power tube protection diode

Status
Not open for further replies.

apa

Distinguished
Jul 28, 2004
53
0
18,580
Archived from groups: rec.audio.pro (More info?)

A bit off topic for RAP, but I thought someone here might know right
off.
I was replacing burnt plate resistors in a Fender Blues Deville and
noticed a diode on each of the power tubes reverse biased between the
plate and cathode. I assume this would only conduct if the plate
voltage went negative (or exceeded PIV and the diode shorted out). What
does this do in terms of circuit protection (I'm assuming that's what
it's there for)? Under what condition would the plate voltage actually
go below 0?
 

apa

Distinguished
Jul 28, 2004
53
0
18,580
Archived from groups: rec.audio.pro (More info?)

Pooh Bear wrote:
> apa wrote:
>
> > A bit off topic for RAP, but I thought someone here might know right
> > off.
> > I was replacing burnt plate resistors in a Fender Blues Deville and
> > noticed a diode on each of the power tubes reverse biased between the
> > plate and cathode. I assume this would only conduct if the plate
> > voltage went negative (or exceeded PIV and the diode shorted out). What
> > does this do in terms of circuit protection (I'm assuming that's what
> > it's there for)? Under what condition would the plate voltage actually
> > go below 0?
>
> That's a novel one ! New to me and I've seen a few.
>
> You're sure it was plate / anode - cathode ? What kind of diode btw ?
>
> Graham

Well the cathode is grounded - so plate to ground. I'm 99% sure that's
what it was, but I'll look again to be sure.
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

apa <tacoma57@hotmail.com> wrote:
>
>Well the cathode is grounded - so plate to ground. I'm 99% sure that's
>what it was, but I'll look again to be sure.

Check to see that it's not the screen grid to ground! That would seem
like a reasonable thing to have.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

apa wrote:

> A bit off topic for RAP, but I thought someone here might know right
> off.
> I was replacing burnt plate resistors in a Fender Blues Deville and
> noticed a diode on each of the power tubes reverse biased between the
> plate and cathode. I assume this would only conduct if the plate
> voltage went negative (or exceeded PIV and the diode shorted out). What
> does this do in terms of circuit protection (I'm assuming that's what
> it's there for)? Under what condition would the plate voltage actually
> go below 0?

That's a novel one ! New to me and I've seen a few.

You're sure it was plate / anode - cathode ? What kind of diode btw ?

Graham
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On 17 Jun 2005 07:48:32 -0700, "apa" <tacoma57@hotmail.com> wrote:

>A bit off topic for RAP, but I thought someone here might know right
>off.
>I was replacing burnt plate resistors in a Fender Blues Deville and
>noticed a diode on each of the power tubes reverse biased between the
>plate and cathode. I assume this would only conduct if the plate
>voltage went negative (or exceeded PIV and the diode shorted out). What
>does this do in terms of circuit protection (I'm assuming that's what
>it's there for)? Under what condition would the plate voltage actually
>go below 0?

This apparently protects from overvoltage/arcing in the output
transformer primary and connected components. This won't normally
happen even at full output with speakers connected, but tube amps can
be damaged if played with the load (speaker) disconnected. High signal
levels under a no-load situation, especially clipping, can cause
inductive 'kick' and high voltage spikes (much higher than the already
high operating voltage on the tubes), high enough to cause arcing
inside the transformer, in the tube, at the tube socket or elsewhere
in the circuitry, and generally causing component damage with
overvoltage.
Specifically, with a high-level input signal into a push-pull
output stage, if one tube conducts hard and then suddenly switches
off, the transformer (with no load, effectively an inductor) voltage
will go up to several times the quiescent plate voltage (and the other
plate will likewise go several times below that), but the diode on the
other side will limit that side's voltage to zero, and through
transformer action the side of the primary that just turned off will
be limited to twice the plate voltage.

Someone asked what type diodes these are, I'm guessing 1N4007 (a
few pennies each in quantity), 1A 1000 PIV, good for outputs powered
at 450V where they should see a peak reverse voltage of no more than
900V.

-----
http://www.mindspring.com/~benbradley
 

apa

Distinguished
Jul 28, 2004
53
0
18,580
Archived from groups: rec.audio.pro (More info?)

Scott Dorsey wrote:
> apa <tacoma57@hotmail.com> wrote:
> >
> >Well the cathode is grounded - so plate to ground. I'm 99% sure that's
> >what it was, but I'll look again to be sure.
>
> Check to see that it's not the screen grid to ground! That would seem
> like a reasonable thing to have.
> --scott
> --
> "C'est un Nagra. C'est suisse, et tres, tres precis."

Definitely the plate. What good would a screen to ground do?
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On Fri, 17 Jun 2005 16:48:32 +0200, apa wrote:

> A bit off topic for RAP, but I thought someone here might know right
> off.
> I was replacing burnt plate resistors in a Fender Blues Deville and
> noticed a diode on each of the power tubes reverse biased between the
> plate and cathode. I assume this would only conduct if the plate voltage
> went negative (or exceeded PIV and the diode shorted out). What does
> this do in terms of circuit protection (I'm assuming that's what it's
> there for)? Under what condition would the plate voltage actually go
> below 0?

The load of the power tubes is highly inductive: a transformer. This can
result in high voltage peaks. If one side of the transformer is going up,
the other side is goging down and will cross zero. A diode from anode to
cathode clamps that peak, and there is a nice resistor to absorb the
energy. When I was using tubes, I sometimes used spark plugs as an
overvoltage protection, low capacitance and very robust.

