Power tubes running red hot - help!

What's Hot
lukedlblukedlb Frets: 488
A '75 Marshall Artiste has just arrived from @Exorcist dad. Beautiful condition. The only worry is the power tubes are running red hot, like the whole tube seems on the point of exploding red hot. I quickly turned off the amp. Sounded great though. This is only when the standby is off.

Is this simply a Biasing issue?
Here's the link to the photos that clearly show the circuit board.:
http://www.thefretboard.co.uk/discussion/152857/

Is the bias control the black disc-like part on the lower RHS of the circuit board above the standby switch?

Does anyone know where I should set up the multi-meter on the artiste?

And lastly regarding draining the caps, is turning the amp off with the standby off exactly the same as a dedicated current drain? Other than touching the bias, I have no plans on fiddling around, having been 
electrocuted once already (holding my mate's bass guitar while he tried to fix his bass amp where the wires in the mains cable on the bass amp had managed to join, sending the mains through the bass through me and out of my guitar into my amp... once bitten, twice shy).
0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
«1

Comments

  • ecc83ecc83 Frets: 1589
    ICBM will know for sure but I think this is the model that "pinches" the bias voltage using a capacitor and resistor from the HT winding? If so it is likely the resistor has gone high or/and the cap O/C (and the cap SHOULD be an "X" or "Y" type and ain't)

    This is a job for a tech, if you are not sure about how and when caps get drained, stay out!

    Dave.
    0reaction image LOL 0reaction image Wow! 1reaction image Wisdom
  • DJH83004DJH83004 Frets: 196
    edited May 2019
    The most probable cause is one of the 10uF bias supply smoothing caps has gone short circuit, which effectively makes the bias supply less negative and so the valves run hotter and eventually cook. It's a job for an amp tech I'm afraid.    
    0reaction image LOL 0reaction image Wow! 1reaction image Wisdom
  • lukedlblukedlb Frets: 488
    A mate just told me the same. A tech trip it is for a full checkup. 
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ICBMICBM Frets: 71960
    DJH83004 said:
    The most probable cause is one of the 10uF bias supply smoothing caps has gone short circuit, which effectively makes the bias supply less negative and so the valves run hotter and eventually cook. It's a job for an amp tech I'm afraid.    
    This. Partial or full failure of these caps is actually much more common than of the main filter caps.

    You can often spot ones that are on the way out if the bias trimmer is right up to the end of its travel, and especially if a previous tech has bodged the feed resistor by changing it to a smaller value or piggybacking another in parallel - this is the *wrong* solution and will actually increase the leakage in the caps even if it temporarily gives a bit more voltage. Given that the circuit works correctly with the original values, it's simply faulty logic to modify the circuit rather than fixing the cause of the problem.

    If either of the caps has failed, always replace both. If it's a '75 it will be a PCB one, and in my opinion it's acceptable to snip the old caps out (as close as possible to the cap body) and solder the new ones to the wires, rather than lift the board which is a big job on these - undo all the pots and jacks, if you really have to.

    "Take these three items, some WD-40, a vise grip, and a roll of duct tape. Any man worth his salt can fix almost any problem with this stuff alone." - Walt Kowalski

    "Just because I don't care, doesn't mean I don't understand." - Homer Simpson

    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • FenderishFenderish Frets: 47
    you can check also the rectifier bridge that provides the bias voltage. if it fails, redplate occurs too (happened to me on a marshall once)
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ICBMICBM Frets: 71960
    Fenderish said:
    you can check also the rectifier bridge that provides the bias voltage. if it fails, redplate occurs too (happened to me on a marshall once)
    There's no bridge on these old models, it's a single diode fed by a resistor. I'm pretty sure I've never seen the diode fail - I think having the resistor does make it inherently less likely since there's no large current surge at start-up.

    A lot of the modern models (JTM60/JCM600, DSLs etc) do use a bridge rectifier and it's yet another of the weak points in the design, although the jumper connectors to the adjuster board on the DSL/TSLs are the most common cause of trouble...

    "Take these three items, some WD-40, a vise grip, and a roll of duct tape. Any man worth his salt can fix almost any problem with this stuff alone." - Walt Kowalski

    "Just because I don't care, doesn't mean I don't understand." - Homer Simpson

    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • FenderishFenderish Frets: 47
    A single diode on those models ? not even 4 separate diodes to make an old fashioned bridge, thus losing half the power in the resistor ? And yes it was a marshall from the nineties with this kind of bridge
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ICBMICBM Frets: 71960
    Fenderish said:
    A single diode on those models ? not even 4 separate diodes to make an old fashioned bridge, thus losing half the power in the resistor ?
    Yes, single diode. Very crude, but it works!

    What was less good is that on a lot of 70s Marshalls, the standby switch works by disconnecting the HT secondary of the power transformer, and the bias voltage is derived from this too via a larger than normal resistor. The problem is that with the large resistor, it takes a long time for the bias voltage to rise and so for the first few seconds after turning the standby to 'play', the power valves run wide-open.

    The solution is luckily simple - move the bias feed wire (the white one) to the PT side of the standby switch so it's always on. The Artiste in the pic in the thread linked above has either had it done, or left the factory like that - the bias supply is the white wire (3rd from the right-hand end of the PCB), the resistor above it and the diode above that.

