GZ34 Rectifier question

[Robert Manktelow @ work]

Hello All

I have a question regarding my valve amps power supply. This is another one
in the series - sorry.

The mains transformer measures 400 - 0 - 400 VAC.
The transformers centre tap is connected to chassis via a thermister which
drops approx 13VAC (measured using a DVM) when 90mA DC is drawn from the
PSU.
This thermister measures approx 3Kohms cold.

In series between the transformer and with each plate of the GZ34 is a 50ohm
power resistor.
The first capacitor is 50uF, after the GZ34, followed by a 10H choke.

With no input signal I measure 465VDC across this first capacitor and this
voltage drops by 20V at full load (45Watts into a dummy load)

Well, if you got this far thanks:)

However in all of the other valve guitar amps schematics I have seen there
are no series resistors OR thermisters in the power supply circuit, just
transformer direct to GZ34.

Assuming I can reduce the transformers secondary to achieve the same DC
voltage output can I safely remove both the two resistors and the
thermister?
This modification should (will it?) increase the dynamic response of the
amplifier - that's my aim.

Note - I have fitted a standby switch, the original design did not have one
of these.

Robert

[Chris Berry]

Post by Robert Manktelow @ work
Hello All
I have a question regarding my valve amps power supply. This is another one
in the series - sorry.
The mains transformer measures 400 - 0 - 400 VAC.
The transformers centre tap is connected to chassis via a thermister which
drops approx 13VAC (measured using a DVM) when 90mA DC is drawn from the
PSU.
This thermister measures approx 3Kohms cold.
In series between the transformer and with each plate of the GZ34 is a 50ohm
power resistor.
The first capacitor is 50uF, after the GZ34, followed by a 10H choke.
With no input signal I measure 465VDC across this first capacitor and this
voltage drops by 20V at full load (45Watts into a dummy load)
Well, if you got this far thanks:)
However in all of the other valve guitar amps schematics I have seen there
are no series resistors OR thermisters in the power supply circuit, just
transformer direct to GZ34.
Assuming I can reduce the transformers secondary to achieve the same DC
voltage output can I safely remove both the two resistors and the
thermister?
This modification should (will it?) increase the dynamic response of the
amplifier - that's my aim.
Note - I have fitted a standby switch, the original design did not have one
of these.
Robert

Sounds familiar...
It looks like someone had previously fitted a couple of diodes and resistors
in a valve base in that amp and then made some adjustments.
You're right - you don't need those components with a GZ34 but the
capacitors are also too large to run the amp without diodes.

Does this look familiar for the supply?
Loading Image...
cb

[Phil Crooks]

Post by Robert Manktelow @ work
Hello All
I have a question regarding my valve amps power supply. This is another one
in the series - sorry.
The mains transformer measures 400 - 0 - 400 VAC.
The transformers centre tap is connected to chassis via a thermister which
drops approx 13VAC (measured using a DVM) when 90mA DC is drawn from the
PSU.
This thermister measures approx 3Kohms cold.
In series between the transformer and with each plate of the GZ34 is a 50ohm
power resistor.
The first capacitor is 50uF, after the GZ34, followed by a 10H choke.
With no input signal I measure 465VDC across this first capacitor and this
voltage drops by 20V at full load (45Watts into a dummy load)
Well, if you got this far thanks:)
However in all of the other valve guitar amps schematics I have seen there
are no series resistors OR thermisters in the power supply circuit, just
transformer direct to GZ34.
Assuming I can reduce the transformers secondary to achieve the same DC
voltage output can I safely remove both the two resistors and the
thermister?
This modification should (will it?) increase the dynamic response of the
amplifier - that's my aim.
Note - I have fitted a standby switch, the original design did not have one
of these.
Robert

I'd guess that the thermister and power resistors are there to protect the
circuit from the inrush of current when the GZ34 starts to conduct. A
400-0-400 transformer will generate 400V * 1.414 = 565V peak when wired up
in the standard manner and the large cap will look like a short circuit for
a fraction of a second. I don't know anything about the GZ34 si I can't say
for sure.

The standby switch might fix things but tread carefully. You may be better
replacing the GZ34 with silicon rectifiers if you want a more "springy"
dynamic sound. That will remove the resistance and sag inherent in a valve
rectifier (most people seem to love it). I'd be tempted to leave the
thermister; it will have very little effect after warmed up and will reduce
the surge at start up from cold.

