>
>
>----- Original Message -----
>From: "Steve Thompson" <g8gsq@qsl.net>
>To: <amps@contesting.com>
>Sent: Tuesday, December 18, 2001 6:51 PM
>Subject: Re: [AMPS] Bypassing Screen of 4cx800a for 50 MHz
>
>
>>
>>
>> -----Original Message-----
>> From: Lamb <k7fm@teleport.com>
>> To: Steve Thompson <g8gsq@qsl.net>; amps@contesting.com
>> <amps@contesting.com>
To: <amps@contesting.com>
>> Date: 18 December 2001 12:47
>> Subject: Re: [AMPS] Bypassing Screen of 4cx800a for 50 MHz
>>
>>
>> >Steve asked:
>> >
>> >"This is an honest question, not a carping comment. Is bypassing the
>> cathode
>> >to ground at -350V easier or less critical than bypassing the screen at
>> >+350V?"
>> >
>> >You were probably always causing problems for your teachers - asking
>> >difficult questions just to see if you could stump them.
>>
>> Only when I wanted to understand something better. Interestingly,
>sometimes
>> my questions elicit no response at all.
>>
>> >The 4CX250 has 4 cathode pins. So, if you are cheap and do not want to
>buy
>> >the proven, but expensive, socket with the built in screen bypassing,
>then
>> >it should be easier to bypass the cathode - since you have 4 shots at
>it -
>> >to get a low inductance bypass.
>>
>> Indeed, Rich and Mike made much the same comment. The view appears to be
>> that it's not as critical as screen decoupling. That's what I had thought,
>> but imperfect cathode decoupling puts a reactance rather than a resistance
>> in the way. The volts across the reactance will reduce the gain, but it
>also
>> offers a point where signals can get fed back to the grid. I wondered if
>the
>> phase element might be troublesome - in practice, I guess it isn't.
>>
>> Steve
>>
>
>Its true that the reactance of the cathode decoupling (or is it coupling hi
>hi :-)
>was considered a problem and the author of the 144MHz 1990 ARRL
>design spoke of using 220pF DM19 Silver Mica capacitors with precisely
>1/4" long leads argueing that this was series resonant at 144MHz and was
>semi-critical in the design/implementation, however I later had
>corrospondance
>with two hams in the USA that had built the beast and they said that the
>make, value and mounting arrangements of the cathode (de)coupling caps
>were non-critical :~)
>
>The 144MHz 1990 ARRL design has around 25-26dB gain and one that
>I used back in 1987 (Mick G1EZF's) was unconditionally stable, needed
>(had) no neutralisation and produced >900W out driven with an IC202S
>barefoot.
>
>If I were going to build a single Tetrode amp on 6m or 2m I'd consider
>running the screen at deck - RF simplicity at (slight) HT complexity (ie.
>need for a B- lead).
>
// A chassis-grounded/earthed screen solves the screen's bypass-C
resonance problem, however, an anode (B+) to ground arc can briefly drive
the HV negative lead to 1000s of volts. Unless precautions are taken,
the arc may cause damage to components that are connected to the HV
negative In a typical filament-type tube amplifier, such as 2,
3-500Zs, such an arc usually will not breakdown the filament-transformer
secondary insulation, but it will often damage the grid-current meter
shunt and/or the grid current meter. In a typical heater-type tube
amplifier, provided that one side of the heater is grounded, a HV+ arc to
chassis can cause the cathode to briefly rise to several kV negative -
which causes an arc between the cathode and the grounded heater (typical
cathode-heater rating: 150v-max). The secondary arc can burn the heater
wire in two, thereby destroying the tube. If the tube is an 8877 or
3cx800A7, the cost is fairly steep.
- The Fix: Unground the heater, connect one side of the heater to the
cathode, add a ?10 - 20 uH, trifilar choke to carry AC heater current and
DC cathode current, add 200a glitch diodes across the grid-current meter
shunt and across the meter movement. Also, c. 20?s of glitch R needs to
be added to the +HV circuit to limit fault current to <200A during a +HV
arc. However, no such precautions are needed for those with a free
supply of new tubes.
- Yesterday, I received a telephone call from an Alpha owner whose model
77 has an 8877 with a burnt-out heater, He was puzzled by the burnt-out
heater because the Alpha-77 has a step-start circuit. I explained that
inrush current to the heater is seemingly not the problem. Although
Eimac sets inrush current limits for some of their thoriated-tungsen
filament tubes, I have seen no such limit for their heater-type tubes.
The likely reason is that filament-type tubes have ?8.3 times less cold-R
(and 8.3 times more cold-current) than hot. Heater-type tubes have much
less difference in hot/cold R (and I) because they operate at a much
lower temperature.
>
Murphy was right -- If anything can possibly go wrong, it will ... ...
... .
cheers, Mike
- R. L. Measures, 805.386.3734,AG6K, www.vcnet.com/measures.
end
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|