--
Chel van Gennip
Visit Serg van Gennip's site http://www.serg.vangennip.com
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

Chel van Gennip wrote:

> On Fri, 17 Jun 2005 16:48:32 +0200, apa wrote:
>
> > A bit off topic for RAP, but I thought someone here might know right
> > off.
> > I was replacing burnt plate resistors in a Fender Blues Deville and
> > noticed a diode on each of the power tubes reverse biased between the
> > plate and cathode. I assume this would only conduct if the plate voltage
> > went negative (or exceeded PIV and the diode shorted out). What does
> > this do in terms of circuit protection (I'm assuming that's what it's
> > there for)? Under what condition would the plate voltage actually go
> > below 0?
>
> The load of the power tubes is highly inductive: a transformer. This can
> result in high voltage peaks. If one side of the transformer is going up,
> the other side is goging down and will cross zero. A diode from anode to
> cathode clamps that peak,

No damn good for an output stage !

It *relies* on the tube with -ve a-k volts being reverse biased at the grid.

That's how tube push-pull amps work !

Graham
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On Fri, 17 Jun 2005 18:49:56 +0200, Pooh Bear wrote:

>> The load of the power tubes is highly inductive: a transformer. This
>> can result in high voltage peaks. If one side of the transformer is
>> going up, the other side is goging down and will cross zero. A diode
>> from anode to cathode clamps that peak,
>
> No damn good for an output stage !
>
> It *relies* on the tube with -ve a-k volts being reverse biased at the
> grid.

You have not understand what I wrote.

> That's how tube push-pull amps work !

When I was working with tubes they never showed any current with a
negative a-k potential.

--
Chel van Gennip
Visit Serg van Gennip's site http://www.serg.vangennip.com
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

Chel van Gennip wrote:

> On Fri, 17 Jun 2005 18:49:56 +0200, Pooh Bear wrote:
>
> >> The load of the power tubes is highly inductive: a transformer. This
> >> can result in high voltage peaks. If one side of the transformer is
> >> going up, the other side is goging down and will cross zero. A diode
> >> from anode to cathode clamps that peak,
> >
> > No damn good for an output stage !
> >
> > It *relies* on the tube with -ve a-k volts being reverse biased at the
> > grid.
>
> You have not understand what I wrote.
>
> > That's how tube push-pull amps work !
>
> When I was working with tubes they never showed any current with a
> negative a-k potential.

They won't. It's impossible. Emission is going the wrong way.

You might get nasty flashovers between grids though.

That why I suggested the control grid should be driven negative.

Graham
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

Chel van Gennip wrote:

> On Fri, 17 Jun 2005 18:49:56 +0200, Pooh Bear wrote:
>
> >> The load of the power tubes is highly inductive: a transformer. This
> >> can result in high voltage peaks. If one side of the transformer is
> >> going up, the other side is goging down and will cross zero. A diode
> >> from anode to cathode clamps that peak,
> >
> > No damn good for an output stage !
> >
> > It *relies* on the tube with -ve a-k volts being reverse biased at the
> > grid.
>
> You have not understand what I wrote.
>
> > That's how tube push-pull amps work !
>
> When I was working with tubes they never showed any current with a
> negative a-k potential.

Sorry, it's unseasonably warm here and my brain went to sleep !

The push-pull stage shouldn't result in -ve a-k volts.

I expect the diodes are indeed a protective measure against inductive
flyback caused by no load or load mismatch.

Graham
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On Fri, 17 Jun 2005 20:19:03 +0200, Pooh Bear wrote:

> You might get nasty flashovers between grids though.

Flashovers (conductivity by ionisation?) are very rare in vacuum tubes,
not so rare on tube sockets.

--
Chel van Gennip
Visit Serg van Gennip's site http://www.serg.vangennip.com
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On Fri, 17 Jun 2005 17:43:34 GMT, Ben Bradley
<ben_nospam_bradley@frontiernet.net> wrote:

> This apparently protects from overvoltage/arcing in the output
>transformer primary and connected components.

<excellent analysis snipped for bandwidth>

In other words, this is how you get to "11".

Chris Hornbeck
 
G

Guest

Guest
Archived from groups: rec.audio.pro (More info?)

On Fri, 17 Jun 2005 19:43:34 +0200, Ben Bradley wrote:

> This apparently protects from overvoltage/arcing in the output
> transformer primary and connected components. This won't normally happen
> even at full output with speakers connected, but tube amps can be
> damaged if played with the load (speaker) disconnected. High signal
> levels under a no-load situation, especially clipping, can cause
> inductive 'kick' and high voltage spikes (much higher than the already
> high operating voltage on the tubes),....

If the amplifier is clipping, high voltage spikes can occur even with
load. The load at the tube annode always is quite inductive at high
frequencies.

> ..high enough to cause arcing inside the transformer, in the tube, at
> the tube socket or elsewhere in the circuitry, and generally causing
> component damage with overvoltage.

Arcs (conductivity by ionisation) wil not occur in vacuum tubes.

> Someone asked what type diodes these are, I'm guessing 1N4007 (a few
> pennies each in quantity), 1A 1000 PIV, good for outputs powered at 450V
> where they should see a peak reverse voltage of no more than 900V.

I would recommend fast diodes with a PIV at least 3 times the voltage of
the power supply.

--
Chel van Gennip
Visit Serg van Gennip's site http://www.serg.vangennip.com
 
Status
Not open for further replies.