Phil

[Stewart Ward]

Post by Phil Crooks
I'd guess that the thermister and power resistors are there to protect the
circuit from the inrush of current when the GZ34 starts to conduct. A
400-0-400 transformer will generate 400V * 1.414 = 565V peak when wired up
in the standard manner and the large cap will look like a short circuit for
a fraction of a second. I don't know anything about the GZ34 si I can't say
for sure.
The standby switch might fix things but tread carefully. You may be better
replacing the GZ34 with silicon rectifiers if you want a more "springy"
dynamic sound. That will remove the resistance and sag inherent in a valve
rectifier (most people seem to love it). I'd be tempted to leave the
thermister; it will have very little effect after warmed up and will reduce
the surge at start up from cold.
Phil

Well, actually, a GZ34 (250mA max) rectifier warms up and starts to conduct
very slowly as do all the other valves... so there's absolutely no need for
'surge limiting' or any other kind of 'soft start' circuitry.

If the amplifier has been converted from GZ34 to diode rectification, then
the HT will be somewhat higher than required. This can cause a strain on
the output valves and reduce their life expectancy. Although, for as long
as they are working, you will get more output power from them. You could
offset this danger by raising the bias from about -34V to -40V to bias the
valves a little more further back round the conduction 'knee', but this will
introduce higher levels of crossover distortion, which is unpleaseant to the
ear.

Some of the early Marshalls were fitted with KT88s, which did require quite
high levels of HT. 560VDC was about normal. So this amp may have a
transformer set intended from those beasts and someone has attempted to make
it safe for EL34 operation.

Whatever you do, it might not be right for EL34s, *probably* with that mains
TX.

[Chris and Mich]

Post by Stewart Ward
Some of the early Marshalls were fitted with KT88s, which did require quite
high levels of HT. 560VDC was about normal. So this amp may have a
transformer set intended from those beasts and someone has attempted to make
it safe for EL34 operation.
Whatever you do, it might not be right for EL34s, *probably* with that mains
TX.

565VDC should be okay for EL34s. The maximum plate voltage is around 800VDC.
I have found a data sheet for Svetlana EL34s at
http://homepage.ntlworld.com/richard.maile/images/EL34.pdf

Maybe the voltage was reduced to give slightly less headroom and earlier
signal breakup.

Just a thought...

Chris

[Chris Berry]

Post by Chris and Mich

Post by Stewart Ward
Some of the early Marshalls were fitted with KT88s, which did require

quite

Post by Stewart Ward
high levels of HT. 560VDC was about normal. So this amp may have a
transformer set intended from those beasts and someone has attempted to

make

Post by Stewart Ward
it safe for EL34 operation.
Whatever you do, it might not be right for EL34s, *probably* with that

mains

Post by Stewart Ward
TX.

565VDC should be okay for EL34s. The maximum plate voltage is around 800VDC.

565 MAY be normal for EL34's but it certainly is high for your 500V/450V
capacitors....
It will work but you'll be having reliability problems a little earlier.
cb

[Stewart Ward]

Post by Chris and Mich
565VDC should be okay for EL34s. The maximum plate voltage is around 800VDC.
I have found a data sheet for Svetlana EL34s at

That May be true, but not with any old output TX. So *not* with the OP TX
as fitted in the amp in question.

You must ensure that the maximum anode power dissipation stays at 25 Watts
or less per valve*. A push-pull output stage can be made with 800 volts HT,
but the anode-anode TX impedance increases to 11k ohms from the 4-5k of a
500 volt HT design.... and the bias changes from -36V to -39V and the Rg2
changes...........!

It ain't that simple to do it right! God help those who have amps that
claim 50 watts from four EL84s (12W/anode dissipation max*). Those poor old
valves must be in for a right old caneing, with their buckling anodes and
fast stripping cathodes!!! You'd need a bottomless pocket to finance
replacement valves for those amps!


* - Data source: Mullard Data Book, issue 3, page EL34 1056-3 & EL84 1057-A.
Scanned copies in .jpg available!

[JNugent]

Post by Stewart Ward

Post by Chris and Mich
565VDC should be okay for EL34s. The maximum plate voltage is around
800VDC. I have found a data sheet for Svetlana EL34s at

That May be true, but not with any old output TX. So *not* with the
OP TX as fitted in the amp in question.
You must ensure that the maximum anode power dissipation stays at 25
Watts or less per valve*. A push-pull output stage can be made with
800 volts HT, but the anode-anode TX impedance increases to 11k ohms
from the 4-5k of a 500 volt HT design.... and the bias changes from
-36V to -39V and the Rg2 changes...........!
It ain't that simple to do it right! God help those who have amps
that claim 50 watts from four EL84s (12W/anode dissipation max*).
Those poor old valves must be in for a right old caneing, with their
buckling anodes and fast stripping cathodes!!! You'd need a
bottomless pocket to finance replacement valves for those amps!
* - Data source: Mullard Data Book, issue 3, page EL34 1056-3 & EL84
1057-A. Scanned copies in .jpg available!