    "Take these three items, some WD-40, a vise grip, and a roll of duct tape. Any man worth his salt can fix almost any problem with this stuff alone." - Walt Kowalski

    "Just because I don't care, doesn't mean I don't understand." - Homer Simpson

    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ExorcistExorcist Frets: 604
    Hi Luke, 
    I've just seen this and feel terrible! It seemed fine when we tried it at my parents house (and fine 20 years ago when I borrowed it for band practice) but it has never been serviced in its life... I certainly don't remember the valves glowing.. but I guess I may have never looked while it was powered on. I sincerely hope its a quick fix :-( Please let me know how you get on.
    Simon
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • lukedlblukedlb Frets: 488
    Simon, it’s totally cool. This is a simple fix with the help of a check up. Although I’m dreading the thought of lifting it back upstairs and into the car. These artistes really are heavy. But by god they’re handsome devils. 
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ExorcistExorcist Frets: 604
    Hi Luke, another thing - this has never seen an amp tech in its life (so 100% original) - so it is possible the power tubes were changed at some point without being biased? I can't remember if I would have done that when I borrowed it - I was super naive 20 years ago so it is possible?
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ExorcistExorcist Frets: 604
    lukedlb said:
    Simon, it’s totally cool. This is a simple fix with the help of a check up. Although I’m dreading the thought of lifting it back upstairs and into the car. These artistes really are heavy. But by god they’re handsome devils. 
    Hi Luke, 
    Thanks for that - I know what you mean about the weight, when he decided to sell it I had to carry it down the stairs for him, I'm not sure my back has fully recovered! :-) 
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ICBMICBM Frets: 71960
    It's very likely that it was apparently working fine before, and - especially if it hadn't been used much or at all for a long time until recently - the caps have just given up...

    Cap failures when they're subjected to voltages they haven't seen for years are very common, which is why I always advise that an old amp that hasn't been used for a long time should not be just 'turned on to see if it still works', but powered up carefully under controlled conditions by a tech. It can be spectacular if it's one of the main filters that goes.

    All electrolytic caps will need replacing sooner or later anyway.

    "Take these three items, some WD-40, a vise grip, and a roll of duct tape. Any man worth his salt can fix almost any problem with this stuff alone." - Walt Kowalski

    "Just because I don't care, doesn't mean I don't understand." - Homer Simpson

    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • ecc83ecc83 Frets: 1589

    The use of a single diode to produce valve bias is not really poor design since the current required is microscopic. There might even be an advantage since the ripple produced is at 50Hz, well below the capabilities of most guitar speakers! A bridge produces 100Hz, about "A" on the bottom string? Mind you, the filter caps needed are low in value and cost in either case.

    Could of course be a hangover from pre-semiconductor days when a valve diode was needed to provide bias for big Mother PA amps? These went to Copper oxide then Selenium rectifiers. Both of which were less reliable than the valve! But then, big M PA amps usually had a relay bias interlock to save the amp's ass if the bias volts failed. I think the big Ampeg has that IC?


    Half or full wave design I still think a dedicated winding is the most elegant solution.

    Dave.

    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • FenderishFenderish Frets: 47
    you're fully right, I even used it without overthinking it on my 2204 :lol: (single diode for bias supply) I thought of it cause this frickin' bias supply bridge caused a problem on my 6101 :lol:

    and yes, only one diode is more reliable than a s.o.c. bridge
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • jpfampsjpfamps Frets: 2723
    ICBM said:
    DJH83004 said:
    The most probable cause is one of the 10uF bias supply smoothing caps has gone short circuit, which effectively makes the bias supply less negative and so the valves run hotter and eventually cook. It's a job for an amp tech I'm afraid.    


    You can often spot ones that are on the way out if the bias trimmer is right up to the end of its travel, and especially if a previous tech has bodged the feed resistor by changing it to a smaller value or piggybacking another in parallel - this is the *wrong* solution and will actually increase the leakage in the caps even if it temporarily gives a bit more voltage. Given that the circuit works correctly with the original values, it's simply faulty logic to modify the circuit rather than fixing the cause of the problem.


    That repair was actually recommended in a popular book on valve amps!
    0reaction image LOL 0reaction image Wow! 1reaction image Wisdom
  • ecc83ecc83 Frets: 1589
    Fenderish said:
    you're fully right, I even used it without overthinking it on my 2204 :lol: (single diode for bias supply) I thought of it cause this frickin' bias supply bridge caused a problem on my 6101 :lol:

    and yes, only one diode is more reliable than a s.o.c. bridge


    Well I wouldn't go THAT far Fends! The HT-40 uses a bridge bias rect and I have never known one fail. It is a wee DIL thing but rated at 1A and so massively over rated for the current required.

    That is the case with bridge rectifiers in general, they are so well above the current demand of even a big valve amp that they hardly ever fail, that is of course assuming the mnfctrs don't pare the spec to the bone! No need because even a 1200V 10A bridge is remarkably cheap. TVs of old sailed closer to the wind with 1.5A devices but again, rare to have one fail.


    Dave.

    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • FenderishFenderish Frets: 47
    well with an IC of 4 diodes vs one diode, given the fact that the bridge needs only one diode failure to fail completely, it's pretty fair to assume the failure risk is 4X higher with the bridge, isn't it ?
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • olafgartenolafgarten Frets: 1648
    Fenderish said:
    well with an IC of 4 diodes vs one diode, given the fact that the bridge needs only one diode failure to fail completely, it's pretty fair to assume the failure risk is 4X higher with the bridge, isn't it ?

    That's not quite how it works, the mean time between failure would be the same. 
    0reaction image LOL 0reaction image Wow! 1reaction image Wisdom
  • FenderishFenderish Frets: 47

    one diode failed in a bridge causes the failure of the whole bridge so the bridge is more likely to fail, hence the formula in the article


    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
Sign In or Register to comment.