[Johnson] Pacemaker

Thu Sep 28 23:05:06 EDT 2017

    Very interesting story. About the resistor, the rise in resistance 
probably comes from the way the leads are attached. If the resistor has 
end caps, fat ends, its possible the connection has become oxidized. I 
think this is rare but can happen.
    Also, my experience with the manufacture of deposited carbon and 
metal resistors is pretty old but when I worked in the industry both 
kinds were trimmed by spirialing although some low values were adjusted 
by linear scraping of the coating. In any case, recent measurements of 
resistor reactance using a Boonton 250A RX meter shows that deposited 
film resistors have _less_ reactance than carbon composition resistors. 
In fact, there is an illustration in the RX meter handbook showing a 
comparison. I measured a number of modern resistors up to about 10Kohm 
and up to 250Mhz (the limit of the instrument). At least up to a couple 
of thousand ohms the film resistors had very low reactance. Of course, 
at 250Mhz the leads become a source of reactance regardless of the value 
or type of resistor.
    Two types of lead fastening are used in film resistors, staked end 
caps and conductive epoxy paint. The factory I worked at made 
established reliability parts for NASA contractors and also MIL spec 
parts. The established reliability resistors were always made with 
cemented leads. They were so strong the ceramic blanks would break in 
tension tests before the leads came off. The staked parts could also be 
reliable if the ends of the blanks were prepared correctly and the 
staking done carefully. I think modern resistors are made in about the 
same way but trimming is now done with laser cutters instead of carbide 
or diamond saws.

On 9/28/2017 7:45 PM, Robert Nickels wrote:
> I thought I'd post a follow-up since there's not much info about 
> troubleshooting the Pacemaker available for others who may come across 
> this thread.
> 
> Unlike many other SSB exciters, the Pacemaker is a single conversion 
> design where the SSB generator operates on a different frequency for 
> each band, 3 Mhz below the lower band edge.  A single mixing operation 
> then takes the signal to the desired frequency, e.g. for 80 meters, the 
> SSB signal is generated at 500KHz, which is mixed with the 3.0-3.5 Mhz 
> VFO.     This makes alignment a bit more tedious than an HT-37 or 20A 
> since there are five RF phasing adjustments to make instead of just one.
> 
> The problem I had was the none of them was working right, and since I 
> knew the audio phasing was OK, that meant the problem had to be 
> something common in the RF phasing unit.   This circuitry is on an 
> L-shaped sub-chassis which is totally inaccessible, so it must be 
> removed from the transmitter.    This is does not represent EF Johnson's 
> finest packaging effort, because to get access requires that the front 
> panel be removed (including all knobs, control hardware, and the VFO 
> bezel).  Removing the RF phasing unit requires disconnecting four wires, 
> removing 3 nuts and 2 screws, and disconnecting the dial cord.   If you 
> masking-tape the dial cord temporarily in position on the bandswitch 
> drum you can disconnect both ends at the RF phasing switch drum and 
> reconnect them afterward with minimum frustration.  Once removed from 
> the transmitter, the RF phasing unit can be powered-up on the bench by 
> applying filament and 300VDC.    I confirmed that coil adjust ment could 
> not produce the required + and - 45 degree lead/lag phasing at the outputs.
> 
> The Pacemaker uses an R-L-C RF phasing network with values appropriate 
> to each range (see page 59 of Jack Brown's SSB Techniques book for an 
> example).   As I originally suspected the coils and dipped mica caps 
> were fine - the problem was the 200 ohm 1% deposited carbon resistor 
> feeding one side of the network was over 10 ohms high, creating a 
> permanent imbalance that was beyond the range of adjustment by the 
> slug-tuned inductors.    Because I was a bit afraid of self-inductance 
> with modern helically-trimmed resistors, I opted to sort out a 220 ohm 
> carbon comp that matched the value of the resistor in the other leg. I 
> later found an article in Electric Radio #284 that described a similar 
> problem with the Pacemaker where one of these original 1% resistors was 
> completely open.   Like the author I'm unable to explain how this 
> failure can occur with only a few volts and essentially no current 
> through these components, but it seems to be a common failure point.
> 
> Reassembly and alignment using Johnson's  specified 2KHz audio tone 
> produced much improved results and typically 30-35 dB of opposite 
> sideband suppression.   All alignment was done using a simple RTL-SDR 
> and SDR# software which provides excellent resolution and an accurate 
> DBFS scale, and which makes adjusting any phasing rig vastly easier than 
> any other method.
> 
> A word to the wise Pacemaker-repairman:   in addition to 110VAC for the 
> Tx light, the plate supply voltage is present on the meter terminals 
> which are uncomfortably close to several audio phasing adjustment 
> pots.   Don't ask why I know this, (!) but use an insulated screwdriver 
> to reach those pots!  Shorting HV to ground WILL get your attention (and 
> hopefully the meter will survive) and if you end up needing to replace 
> the plate meter shunt, it's a standard .51 ohm job that's readily 
> available.    Another cautionary note - there is B+ on the exposed shaft 
> of the driver neutralizing variable cap that is mounted thru the side of 
> the chassis.   Some red paint or other marking will remind you to be 
> extra careful around that part of the chassis until the transmitter is 
> safely back in the cabinet.
> 
> I think the Pacemaker's reputation for being difficult to adjust is 
> mostly undeserved, especially if you use an X-Y mode scope to set the 
> audio phasing and an SDR or spectrum analyzer to adjust the RF phasing 
> for maximum opposite sideband suppression for each band. Like some other 
> Johnson transmitters, the RF stages are tracked with the VFO which is 
> just like aligning a superhet receiver.   Both driver and PA are 
> neutralized for stable operation.
> 
> I'm looking forward to having the Pacemaker on the air for vintage SSB 
> and as the RF driver for my newly-acquired Desk KW.
> 
> 73, Bob W9RAN
> ______________________________________________________________
> Johnson mailing list
> Home: http://mailman.qth.net/mailman/listinfo/johnson
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:Johnson at mailman.qth.net
> 
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> 

-- 
Richard Knoppow
1oldlens1 at ix.netcom.com
WB6KBL