[Amps] Alpha 77Dx -Adding Grounded Grid Filament Choke

[Tony King - W4ZT]

Tomm,

I believe that what Rich is saying here is that we should treat the 
GS-35B just as though it were a directly heated cathode when dealing 
with the question of protection for the cathode and heater circuit.  A 
bifiler choke with a center tapped filament transformer along with the 
normal protection discussed for B- and metering are in order.  The 
GS-35B, as Will said, is unique because of the internal connection 
between the indirectly heated cathode and the filament. The choke should 
be capable of not only handling the filament current plus the normal 
plate current but also be sized up enough to handle a large glitch. This 
is probably only a factor if building a single tube amp in which one 
might be tempted to down size the choke because of the low filament 
current requirement.

Our friends making triode boards all recognize the need for and 
recommend protection against the B- rail soaring when we're visited by 
the big glitch. Rich's suggestion for the series string of 200 peak amps 
diodes is a common method for providing that protection.

73, Tony W4ZT

R.Measures wrote:
> On Mar 5, 2005, at 9:54 AM, Paul Hewitt wrote:
> 
> 
>>Hi Tomm
>>With the GS-35B, (cathode tied to filament internally) I would use a
>>filament choke and ground one side of the filament secondary, 
>>transformer
>>side of the choke of course.  A spark gap at the cathode wouldn't hurt 
>>but
>>might be over-kill.
> 
> 
> I would:  not ground either side of the heater supply, use a glitch-R 
> to limit I-pk to 200a, and I would put a string of 20, 200a pk glitch 
> diodes from the HV neg to gnd.
> 
> 
>>Paul
>>
>>PAUL HEWITT
>>WD7S PRODUCTIONS
>>QRO HOMEBREW COMPONENTS
>>http://home.earthlink.net/~wd7s
>>
>>
>>
>>----- Original Message -----
>>From: "Tomm Aldridge" <KD7QAE at ARRL.NET>
>>To: "Paul Hewitt" <wd7s at earthlink.net>
>>Cc: "R. Measures" <r at somis.org>; "AMPS" <amps at contesting.com>
>>Sent: Saturday, March 05, 2005 9:45 AM
>>Subject: Re: [Amps] Alpha 77Dx -Adding Grounded Grid Filament Choke
>>
>>
>>
>>>As I understand this, I really should be adding a very high speed 
>>>clamp
>>>to ground from each of my GS35B cathode connections in the form of a
>>>triggered device or a spark gap.  Correct?  What I don't understand is
>>>what failure or operational modes cause thsi to happen in a well
>>>designed amp?
>>>
>>>Tomm
>>>
>>>Paul Hewitt wrote:
>>>
>>>>The gap I'm using is a Bourns 2035 series, 90 volt, 20KA (single 
>>>>pulse),
>>>>hope the glitch resistor holds !!  The gap through Mouser is $3.22
>>
>>single
>>
>>>>piece price
>>>>
>>>>Arc to ground or internal arc to cathode, same problem for the 
>>>>filament.
>>
>>If
>>
>>>>the filament secondary is not grounded the cathode to filament 
>>>>breakdown
>>>>rating becomes 150v plus the transformer winding breakdown rating, 
>>>>(or
>>>>feed-thru cap breakdown rating, connecting wire insulation, etc).  
>>>>The
>>>>90-150 volt spark gap still saves the filament.
>>>>
>>>>I have had unintentional spark gaps between the B- buss and ground of
>>
>>.5"
>>
>>>>fire,(with catastrophic results) while a 1N5408 "glitch" diode sat 
>>>>there
>>
>>and
>>
>>>>smiled at me.  The spark gap is faster than the diode, race them
