| Author |
Message |
ben
Guest
|
 Posted:
Fri Sep 30, 2005 4:43 am Post subject:
Amp - loss of power? |
|
|
Request some assistance diagnosing a problem in the power amp of a
Marshall Artist 4203.
Amp runs 2xEL34's,12AX7 phase inverter and a solid state rectifier.
The problem is loss of power.
Usually within a minute or two of start up.
It can occur within a few seconds of start up, which makes me think it
is not necessarily heat related.
I have replaced tubes and filter cap.
The interesting clue is when I open the standby with signal present,
there is a brief surge of near full power.
I am guessing, but it's like the charge in the filter cap drives it
once the supply from the rectifier is removed.
Does this automatically suggest rectifier diodes as the problem?
I am new at this and have only a basic theoretical understanding - any
help greatly appreciated.
Schematic available here:
http://www.drtube.com/schematics/marshall/3203pwr.gif
Thanks.
|
|
| Back to top |
|
 |
Phil Allison
Guest
|
| Posted:
Fri Sep 30, 2005 4:43 am Post subject:
Re: Amp - loss of power? |
|
|
"ben"
| Quote: |
Request some assistance diagnosing a problem in the power amp of a
Marshall Artist 4203.
Amp runs 2xEL34's,12AX7 phase inverter and a solid state rectifier.
The problem is loss of gain*****.
Usually within a minute or two of start up.
It can occur within a few seconds of start up, which makes me think it
is not necessarily heat related.
I have replaced tubes and filter cap.
The interesting clue is when I open the standby with signal present,
there is a brief surge of near full power.
I am guessing, but it's like the charge in the filter cap drives it
once the supply from the rectifier is removed.
Does this automatically suggest rectifier diodes as the problem?
I am new at this and have only a basic theoretical understanding - any
help greatly appreciated.
Schematic available here:
http://www.drtube.com/schematics/marshall/3203pwr.gif
|
** Can you measure the -39 volt bias voltage (at cathode end of D1) to the
output tubes ?
Does it change in the low and high gain modes ?
Does the 74.5 volt AC change ?
Replace R1 (56 kohms) and C6 ( 0.047 uF X-cap) if any peculiar changes are
going on here.
Use a 1 watt type for R1.
.......... Phil |
|
| Back to top |
|
|
Phil Allison
Guest
|
| Posted:
Fri Sep 30, 2005 4:43 am Post subject:
Re: Amp - loss of power? |
|
|
"ben"
| Quote: |
The problem is loss of power.
|
** In your world are "power" and "gain" the same thing ?
IOW - kindly describe the symptom PROPERLY !!!
......... Phil |
|
| Back to top |
|
|
ben
Guest
|
| Posted:
Fri Sep 30, 2005 4:43 am Post subject:
Re: Amp - loss of power? |
|
|
I guess I mean gain..
Tubes stay lit so power is present.
Should have described it "loss of volume". |
|
| Back to top |
|
|
JamesG
Guest
|
| Posted:
Fri Sep 30, 2005 4:43 am Post subject:
Re: Amp - loss of power? |
|
|
Ben,
It sounds like you know enough to work on the amp safely, which is good.
You need more information
The first step in diagnosing a problem of this type is to measure the
voltages, particularly the power stage B+ and bias voltages. One
possibility is that the tubes are at the wrong operating point. Placing the
amp on standby will cause the B+ to sag, perhaps placing the the tubes
momentarily at a more appropriate operating point.
There are individuals on this newsgroup who can give you excellent advice. I
would not change any components until you have clearer idea of what is
happening. A good place to get you started is:
http://www.geofex.com/index.htm
"ben" <o2xtal@swiftdsl.com.au> wrote in message
news:1128041708.167725.55160@g49g2000cwa.googlegroups.com...
| Quote: | Request some assistance diagnosing a problem in the power amp of a
Marshall Artist 4203.
Amp runs 2xEL34's,12AX7 phase inverter and a solid state rectifier.
The problem is loss of power.
Usually within a minute or two of start up.
It can occur within a few seconds of start up, which makes me think it
is not necessarily heat related.
