[ARC5] Command Transmitter circuit design

[Brian]

Hello folks,

What I am about to say has nothing to do with chirp - enough noise has been 
generated and mis-absorbed on that already. However, some of the claims 
about the 1626 and the oscillator circuit need careful consideration.

Those RCA notes are about application. They are not 1626 design parameters - 
you would need access to patent and in-house design engineering info for 
that.

Yes, the app notes do say to keep the potential difference between the 
cathode and its heater low. However, the claim that the 1626 was designed 
for fixed DC bias is not supported by the app note. Further, the required 
grid-cathode bias Voltage mentioned in the app note (at least -35 V for CCS 
operation) cannot be achieved by fixed DC bias in the Command Tx.

So, let's look at the Command Tx circuit design: sure enough, one side of 
the heater is connected to the cathode, the low Voltage side. So, the 
cathode has negligible DC on it - perhaps 10 mV; however, it will have a 
fair amount of RF Voltage. Any bias between the grid and the cathode will be 
developed by rectification of the RF between the grid and the cathode - an 
Armstrong patented idea from his super-regeneration period. This is probably 
also the way that the oscillator output was stabilised. Attach a suitable 
CRO probe in the oscillator circuit - I bet the output is slightly flattened 
as a result of single-ended rectification. Such distortion will not appear 
in the Tx output because of filtering by the RF PA tank circuit.

If you want to run the heater circuit from 12 V, then to maintain the 
bifilar winding in the 1626 heater circuit, merely connect the incoming 12 V 
line to the lead running to the 1629 heater; if you don't want to use the 
1629, this 12 V operation can be easily achieved by shorting pins 2 and 7 on 
the 1629 socket. This will maintain the original oscillator heater design. 
Rewiring the 1625 heaters will present no problems. However, there will be a 
problem with the K-53 relay and the Tx-select relay; this latter can be 
rewired for 12 V operation, but K-53 will need a different solenoid coil. I 
can see no effect from running the heaters on AC or DC: 50 or 60 Hz is so 
far removed from the oscillator frequency that any effect will be massively 
attenuated by the resonant circuit; however, the relays will misbehave. If 
you only wish to use one Tx, easily reversed mods are to jam a toothpick 
under the armature of K-53, and tie the spring contact of the Tx-select 
relay in the selected position.

So, I believe that a small amount of ham hacking is permissible and will 
have little effect, if the above observations are taken into consideration. 
In Australia, such modifications were routinely undertaken by 
government-accredited firms, such as AWA, to provide comms for the Avro 
Anson. So, folks, it's all been done before by the professionals.

73 de Brian, VK2GCE.

On Wednesday, October 08, 2014 2:37 AM , Ken responded to Jim thus:

On 6 Oct 2014 at 20:54, J Mcvey via ARC5 wrote:

> Well, when in doubt refer back to the original design parameters. I looked 
> up
> the 1626 datasheet and I think it answers some of the questions of the Why 
> and
> how of the MO circuit. http://www.r-type.org/pdfs/1626.pdf

Ah, yes! Two "issues", 1) keep the voltage difference between the cathode
and filament as low as possible, and 2) the source for grid bias, how much,
and why.

> So, from what I gleaned from the notes, the circuit was designed to use a 
> fixed
> DC bias otained from the LV line. I believe that applying AC to the 
> filliments
> will affect the MO circuit deleteriously . a) 60HZ modulation b) loss of
> stability due to lack of, or improper bias c) possible damage shortening 
> the
> useable life?

Well, Gentlemen, from all the foregoing discussion it seems very clear to me
now that 1) leaving the circuit exactly as designed, and 2) using DC on the
filaments may go some considerable distance towards a) eliminating some
chirp, and b) contributing towards greater stability.

All of this is what I suspected from the beginning. After all, as I have
REPEATEDLY stated here, IMHO, the designers of our favorite equipment
were sheer geniuses. I have yet to find anything they DIDN'T address.

Conversely, ham-hacking results in WORSE operation of the equipment, not
better.

Ken W7EKB