>>
>>sometime.
>>
>>>>I now use a spark gap paralleled with the "glitch diode plus a spark 
>>>>gap
>>
>>at
>>
>>>>the cathode and all is good in river city.
>>>>Paul
>>>>
>>>>PAUL HEWITT
>>>>WD7S PRODUCTIONS
>>>>QRO HOMEBREW COMPONENTS
>>>>http://home.earthlink.net/~wd7s
>>>>
>>>>
>>>>
>>>>----- Original Message -----
>>>>From: "R. Measures" <r at somis.org>
>>>>To: "Paul Hewitt" <wd7s at earthlink.net>
>>>>Cc: "AMPS" <amps at contesting.com>; "Paul Christensen" <w9ac at arrl.net>
>>>>Sent: Saturday, March 05, 2005 8:14 AM
>>>>Subject: Re: [Amps] Alpha 77Dx -Adding Grounded Grid Filament Choke
>>>>
>>>>
>>>>
>>>>
>>>>>On Mar 5, 2005, at 7:57 AM, Paul Hewitt wrote:
>>>>>
>>>>>
>>>>>
>>>>>>Hi Rich
>>>>>>For the 8877 at HF you can save the filament choke, a simple gas
>>>>>>discharge
>>>>>>spark gap at the cathode will fire long before an internal arc to
>>>>>>cathode
>>>>>>will jump to the filament.
>>>>>
>>>>>What is the peak-I rating of the gas-discharge device, the V-drop
>>>>>across it at the peak-I rating, and what is its cost?
>>>>>
>>>>>
>>>>>
>>>>>>The discharge tube insures the cathode stays no
>>>>>>more than the gaps voltage rating above ground potential during a 
>>>>>>HV
>>>>>>arc to
>>>>>>cathode
>>>>>
>>>>>The problem is not a HV arc to cathode, the problem is a +HV arc to
>>>>>chassis ground -- which in turn tries to elevate the cathode to the 
>>>>>neg
>>>>>amount of HV.  Since this is typically several neg. kV, and the 
>>>>>max-V
>>>>>rating between the heater and the cathode is c, 150v, there's 
>>>>>trouble
>>>>>in river city.
>>>>>
>>>>>
>>>>>
>>>>>>and will fire long before a "glitch diode" will turn on and 
>>>>>>conduct.
>>>>>
>>>>>Glitch diodes are thoroughly on at 1v per junction.
>>>>>
>>>>>
>>>>>
>>>>>>Eimac recommends this on the 8877's bigger brothers. Un-grounding 
>>>>>>the
>>>>>>filament secondary is a good idea also, no sense having the 
>>>>>>filament
>>>>>>looking
>>>>>>like a lower impedance path to ground than the spark gap !!
>>>>>>Cheers
>>>>>>
>>>>>>PAUL HEWITT
>>>>>>WD7S PRODUCTIONS
>>>>>>QRO HOMEBREW COMPONENTS
>>>>>>http://home.earthlink.net/~wd7s
>>>>>>
>>>>>>
>>>>>>
>>>>>>----- Original Message -----
>>>>>>From: "R.Measures" <r at somis.org>
>>>>>>To: "Paul Christensen" <w9ac at arrl.net>
>>>>>>Cc: "AMPS" <amps at contesting.com>
>>>>>>Sent: Saturday, March 05, 2005 12:55 AM
>>>>>>Subject: Re: [Amps] Alpha 77Dx -Adding Grounded Grid Filament Choke
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>>On Mar 4, 2005, at 6:30 AM, Paul Christensen wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>Whoops - my mistake.  The 77Dx's 8877 does not tie the cathode 
>>>>>>>>and
>>>>>>>>filament
>>>>>>>>together.  One side of the filament is grounded.  Would it still 
>>>>>>>>be
>>>>>>>>advisable to add the common-mode filament choke anyway?
>>>>>>>
>>>>>>>Not if you have a free supply of 8877s.
>>>>>>>-  I would add a bifilar RFC, unground the filament/heater 
>>>>>>>winding,
>>>>>>>and
>>>>>>>wire the cathodes to one side of the heater.
>>>>>>>
>>>>>>>
>>>>>>>>-Paul, W9AC
>>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>Richard L. Measures, AG6K, 805.386.3734.  www.somis.org
  > Richard L. Measures, AG6K, 805.386.3734.  www.somis.org