[Amps] Modern Op-Amp shunt stabilizator power supply (for tetrodes cr

Thu Mar 3 13:52:38 EST 2022

Hi Jeff,
Here are some thoughts on short circuit current that others may be able to expand upon.  
Let me use my SB-220 as an example.  

The plate voltage is about 3,000V and the effective power supply output capacitance is 
25fuF.   I have a 15 ohm glitch resistor in the plate circuit.  Allowing about 5 more 
ohms of circuit resistance and 10 ohms for the grid-to-ground short, the total resistance 
is about 30 ohms.  This is just a rough guess.  The initial current is about 
3,000V/30ohms or approximately100 amps.  The time constant with 25uF and 30 ohms is 
~3/4millisecond.  At the end of one time constant, the voltage is now 1,125 volts (37% 
Vmax).  

The AVERAGE current during the first time constant is C*(deltaV)/(deltaT)= 
25e-6*(3,000-1125)/.00075= 62.5 amps.  During the second time constant (37% to 13.5% 
Vmax), the current will be 25e-6*(1125V-405V)/.00075 = 24 amps.  I have ignored the 
transformer because I am assuming that the inductance is large enough so the transformer 
does not contribute much to the current flow during the first two time constants.

If the inductance of transformer secondary is known, the transformer's contribution to 
the total current could be added in at the appropriate time.

This should provide a starting point.  In general, I would expect that the current will 
be under 100 amps and decay towards zero in a few milliseconds.
        Gl & 73
.... Bill K3HZP

---------- Original Message ----------
From: "hzp_electronics at juno.com" <hzp_electronics at juno.com>
To: wa1hco at gmail.com
Subject: Re: [Amps] Modern Op-Amp shunt stabilizator power supply (for tetrodes cr
Date: Thu, 3 Mar 2022 18:48:34 GMT

Hi Jeff,
Here are some thoughts on short circuit current that others may be able to expand upon.  Let me use my SB-220 as an example.  

The plate voltage is about 3,000V and the effective power supply output capacitance is 25fuF.   I have a 15 ohm glitch resistor in the plate circuit.  Allowing about 5 more ohms of circuit resistance and 10 ohms for the grid-to-ground short, the total resistance is about 30 ohms.  This is just a rough guess.  The initial current is about 3,000V/30ohms or approximately100 amps.  The time constant with 25uF and 30 ohms is ~3/4millisecond.  At the end of one time constant, the voltage is now 1,125 volts (37% Vmax).  

The AVERAGE current during the first time constant is C*(deltaV)/(deltaT)= 25e-6*(3,000-1125)/.00075= 62.5 amps.  During the second time constant (37% to 13.5% Vmax), the current will be 25e-6*(1125V-405V)/.00075 = 24 amps.  I have ignored the transformer because I am assuming that the inductance is large enough so the transformer does not contribute much to the current flow during the first two time constants.

If the inductance of transformer secondary is known, the transformer's contribution to the total current could be added in at the appropriate time.
This should provide a starting point.  In general, I would expect that the current will be under 100 amps and decay towards zero in a few milliseconds.
        Gl & 73
.... Bill K3HZP

---------- Original Message ----------
From: jeff millar <wa1hco at gmail.com>
To: Steve Thompson <g8gsq72 at gmail.com>
Cc: Amplifier Mailing List <amps at contesting.com>
Subject: Re: [Amps] Modern Op-Amp shunt stabilizator power supply (for tetrodescr
Date: Thu, 3 Mar 2022 10:14:46 -0500

Consider the arc as a zero ohm resistor.  The current peak is limited by
the glitch resistor and the impedance of the HV caps

jeff, wa1hco

On Thu, Mar 3, 2022 at 8:35 AM Steve Thompson <g8gsq72 at gmail.com> wrote:

> A while ago I breadboarded a screen regulator supply with some
> similarities to Jim's design. One of these was that the 'pull down'
> transistor could act as a clamp if there was a flashover to the screen.
>
> I was fortunate enough not to test that function but I had a couple of
> questions which I never found answers for:
>
> How much current flows during a flashover to the screen (assuming the
> amp has a sensible glitch resistor in the HV supply) and
>
> How long does a flashover last?
>
> Worst case guesses are simple enough but I don't know how the physics
> of the arc affects things in practice.
>
> Steve G8GSQ
>
> On Mon, 28 Feb 2022, 21:37 Jim Garland, <4cx250b at miamioh.edu> wrote:
>
> > Attached is the screen voltage regulator I designed for my 160m/80m
> > duo-band amplifier using three GU74B tetrodes. From your msg, I suspect
> > it’s similar in some ways to the circuit you’ve described. The circuitry
> is
> > deceptively subtle, however, in that it provides protection against
> damage
> > from grid-plate flashovers in the tubes, which can be a problem,
> especially
> > with new (NOS) Russian tetrodes. I’ve used this circuit for more than
> five
> > years and found it to be always stable, with tight regulation. The
> > circuitry mounts on a small file card-sized PCB. The screen voltage is
> > adjustable and I set it at a nominal 340V.
> >
> >
> >
> > If the attachment doesn’t come through (I’m away from my home QTH and
> have
> > limited internet coverage, here’s a link to the complete schematic
> package
> > for the amplifier. Pse email me if you can’t get any of the diagrams to
> > load!
> >
> >
> >
> >
> http://www.w8zr.net/160amplifier/images/Amplifier%20Schematic%20Package.pdf
> >
> > 73,
> >
> > Jim W8ZR
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps