spikes on unkey...

[N2DTS]

I have been working on trying to remove some random spikes of RF when I unkey the 3X4D32 rig.
It was not neutralized, so I tried to neutralize it using the usual lifting of the grid coil above ground and inserting feedback there, but it does not seem to neutralize well, even with a very little cap there.

So I wonder if there is any good way to damp out any RF on unkey.

The spike on unkey does not seem to cause any problems, but its not normal...

Brett

[w3jn]

If your ant relay runs on DC, put a 10 uF cap or so across the coil.  This will delay the relay dropout a bit and let the final drop out while it still has a load on it.

If everything is switched simultaneously (IE B+, final cutoff, ant relay) the final could still have power applied for a few milliseconds while the ant relay has already done its job - thus, no load on the final and noplace for the RF to go except to be dissipated in the tank circuit and final tube.

[k4kyv]

You need a relay sequencer.  The 10 uf cap will also delay the antenna change on start-up, so the final will be unloaded for a fraction of a second at the beginning of each xmsn instead of at the end.

I had this  same problem when I first put the 160m Gates on line.  It resulted in turning a Dow-key into a solid block of charcoal.

I built a little sequencer circuit using 3 control relays in a latching arrangement so that the transmitter is delayed on start-up, but the transmitter drops out immediately while the antenna relay is delayed and last to drop-out. The latching relay toggles the delay capacitor between the antenna pilot relay and the transmitter pilot relay. I have a copy of the circuit on file, plus I did a write-up in Electric Radio about 6 years ago.

A cheap, simple and dirty way would be to use brute-force mechanics, employing an open-frame multi-pole "pilot" relay to activate the main antenna and T/R relays.  Bend the reeds that hold the contacts so that the set that runs the antenna relay makes contact first and drops out last. When the relay is unenergised, the set that runs the antenna relay should be adjusted to have as small a gap as possible, while the the set that activates the transmitter has as wide a gap as possible. It might take a little experimenting to get it working reliably.

[N2DTS]

Thanks Don.
The relays are sequenced, in the form of one relay energizes another...
The ant relay is a BIG contactor run off 120 vac....

It does it sometimes, about 10% of the time, adjusting the grid bias, grid leak, screen dropping resistor, seem to have no effect, maybe the final tuning does a little.

I think its in the RF deck maybe, since I cant seem to neutralize it at all.
Three 4D32 tubes in pie net might not be the best choice, I think it worked fine with two, but that is only a a piss weak 200 watts!
 

Brett


 

[w3jn]

THe cap won't delay the ant relay engagement if the relay's power supply is stiff enough.  I've used this arrangement for years with no problem.

[WA1GFZ]

You want to add a diode ahead of the relay and cap to keep the cap voltage across the relay coil and not back feed other circuits. The cap holds the antenna relay in long enough to discharge all the energy in the final and modulator.

sequence easy to do with relays.
bias modulators to cut off
bias final to cutoff
kill drive
open antenna relay

Goint to TX
antenna relay to TX
turn on drive
operating bias to final
operating bias to modulator

[WU2D]

Brett,

Rice (grid) neutralization with a true balanced center tapped coil will work better than the compromise "split cap lifted driver tank" type of neutralization used on most 6146 rigs.

And Hazeltine (Plate) neutralization with a center tapped tank will work better still.

Mike

[N2DTS]

Mike,
yes, I should look into the balanced center tapped grid coil setup.
As it is, its not neutralized at all, I added the rings around the base of the 4D32's and that did not help matters at all.

I have some more experimenting to do, I might go back to no neutralization since what I did is not working.
I put the grid stuff in a metal box, no improvement, did the neutralization thing, no difference, installed the base rings, no difference...
The spike on unkey might have nothing to do with neutralization....

Brett

[Opcom]

Is the spike on the operating frequency or elsewhere? There could be some other issue at work. I been fighing that battle.

[N2DTS]

It seems to be on freq, or broad banded.
When I unkey, the mod monitor mod meter spikes up somewhat, the wattmeter spikes maybe 20 watts, the panadapter in the flex shows the entire bandwidth well above the noise level.
It does not happen all the time, about 10%, adjusting the final tuning on the rf deck one way past the dip makes it happen all the time it seems, the other way 10% of the time.

I did a brief test running full bore with no drive, no RF out, but I did not try adjusting the tuning and loading much.
It has run a long time this way, but I don't think it always did it, I don't remember having the issue in the past.

I might increase the protection bias which is a little low I think. Grid leak bias is high....

Brett

 

[W3GMS]

Brett.
I never heard an answer on the sequencing times.  What happens if you put a dummy load on the output of the amp and leave it connected all the time (no TR relay).  If you still have the problem after doing that, I would agree thats it something else.
Joe, W3GMS 

[WQ9E]

Good advice Joe!

Like JN, I use the cap across a DC relay coil to provide a little delay on release.  Johnson includes notes on that in the Desk KW manual and with mine I use a Jennings 28 volt vacuum relay and use the filter cap on the 28 volt supply to provide the slight release delay.  The cap should not create a delay for closing the relay unless you have some serious resistance between the rectifier and the capacitor and if you do there is a problem; the normal power supply for a relay should not have enough series resistance (from the transformer windings and rectifier) to provide for a significant RC type circuit.  I wouldn't use a vacuum tube rectifier for this relay supply of course

I ran into a similar problem with my Viking 500 which has a built in delay upon release circuit for antenna changeover provided by a "mechanical delay" relay.  I guess after 50 years the mechanical delay is no more and the transmitter started firing across the protective spark gap once in awhile upon PTT release.  I installed a small capacitor with an isolating diode across the "mechanical delay" relay a couple of years ago and that provides a reliable short delay upon PTT release and no more fireworks. 

By the way, my Viking 500 has the much maligned original connectors between the modulator/PS deck and the RF unit and they are working fine.  I imagine that many of those who experienced arc over with those had it because of the lack of antenna relay delay upon PTT release and they had also followed the instructions in the Johnson manual to slightly increase the spacing on the protective gaps if nuisance arc over occurred.  Once those gaps are wide enough then the chassis connectors themselves become the protective gap.  Replacing these with "better" connectors and having too wide a setting on the protective gaps is a perfect setup for some expensive damage.

Rodger WQ9E