I have replaced tubes and filter cap.
The interesting clue is when I open the standby with signal present,
there is a brief surge of near full power.
I am guessing, but it's like the charge in the filter cap drives it
once the supply from the rectifier is removed.
Does this automatically suggest rectifier diodes as the problem?
I am new at this and have only a basic theoretical understanding - any
help greatly appreciated.
Schematic available here:
http://www.drtube.com/schematics/marshall/3203pwr.gif
Thanks.
|
|
|
| Back to top |
|
|
Ian Iveson
Guest
|
| Posted:
Fri Sep 30, 2005 4:43 am Post subject:
Re: Amp - loss of power? |
|
|
"ben" wrote
| Quote: | ...The interesting clue is when I open the standby with signal
present,
there is a brief surge of near full power.
I am guessing, but it's like the charge in the filter cap drives
it
once the supply from the rectifier is removed...
|
At a guess, the bias voltage is running high so the valves switch
off. The bias is powered from the HT, so when you switch to standby
the bias drops until it approaches its proper value, and that is
when you get the burst of sound, until the HT caps discharge.
Phil is probably quite right. Just thought I might explain the
symptoms. Possibly the cap or resistor ahead of the bias diode has
failed to a short, or the resistor after the diode has failed to an
open circuit. I can't remember which way x caps fail. Phil will
know...
cheers, Ian |
|
| Back to top |
|
|
Phil Allison
Guest
|
| Posted:
Fri Sep 30, 2005 10:16 am Post subject:
Re: Amp - loss of power? |
|
|
"Ian Iveson"
| Quote: |
Phil is probably quite right.
|
** As usual.
| Quote: | Just thought I might explain the symptoms. Possibly the cap or resistor
ahead of the bias diode has failed to a short, or the resistor after the
diode has failed to an open circuit. I can't remember which way x caps
fail. Phil will know...
|
** X-caps are usually made by the "two in series" winding method.
So, if one of the two is shorting the cap's value will double hence driving
the bias voltage far more negative.
Also, the crappy 56 kohm resistors Marshall used are prone to going high
when stressed with high voltage.
........... Phil |
|
| Back to top |
|
|
Joseph Meditz
Guest
|
| Posted:
Fri Sep 30, 2005 10:57 am Post subject:
Re: Amp - loss of power? |
|
|
| Quote: | "ben" wrote
...The interesting clue is when I open the standby with signal
present,
there is a brief surge of near full power.
I am guessing, but it's like the charge in the filter cap drives
it
once the supply from the rectifier is removed...
|
| Quote: | At a guess, the bias voltage is running high so the valves switch
off. The bias is powered from the HT, so when you switch to standby
the bias drops until it approaches its proper value, and that is
when you get the burst of sound, until the HT caps discharge.
Phil is probably quite right. Just thought I might explain the
symptoms. Possibly the cap or resistor ahead of the bias diode has
failed to a short, or the resistor after the diode has failed to an
open circuit. I can't remember which way x caps fail. Phil will
know...
cheers, Ian
|
Yes, I think Phil nailed it! C6 and R1 form a voltage divider. If C6
fails toward short the bias will go up. Then, as you said, when the
amp is put in standby the bias voltage drops to allow the tube to turn
on more. The high voltage across C6 may be overstressing it and
breaking it down enough to change the voltage divider enough to raise
the bias. When entering standy the supply voltage collapses to the
point where the weakened C6 can handle it.
On the other hand, if R1 failed toward open that would also raise the
bias, but I think it would stay open and not provide that little burst
of volume when entering standby.
So, my guess is C6.
Best regards,
Joe |
|
| Back to top |
|
|
Phil Allison
Guest
|
| Posted:
Fri Sep 30, 2005 11:46 am Post subject:
Re: Amp - loss of power? |
|
|
"Joseph Meditz"
| Quote: |
Yes, I think Phil nailed it!
|
** Long as those " nails " are not in my coffin I suppose that is a
compliment.
see http://www.drtube.com/schematics/marshall/3203pwr.gif
| Quote: | C6 and R1 form a voltage divider.
|
** Not really that simple.
Poly cap C6 gives electro C3 dose of charge via D1 during the negative half
cycle of the AC supply, plus also charges itself - then on the positive
half cycle R1 discharges C6, reversing its previous charge polarity so it
can repeat the dosage to C3.