Hmmm... I have such an amp (a Carvin Nomad).

Mind you, I take the "50 watt" claim with a teaspoonful of salt - and
certainly don't push the volume past equivalent AC30 levels (and yes, I have
one of those as well - a real one... eat your heart out, George Weston).


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.658 / Virus Database: 421 - Release Date: 09/04/04

[George Weston]

Post by JNugent

Post by Stewart Ward

Post by Chris and Mich
565VDC should be okay for EL34s. The maximum plate voltage is around
800VDC. I have found a data sheet for Svetlana EL34s at

That May be true, but not with any old output TX. So *not* with the
OP TX as fitted in the amp in question.
You must ensure that the maximum anode power dissipation stays at 25
Watts or less per valve*. A push-pull output stage can be made with
800 volts HT, but the anode-anode TX impedance increases to 11k ohms
from the 4-5k of a 500 volt HT design.... and the bias changes from
-36V to -39V and the Rg2 changes...........!
It ain't that simple to do it right! God help those who have amps
that claim 50 watts from four EL84s (12W/anode dissipation max*).
Those poor old valves must be in for a right old caneing, with their
buckling anodes and fast stripping cathodes!!! You'd need a
bottomless pocket to finance replacement valves for those amps!
* - Data source: Mullard Data Book, issue 3, page EL34 1056-3 & EL84
1057-A. Scanned copies in .jpg available!

Hmmm... I have such an amp (a Carvin Nomad).
Mind you, I take the "50 watt" claim with a teaspoonful of salt - and
certainly don't push the volume past equivalent AC30 levels (and yes, I have
one of those as well - a real one... eat your heart out, George Weston).

Oh - these "vintage" aficionados!
One day, Jim, I'd welcome a shoot-out with your "real" one and my "virtual"
AC30... :-)
George

[Phil Crooks]

Post by Robert Manktelow @ work

Post by Phil Crooks
I'd guess that the thermister and power resistors are there to protect the
circuit from the inrush of current when the GZ34 starts to conduct. A
400-0-400 transformer will generate 400V * 1.414 = 565V peak when wired up
in the standard manner and the large cap will look like a short circuit

for

Post by Phil Crooks
a fraction of a second. I don't know anything about the GZ34 si I can't

say

Post by Phil Crooks
for sure.
The standby switch might fix things but tread carefully. You may be better
replacing the GZ34 with silicon rectifiers if you want a more "springy"
dynamic sound. That will remove the resistance and sag inherent in a valve
rectifier (most people seem to love it). I'd be tempted to leave the
thermister; it will have very little effect after warmed up and will

reduce

Post by Phil Crooks
the surge at start up from cold.
Phil

Well, actually, a GZ34 (250mA max) rectifier warms up and starts to conduct
very slowly as do all the other valves... so there's absolutely no need for
'surge limiting' or any other kind of 'soft start' circuitry.

Absolutely... Unless they've already been warmed up. I too am a little
confused by the need for two sets of soft start circuitry.

Phil

[icarusi]

Post by Robert Manktelow @ work
Note - I have fitted a standby switch, the original design did not have one
of these.

It sounds like some sort of auto soft start circuitry, not required if
you've a standby switch. Older TVs used thermistors to auto-switch off the
degausing coils after switch-on.

Icarusi
--
remove the 00 to reply

[Robert Manktelow @ work]

Post by Robert Manktelow @ work
Hello All
I have a question regarding my valve amps power supply. This is another one
in the series - sorry.
The mains transformer measures 400 - 0 - 400 VAC.
The transformers centre tap is connected to chassis via a thermister which
drops approx 13VAC (measured using a DVM) when 90mA DC is drawn from the
PSU.
This thermister measures approx 3Kohms cold.
In series between the transformer and with each plate of the GZ34 is a 50ohm
power resistor.
The first capacitor is 50uF, after the GZ34, followed by a 10H choke.
With no input signal I measure 465VDC across this first capacitor and this
voltage drops by 20V at full load (45Watts into a dummy load)
Well, if you got this far thanks:)
However in all of the other valve guitar amps schematics I have seen there
are no series resistors OR thermisters in the power supply circuit, just
transformer direct to GZ34.
Assuming I can reduce the transformers secondary to achieve the same DC
voltage output can I safely remove both the two resistors and the
thermister?
This modification should (will it?) increase the dynamic response of the
amplifier - that's my aim.
Note - I have fitted a standby switch, the original design did not have one
of these.
Robert

Thanks for all the help, much appreciated.
Just to fill in a few missing bits in my original post.

