[Boatanchors] 30S1 bis query

[Gary Schafer]

Hi Jack,

Glad to see you got it going! But a 10 amp fuse for the screen supply sounds
mighty big. Put the right size in it! Those parts are expensive.

As to the 20' cable to the exciter; The 30S1 is actually a grounded grid
amplifier with the cathode driven except it has grid and screen voltages
just like a conventional grid driven AB1 amplifier. The only difference here
is that the drive is applied to the cathode instead of the grid. No grid
current should be drawn on the 4CX1000.

With a cathode driven amplifier normally there is a pi network input tuned
circuit to match the impedance and to help provide lower intermodulation
distortion. Part of the job of the pi network circuit is to provide
capacitance directly to ground at the cathode of the tube so that the
harmonics of the plate current pulses have a direct low impedance path to
ground. That is what reduces IM distortion. The plate is in series with the
cathode so the plate current pulses are seen at the cathode.
The 30S1 does not have a tuned input circuit.

If you place a pi network tuned input circuit some distance away from the
tube, then depending on the length of the interconnecting cable, it may
reflect back some impedance to the cathode other than the low impedance
wanted that the pi network capacitor provides. In this case the pi network
is located in the exciter (Collins KWM2 or 32S1). If the interconnecting
cable length is carefully chosen it will reflect the proper low impedance of
the pi network output capacitor in the exciter up to the cathode of the tube
in the 30S1. If some other length is used then that impedance seen at the
cathode could be all over the map. It could actually be lower, which would
be good, but then it could also be higher which is not good.
The 20.5' length is no doubt a compromise on some bands but provides enough
of an impedance transformation to do the job.

If you are driving the amp with a more modern exciter that does not have a
pi network output or you use a tuner that is a T type tuner, as most are,
you may not have a capacitor to ground in the network that feeds the amp.
This will not give the proper impedance at the harmonics in the tube and IM
products will be higher.

73
Gary  K4FMX

> -----Original Message-----
> From: boatanchors-bounces at mailman.qth.net [mailto:boatanchors-
> bounces at mailman.qth.net] On Behalf Of Jack Taylor
> Sent: Wednesday, August 16, 2006 6:36 PM
> To: boatanchors at mailman.qth.net
> Subject: Re: [Boatanchors] 30S1 bis query
> 
> Regarding the gremlins in our 30S1 amplifier, today saw a major
> breakthrough.
> Following Gary's suggestion, we were able to measure the screen voltage
> without
> being electrocuted and still keeping the interlock system in place.  The
> first test
> point yielded a measurement of approximately 12 VDC.  Hmm, we thought,
> that
> can't be right so we found another measurement point.  This one yielded
> approximately
> 12 VDC.  These readings appeared both in the CW and SSB mode positions.
> In
> CW we were seeing over 500 watts output with drive and normal indications
> on
> all meter positions.
> 
> Puzzling over this, we pulled the 12 amp screen supply fuse.  It was open!
> Whether it just got tired or it blew due to excessive screen current, we
> haven't
> yet determined.  To be on the safe side we replaced it with a 10 amp fuse,
> crossed our fingers and applied high voltage.  It held.  We then went
> ahead
> and tuned up the amplifier.  It takes about -55 volts of bias to adjust
> the
> plate
> current meter for 200 ma in SSB mode with no drive applied.  In CW mode,
> we gained about 200 watts over that delivered when there was no screen
> voltage, strange!
> 
> Gentlemen, thank you for your suggestions.!  Getting acquainted with the
> 30S1 has been an interesting experience for me.
> 
> One additional mystery remains.  What is the theory/technology behind
> using
> a 20.5 ft length of "RF cable" between the Collins exciters and the
> amplifier?
> The instruction manual says "This length is optimum in maintaining the
> low-
> distortion figure for which the equipment was designed."  Does anyone have
> insight as to what's going on here?
> 
> 73 de Jack --  N7OO