Johnson DKW help

[W9BHI]

Hello all,
I need to make a fast make slow break antenna relay for my Johnson DKW.
I would like to use a Dow-Key 115vac relay.
I saw the circuits in the manual but They don't specify the coil resistance for the AC circuit.
Any help would be appreciated cause I'm about ready to put this puppy on the air!
Thanks,
Don W9BHI

[AB2EZ]

Don

Hi!

You posted a number of previous requests for help/suggestions. You received a number of responses.

Now that you have the Desk KW up and running... can you comment on what you ended up doing with respect to the audio driver and the bias/screen supplies?

Best regards
Stu

[WD8KDG]

Hi Don,

Guessing here that you read the KW manual, page 23. It sort of shows a skezmatic of how to use an AC relay. The bigger question is; was the Dow-Key 115AC relay meant to work on DC? Reading more on that page, future issues might arise such as the coil becoming magnitized. I haven't been down that road, maybe someone else has it working.

The chosen path here in Springtucky was to use the Johnson KW Matchbox. The delay relay is built-in, will handle full scroat, open wire, etc.

Craig,

[WQ9E]

Don,

Unless this is a newer DowKey relay I wouldn't trust using an old one with the Desk KW.  They weren't great relays to begin with and they age very poorly (burned contacts, sticking contacts, etc.).

I built a relay box for my Desk KW with a Jennings vacuum relay for RF switching.  The relay is powered from a simple 24 volt supply built into the box consisting of a transformer, bridge rectifier, and a filter capacitor.  The transformer primary is powered by the switched 110V AC from the Desk KW and the filter cap for the supply provides the delay on break.  The vacuum relay switches very quickly so that the Desk KW isn't looking into no load while the dowkey or similar slower relay is settling.  A second  regular relay handles muting.

If you use the Johnson circuit you are going to have to experimentally adjust the value of current limiting resistor R3 so that the relay picks up quickly enough but doesn't overheat.  This value will change as you adjust R1 for proper release delay and slightly if the coil is heating some which will cause its resistance to increase.  If R3 is too low in value you will overheat the relay coil and if it is too large you will have slow make which will burn the contacts while subjecting the Desk KW output network and modulation circuit to high peak voltages from an unloaded output circuit.  Any loud audio during a slow make may well create a high enough voltage transient inside the Dowkey to arc to the receiver antenna contact.  If someone comes up with values from old notes check to see if they were using a selenium rectifier because the voltage drop is considerably less with a modern silicon rectifier.

The Dowkey relay is convenient but I wouldn't chance one of unknown age and history in this application.

[W9BHI]

Stu,
I left the bias and screen supplies alone, they worked as Johnson intended.
No sense in reinventing the wheel.
I went with a 30 watt backwards P-P output transformer that I got from Robert (W0VMC).
He uses them in his KW1 modifications and they work well to modulate a pair of 810's.
I am using a TOA BG-130 30 watt mono paging amp for a speech amp, which works well with that transformer.
the DKW is working in CW and I am close to getting it on the air for AM.
I need to get my keying circuits built for the FT-950 and R-390 and DKW and I should be on the air.

ALL you guys that made suggestions and offered help to me are greatly appreciated.
I will keep posting on my results.
Thanks,
Don W9BHI

[W9BHI]

I may use the circuit that is used in the Johnson Kilowatt tuner, it has a delay.
I can get the 5000 ohm relay and use it to switch the Dow-Key.
BTW the Dow-Key I have is new so it shouldn't have any issues (I hope).
 
Don W9BHI

[AB2EZ]

Don

I suggest that you build a sequencer box. There are several ways to make a sequencer, and it is an inexpensive and very worthwhile addition to your shack.

When you want to transmit, you close your push-to-talk switch... to activate the sequencer (not the Desk KW directly). One output (call it the 1st output) of the sequencer will provide a path the ground, almost immediately after you close the push-to-talk switch. You can use the 1st output to activate the antenna relay. Another output (call it the 2nd output) of the sequencer will provide a path to ground about 0.5 seconds after the first output provides a path to ground. You can use the 2nd output to turn on the Desk KW.

When you want to stop transmitting, you open the push-to-talk switch. The 2nd output's path to ground opens up almost immediately... turning off the Desk KW. The 1st output's path to ground open's up about 0.5 seconds later, deactivating the antenna relay.

