory. The amp was over 10 years old. It finally oxidized enough to be intermittent.