This amp has always had valve rectifiers, initially GZ32 then later a GZ34 -
before it got to me.
The op valves were originally KT66 but are now 6L6GC.
Grampian (the manufacturer) made a range of amps (including KT88 based ones)
I assume that they used the same mains transformer for this one (adding
resistors and thermistor) to drop the rail - I think mine was rated at
20Watts originally.
The output transformer secondary measures 80 ohms between each 400V and the
centre tap (total 160 ohms).

So.. judging by your replies my plan is: -
Keep the thermister and the standby switch
Remove the two 50 ohm resistors in the GZ34 plates
Add a MOSFET/zener voltage drop circuit between the transformers centre tap
and ground to "loose" some of the AC to get 465V DC across the first
capacitor.

Replacing the GZ34 with solid state diodes will give me even more of a
problem ( more volts to loose) so I will stay away from this option.

I have a variac ect so plan to "creep up" on this mod to see if it does what
it should do.

It's at times like this that I wonder why I did not just buy a guitar valve
amp from Fender etc. but I guess learning never came cheap!

Robert

[Stewart Ward]

Post by Robert Manktelow @ work
Add a MOSFET/zener voltage drop circuit between the transformers centre tap
and ground to "loose" some of the AC to get 465V DC across the first
capacitor.
Robert

Errm, not a good plan, as you will need a dropper resistor to lose the volts
across. So you'll be back to square one with unwanted resistance & heat in
the HT supply line. With a valve amp, the last thing you want is *more*
heat to wreck the already 'only-just-good-enough' components in old amps!

[Chris Berry]

Post by Robert Manktelow @ work

Post by Robert Manktelow @ work
Hello All
I have a question regarding my valve amps power supply. This is another

one

Post by Robert Manktelow @ work
in the series - sorry.
The mains transformer measures 400 - 0 - 400 VAC.
The transformers centre tap is connected to chassis via a thermister which
drops approx 13VAC (measured using a DVM) when 90mA DC is drawn from the
PSU.
This thermister measures approx 3Kohms cold.
In series between the transformer and with each plate of the GZ34 is a

50ohm

Post by Robert Manktelow @ work
power resistor.
The first capacitor is 50uF, after the GZ34, followed by a 10H choke.
With no input signal I measure 465VDC across this first capacitor and this
voltage drops by 20V at full load (45Watts into a dummy load)
Well, if you got this far thanks:)
However in all of the other valve guitar amps schematics I have seen

there

Post by Robert Manktelow @ work
are no series resistors OR thermisters in the power supply circuit, just
transformer direct to GZ34.
Assuming I can reduce the transformers secondary to achieve the same DC
voltage output can I safely remove both the two resistors and the
thermister?
This modification should (will it?) increase the dynamic response of the
amplifier - that's my aim.
Note - I have fitted a standby switch, the original design did not have

one

Post by Robert Manktelow @ work
of these.
Robert

Thanks for all the help, much appreciated.
Just to fill in a few missing bits in my original post.
This amp has always had valve rectifiers, initially GZ32 then later a GZ34 -
before it got to me.
The op valves were originally KT66 but are now 6L6GC.
Grampian (the manufacturer) made a range of amps (including KT88 based ones)
I assume that they used the same mains transformer for this one (adding
resistors and thermistor) to drop the rail - I think mine was rated at
20Watts originally.
The output transformer secondary measures 80 ohms between each 400V and the
centre tap (total 160 ohms).
So.. judging by your replies my plan is: -
Keep the thermister and the standby switch
Remove the two 50 ohm resistors in the GZ34 plates
Add a MOSFET/zener voltage drop circuit between the transformers centre tap
and ground to "loose" some of the AC to get 465V DC across the first
capacitor.
Replacing the GZ34 with solid state diodes will give me even more of a
problem ( more volts to loose) so I will stay away from this option.

Just go back to GZ32?
cb

[Chris Berry]

Another thought: contact webervst.com and they'll make you something with
the right characteristics.
cb