Amp noise problem

Dominic

20w PRS  HNDX..........bit dark but great little Amp 
Awful noise ( crackle /rustle /wobbleboard sound ).......I checked to make sure the ghost of Rolf Harris wasn't hiding inside but he's not .
The noise is worst for first 4 or 5 mins on start-up/warm up even if not playing and then much less noticeable but occaisional as playing together with the odd split second stammer /cutting out.
Leads / board / inputs all checked good ..........got to be a valve BUT which one (s) ?????
Pre or Power ?
Most likely candidate ?

ecc83

Well you can pull all three 12AX7s and fire it up from cold. If it crackles then you have a bad OP valve or a circuitry problem, dry joint say. Since new OP bottles will need a bias setting, tech time.

But hopefully not the big jobs so put the Triodes back starting with the one next to the OP valves.

N.B. Even if you find valve 2 say (number them with a Sharpie) seems to be the culprit the noise could still be in the CIRCUIT associated with that valve so then swap it for another one.

FWIIW I always found the TAD 7025 HG a good, low noise Triode at a reasonable price but the valve world may have moved on!

Dave.

ecc83

??

Dominic

Thanks.........prob don't need a tech ;I can use a multimeter ....but thanks for the tips on elimination
sounds a good way of logically going about it

ecc83

Dominic said:

Thanks.........prob don't need a tech ;I can use a multimeter ....but thanks for the tips on elimination
sounds a good way of logically going about it

Yes, I saw it had a user bias trim but did not know if you had the kit or the knowledge. Do you know what the bias setting is by any chance?

Dave.

Dominic

offhand no but will check tonight.....it might suit my needs to run a touch hot anyway

ecc83

Dominic said:

offhand no but will check tonight.....it might suit my needs to run a touch hot anyway

I wouldn't. All the 25W pentode amps I dealt with (and '88's) were set for Ik =25mA and I never had a customer ask for anything hotter.

You are IMO just cooking valves, electro-caps and transformers to no good purpose.

Dave.

Dominic

noted.

ICBM

Don’t pull all the preamp valves at the same time - that will cause the voltage at the far end of the B+ chain to rise to the full HT voltage, which can exceed the cap ratings. Less common on modern amps admittedly - usually all the caps are 450V or 500V nowadays - but still not a good idea.

Just pull the phase inverter first - usually nearest the power valves. If the noise doesn’t stop it’s a power stage problem. If it does, pull each preamp valve in turn by itself, with the others still in.

Dominic

thankyou

ecc83

Dominic said:

noted.

Not often I disagree with me hon'friend ICBM but surely 95% of the HT load is the output valves and I do NOT suggest running the amp without those*.

I shall look for a schematic then we can work it out. Very unlikely it will show voltages (why almost never?) but we can assume top volts on the traff tap and go from there.

*But then maybe I am too used to working on PROPERLY designed amps with "worse case" rated components!

Dave.

ICBM

ecc83 said:

Not often I disagree with me hon'friend ICBM but surely 95% of the HT load is the output valves and I do NOT suggest running the amp without those*. 

I shall look for a schematic then we can work it out. Very unlikely it will show voltages (why almost never?) but we can assume top volts on the traff tap and go from there.

*But then maybe I am too used to working on PROPERLY designed amps with "worse case" rated components!

The problem is that even the small current draw of the preamp valves is sufficient to pull down the HT quite considerably at the far end of the chain, since it's going through tens of kohms of resistance. Typical values for, say, a 100W Marshall would be 470V at N1 (OT supply), 465V at N2 (screen grids), 375V at N3 (phase inverter), 300V at N4 (V2) and 275V at N5 (V1). The first two nodes use stacked 350V caps not single 450Vs, the remainder are single 450Vs.

So you should be able to see the problem - pull all three preamp valves and the voltage at N3, N4 and N5 will rise to 465V because there is no current draw down the B+ chain - this can cause the caps to fail, even though it's only just over the voltage rating, as they will have got conditioned to operating at the lower voltages. This is *not* bad design - there was no reason for the designers to think that the amp would ever be operated like that, so no need to rate them higher than 450V or use stacked pairs.

Yes, if you use 500V caps throughout this theoretically isn't a problem, but they were not easily available for most of the history of guitar amplifiers - 450V was the highest, and most higher-powered amps use a higher HT voltage than that. Even then, it's surprisingly common for the cap at the end of the chain to fail.

I learned this the hard way by the way .

Counterintuitively, pulling all the power valves is in fact safe as it will cause the voltages to rise by only about 10V at most, probably less - there's not much resistance upstream.

ecc83

ICBM said:

ecc83 said:

Not often I disagree with me hon'friend ICBM but surely 95% of the HT load is the output valves and I do NOT suggest running the amp without those*. 

I shall look for a schematic then we can work it out. Very unlikely it will show voltages (why almost never?) but we can assume top volts on the traff tap and go from there.

*But then maybe I am too used to working on PROPERLY designed amps with "worse case" rated components!

The problem is that even the small current draw of the preamp valves is sufficient to pull down the HT quite considerably at the far end of the chain, since it's going through tens of kohms of resistance. Typical values for, say, a 100W Marshall would be 470V at N1 (OT supply), 465V at N2 (screen grids), 375V at N3 (phase inverter), 300V at N4 (V2) and 275V at N5 (V1). The first two nodes use stacked 350V caps not single 450Vs, the remainder are single 450Vs.

So you should be able to see the problem - pull all three preamp valves and the voltage at N3, N4 and N5 will rise to 465V because there is no current draw down the B+ chain - this can cause the caps to fail, even though it's only just over the voltage rating, as they will have got conditioned to operating at the lower voltages. This is *not* bad design - there was no reason for the designers to think that the amp would ever be operated like that, so no need to rate them higher than 450V or use stacked pairs.

Yes, if you use 500V caps throughout this theoretically isn't a problem, but they were not easily available for most of the history of guitar amplifiers - 450V was the highest, and most higher-powered amps use a higher HT voltage than that. Even then, it's surprisingly common for the cap at the end of the chain to fail.

I learned this the hard way by the way .

Counterintuitively, pulling all the power valves is in fact safe as it will cause the voltages to rise by only about 10V at most, probably less - there's not much resistance upstream.

I am PROSTRATE with apology IC! I have just worked through a Fender schemo and yes, I see the problem, although all the caps in the circuit I happened on were rated at 450V and the +A rail only 435V or so.

Of course the same problem would occur if the pre amp stages lost heater volts or indeed a heater went O/C. Fortunately that is a surprisingly rare event.

Dave.

ICBM

ecc83 said:

I am PROSTRATE with apology IC!

You don't need to be . I agree it's a non-obvious problem if you're not used to working on this very narrow little niche of older guitar amps that I'm overly familiar with .

It never occurred to me either until the first time I was testing an amp with all the valves out - I can't even remember why now, it was about 30 years ago - and a filter cap went pop! Quite a learning experience - luckily I've always been the sort of person who finds out *why* something went wrong.

Boromedic

I had similar on my HDRX20, turned out to be V1. The HDRX has external bias measuring points and a trimmer, pretty good for a budget amp

Note for Dominic though, mines anything but dark tbh, dialling out a bit of fizz usually!

Dominic

interesting thanks