I made the sequencer in my shack using old fashioned relay technology. I needed two 12V DPDT relays, and a capacitor with (roughly) 3300uF of capacitance to provide the needed delay. My sequencer provides 12V to the antenna relay from output 1; but it is easy to modify this design, to provide a path to ground at output 1.

The wiring diagrams are attached. The first attachment is the complete wiring diagram. The 2nd and 3rd attachments are partial diagrams to show (separately) the circuitry for switching the antenna relay (the 1st output) and the circuitry for switching the transmitter (the 2nd output).

In this design, you can use relays that have coils with 360 ohms of series resistance, and requiring 23.3mA of coil current to operate.

http://www.mouser.com/ds/2/418/RT2pb0411-205066.pdf

With 100 ohms in series ... for the delay ... 12 volts across 460 ohms produces 26mA of current. If 26mA of current is not quite enough... to activate the delayed-turn-on relay... then use less resistance (e.g. 47 ohms) and more capacitance (e.g. 2 x 3300uF)

The delay time will be roughly two RC time constants = 2 x 78 ohms x 3300uF = 0.5 seconds  

Separately:

I don't know whether or not Dow-Key relays are reliable enough. I use vintage Dow-Key relays; and also new Tohsu relays, which are much less expensive than new Dow-Key relays. The Dow-Key relays do require occasional contact cleaning... and sometimes the SO-239 sockets on my Dow-Key relays don't make good contact with the mating PL-259 connectors. To fix that, I usually roll up a 3/8 inch wide strip of aluminum foil into a 1/8 inch diameter cylinder... and insert it into the Dow-Key SO-239 center sleeve... before attaching the PL-259 connector.

Stu

[WQ9E]

I may use the circuit that is used in the Johnson Kilowatt tuner, it has a delay.
I can get the 5000 ohm relay and use it to switch the Dow-Key.
BTW the Dow-Key I have is new so it shouldn't have any issues (I hope).
 
Don W9BHI

Don,

Don't use another relay to switch the DowKey because you will have too much delay on make leading to Desk KW output with no load along with hot switching.

As Stu noted in his response a sequencer is useful and is easy to build.  I have sequencers on other rigs but with the relay setup I use on the Desk KW I have had no problems with it.  But with the sequencer I would still use a vacuum relay for the output with less delay for the sequencer unless you train yourself not to say anything for .5 second otherwise you will clip the first word of every transmission which will become annoying to others in a QSO/roundtable.

 I believe when the Desk KW was young a lot of users used a Johnson electronic T/R switch with it which carries its own set of benefits and drawbacks.

[W9BHI]

Got the JDKW on the air last night.
Great audio reports.
Does anybody know what the input impedance to the rf deck is?
It sure ain't 50 ohms.
The tuner in the FT-950 wouldn't tune it (was more than 3:1).
I put a 50 ohm load in parallell with the input and then the Yaesu would tune it.
Is the input impedance too high or too low?
Thanks,
Don W9BHI

[WD8KDG]

Don,

Seems to be an issue with using a ricebox as the driver/exciter for a boatanchor amp. Other lashups used by hams in the PNW have been there using TBolts, etc. One fix was to add another tuna between driver & amp

Anyway......in Table A there are suggested settings for the Grid Capacitor & Grid Inductor. You will have to experiment keeping fingers crossed.

Craig,

[WQ9E]

Around 100-200 ohms non-inductive swamping should make the FT-950 happy.  If the FT-950 has enough output to drive it with the 50 ohm load in parallel then you can use it that way.

As Craig stated the input impedance is going to vary a lot with grid tuning.

[W9BHI]

Craig,
I am using the suggested grid tuning settings in table A.
The SWR reading on the 950 is lowest when the JKW grid is peaked for max.
The 950 is running at about 27 watts into the 50 ohm load to give 20 Ma. of grid drive.

Don W9BHI

[W9BHI]

I put an MFJ analyzer on the input of the JDK and got the input to tune down to 1.8:1 SWR @33 ohms on 7.295.
The tuning is VERY sharp and was very easy to go past the sweet spot.
This is why I couldn't get it to tune with the FT-950.
After using the MFJ, I put the FT-950 back on without the 50 ohm load and the internal tuner tuned it instantly.
To make a long story short, when tuning the input using a 50 ohm rice box, I will need to tune the input VERY slowly looking at the SWR on the radio.
BTW, now it only takes about 9 watts to get the 20 Ma. of grid drive.
I got the audio processing chain connected to the outboard speech amp and it now passes
the W0VMC test  

Don W9BHI