[Amps] 3.5 kV 2A REGULATED Power Supply: Schematic ?

[R.Measures]

>
>Rich said:
>
>>Stabilizing the screen supply with a series pass regulator is not
>>good engineering practice unless a screen supply bleeder is used to
>>protect against reverse screen current.
>
>I didn't make it clear that the  'low impedance' screen supply means low
>impedance both for source and sink - i.e good regulation for both source
>and sink..  For 6146B's, I use zeners - not ideal, but the screen current
>doesn't go very negative - rarely more than 1 or 2mA.
>
**  Zeners are plenty good enough at 1/3 of their dissipation rating --  
and they are fail-safe.  For screen potentials of less than 700v, a zener 
string is a good choice.

>>Stabilizing the grid supply in AB1 is of little value since there is
>>neither forward or reverse grid current.
>
>Stabilising has a lot of value if the screen supply is stabilised.
>Regulation is the parameter that has little value. Stabilising both is best
>to get around mains variations.

**   However, both of the Ugly Amplifiers on my Web site were sold to 
operators who eventually experienced trouble with the screen-regulator 
electronics.  As a result, the mo'-linear amplifier I am currently 
gathering parts for will use a 1500V or so resonant-choke screen supply 
with a 30mA bleeder.  Hopefully, this will be adequate.  If not, I will 
add a FET-pass positive floating regulator shown in:
http://www.somis.org/D.a.08.GIF
However, since the speaking screen-I on a 4cx10-kJ is only c, 35mA 
average, the regulation shouldn't be all that bad.  

--  RE:  mains variations --  With an AB1 tetrode, when the mains sag, 
the anode-potential sags and  ZSAC drops.  If the grid bias supply 
potential also sags, ZSAC increases - which tends to offset the ZSAC drop.

cheers, Peter.
>
>Definitions:
>'Regulation' is a measure of the variation of output voltage with load.
>'Stabilisation' is a measure of the variation of output voltage with input
>(mains) variation.
>
>
>73
>
>Peter G3RZP
>
>
>
>
>
>
>                                                                           
>                                                   
>                      "R. Measures"                                        
>                                                   
>                      <r at somis.org>            To:       "Peter C," 
><peter.chadwick at Zarlink.Com>, "Ian White, G3SEK"          
>                                                <g3sek at ifwtech.co.uk>      
>                                                   
>                      05-Jan-2004 01:31        cc:       " AMPS" 
><amps at contesting.com>, <amps-bounces at contesting.com>         
>                                               Subject:  Re: [Amps] 3.5 kV 
>2A REGULATED Power Supply: Schematic ?             
>                                                                           
>                                                   
>
>
>
>
>**  Stabilizing the screen supply with a series pass regulator is not
>good engineering practice unless a screen supply bleeder is used to
>protect against reverse screen current.   Stabilizing the grid supply in
>AB1 is of little value since there is neither forward or reverse grid
>current.
>
>>I have found problems when using high impedance supplies with 6146s,
>>especially  as they get older.  Especially bad is the Yaesu 'anti flash
>>over diode' in series with the screen grid - any negative screen current
>>pushes up the screen volts, and the anode current runs away, with
>>disastrous results for the valves.
>>
>>Grid bias and screen  supplies should either both be stabilised, or
>>neither. If neither supply is stabilised, variations in supply volts tend
>>to cancel. I prefer to stabilise them both, and keep the supply impedance
>>(at low frequencies, including zero frequency) low. I'm still using the
>>original 6146Bs in my FT102, and after 20 years of hard work, they're
>still
>>fine.
>>
>>73
>>
>>Peter G3RZP
>>
>>
>>_______________________________________________
>>Amps mailing list
>>Amps at contesting.com
>>http://lists.contesting.com/mailman/listinfo/amps
>>
>
>
>
>
>
>