The is an optimum value for R1, either side of which the effectiveness of
C6 diminishes and the negative bias DC created also.
It has to be the *crappiest* way known to man of generating negative tube
bias - since any leakage current in the storage electro ( highly
temperature dependant of course) will have a large effect on the final
voltage achieved.
Plus the poly cap is being overstressed - hence the spec for a class X cap.
| Quote: | The high voltage across C6 may be overstressing it and
breaking it down enough to change the voltage divider enough to raise
the bias.
|
** In fact, some examples of this amp have a standard 400 volt poly cap for
C6 - ie bound to fail.
| Quote: | On the other hand, if R1 failed toward open that would also raise the
bias,
|
** An open R1 allows C6 to charge to the peak AC voltage and then do
nothing.
.............. Phil |
|
| Back to top |
|
|
P. Kärhä
Guest
|
| Posted:
Fri Sep 30, 2005 1:07 pm Post subject:
Re: Amp - loss of power? |
|
|
You might also try "the Fonz trick," that is, bang the cabinet. If it then
works for a while, you have a bad soldering somewhere. Go then through the
solderings with a soldering iron.
Pete
"ben" <o2xtal@swiftdsl.com.au> wrote in message
news:1128041708.167725.55160@g49g2000cwa.googlegroups.com...
| Quote: | The problem is loss of power.
Usually within a minute or two of start up.
It can occur within a few seconds of start up, which makes me think it
is not necessarily heat related. |
|
|
| Back to top |
|
|
Joseph Meditz
Guest
|
| Posted:
Fri Sep 30, 2005 10:22 pm Post subject:
Re: Amp - loss of power? |
|
|
| Quote: | see http://www.drtube.com/schematics/marshall/3203pwr.gif
C6 and R1 form a voltage divider.
** Not really that simple.
Poly cap C6 gives electro C3 dose of charge via D1 during the negative half
cycle of the AC supply, plus also charges itself - then on the positive
half cycle R1 discharges C6, reversing its previous charge polarity so it
can repeat the dosage to C3.
|
Thanks for explaining that Phil. The truth is I had some questions
about that CKT since the reactance of C6 and resistance of R1 are about
the same. So, if I understand correctly, this CKT is actually similar
in behavior to a voltage doubler.
| Quote: | The is an optimum value for R1, either side of which the effectiveness of
C6 diminishes and the negative bias DC created also.
It has to be the *crappiest* way known to man of generating negative tube
bias - since any leakage current in the storage electro ( highly
temperature dependant of course) will have a large effect on the final
voltage achieved.
|
So, let me guess. This CKT is more efficient than a simple resistive
voltage divider allowing them to use a smaller power transformer.
| Quote: | Plus the poly cap is being overstressed - hence the spec for a class X cap.
The high voltage across C6 may be overstressing it and
breaking it down enough to change the voltage divider enough to raise
the bias.
** In fact, some examples of this amp have a standard 400 volt poly cap for
C6 - ie bound to fail.
On the other hand, if R1 failed toward open that would also raise the
bias,
** An open R1 allows C6 to charge to the peak AC voltage and then do
nothing.
|
OK.
| Quote: | ............. Phil
|
Joe |
|
| Back to top |
|
|
Bob
Guest
|
| Posted:
Sat Oct 01, 2005 4:42 am Post subject:
Re: Amp - loss of power? |
|
|
On 30 Sep 2005 10:22:53 -0700, "Joseph Meditz" <josephmeditz@yahoo.com> wrote:
| Quote: | see http://www.drtube.com/schematics/marshall/3203pwr.gif
C6 and R1 form a voltage divider.
** Not really that simple.
Poly cap C6 gives electro C3 dose of charge via D1 during the negative half
cycle of the AC supply, plus also charges itself - then on the positive
half cycle R1 discharges C6, reversing its previous charge polarity so it
can repeat the dosage to C3.
Thanks for explaining that Phil. The truth is I had some questions
about that CKT since the reactance of C6 and resistance of R1 are about
the same. So, if I understand correctly, this CKT is actually similar
in behavior to a voltage doubler.
|
Not really, it's like a half wave rectifier power supply... which it is...
| Quote: | The is an optimum value for R1, either side of which the effectiveness of
C6 diminishes and the negative bias DC created also.
It has to be the *crappiest* way known to man of generating negative tube
bias - since any leakage current in the storage electro ( highly
temperature dependant of course) will have a large effect on the final
voltage achieved.
So, let me guess. This CKT is more efficient than a simple resistive
voltage divider allowing them to use a smaller power transformer.
|
The reason they use this circuit topology is because of the non-center-tapped
transformer secondary and bridge rectifier. The positive supply bridge is
grounded, therefore there can be no way to get a DC path from the transformer
with a negative output voltage. The capacitor provides an AC signal to the diode
and capacitor which can now produce a voltage of any polarity - much like a
radio diode detector! In this case the diode is wired to provide negative
voltage. It's simply a half wave power supply. As Phil said, the first cap must
be discharged every cycle by the first resistor, so that it can pump a charge
into the second capacitor, which acts as a filter. This type of action is common
in keyboard diode keyer circuits, and I bet Phil has worked on those too!
| Quote: | Plus the poly cap is being overstressed - hence the spec for a class X cap.
The high voltage across C6 may be overstressing it and
breaking it down enough to change the voltage divider enough to raise
the bias.
** In fact, some examples of this amp have a standard 400 volt poly cap for
C6 - ie bound to fail.
On the other hand, if R1 failed toward open that would also raise the
bias,
** An open R1 allows C6 to charge to the peak AC voltage and then do
nothing.
OK.
............. Phil
Joe |
|
|
| Back to top |
|
|
Phil Allison
Guest
|
| Posted:
Sat Oct 01, 2005 4:42 am Post subject:
Re: Amp - loss of power? |
|
|
"Joseph Meditz"
Phil Allison
** The proper method is to have a winding on the PT just for tube bias -
or even its very own tranny that is powered up whenever the amp is switched
on.
The circuit arrangement used in that Marshall is an *abortion * - since
the "standby" switch disables the bias supply as well as the HT supply.
Consider what happens to the output tubes when the standby switch is first
switched on !!!
.......... Phil |
|
| Back to top |
|
|
Ian Iveson
Guest
|
| Posted:
Sat Oct 01, 2005 4:42 am Post subject:
Re: Amp - loss of power? |
|
|
"Phil Allison" wrote
| Quote: | The circuit arrangement used in that Marshall is an *abortion * -
since the "standby" switch disables the bias supply as well as
the HT supply.
Consider what happens to the output tubes when the standby switch
is first switched on !!!
|
I assume there must be a reason for this. The bias could be just the
same but taken before the switch. I wonder how long it takes to
reach full voltage, compared to the HT.
At least it seems to fail safe.
cheers, Ian |
|
| Back to top |
|
|
Phil Allison
Guest
|
| Posted:
Sat Oct 01, 2005 9:44 am Post subject:
Re: Amp - loss of power? |
|
|
"Ian Iveson"
| Quote: | "Phil Allison"
The circuit arrangement used in that Marshall is an *abortion * - since
the "standby" switch disables the bias supply as well as the HT supply.
Consider what happens to the output tubes when the standby switch is first
switched on !!!
|
** See http://www.drtube.com/schematics/marshall/3203pwr.gif
| Quote: | I assume there must be a reason for this.
|
** Sheer stupidity.
Similar Marshall models ( ie JCM900s) ot it right.
| Quote: | The bias could be just the same but taken before the switch. I wonder how
long it takes to reach full voltage, compared to the HT.
|
** HT arrives is almost instantly while the bias takes about 10 seconds to
reach 39 volts.
| Quote: | At least it seems to fail safe.
|
** Not when R1 goes open - which it does.
.......... Phil |
|
| Back to top |
|
|
|
|
|
|
Profile
Log in to check your private messages
Log in