relays

[KC2ZFA]

hello all,

I'm moving forward to the ps for my p-p 812A stage and the 811A modulator
(both to be run at 1250 V).

I went through a myriad of posts on step-start/inrush, PTT, and sequencing
issues (amfone is a gold mine! you guys rock) as well as the ancient holy
books. I will put the resistor inrush limiter with a 1 second delay on both
xfmrs. But the iron is old (stancor p-8025 and kenyon t-669) and I'm hesitant
to do PTT this way.

Can I use the pictured Cutler relay to switch the pa and mod B+ _simultaneously_ ?
The mod iron is a 300w thordarson multimatch and, to begin with, I won't have the
heising gizmo as I haven't been able to find a suitable choke yet.

If I can do this then to transmit I will be changing over the antenna, exciter B+
on, pa/mod B+ on. The biases on the exciter+PA will be always on.

[WA1GFZ]

No, look at the contact voltage rating on the tag. These relays are for switching primary power. I have a pair I ripped out of an old Deicer project breadboard my boss gave me. They were used to switch 3 phase 400hz power. This is the perfect realy to use on a 240 volt line 50 amps should be enough rating. (28 VDC coil)

[KC2ZFA]

OK, great, one mishap averted 

how about the principle of simultaneously switching pa and mod B+, say
with a DPST vacuum relay ?

Also, the relays pictured below should _not_ be used for this kind of switching
with 1200-1700V...right ?

[Opcom]

I used a jennings vacuum relay to select the 1200 or 2400V tap on a 0.25A plate transformer, it never failed in 5 years. It was probably overkill. With DC, watch not to create an arc caused by any indicturs in series.

[W3SLK]

That P&B on the right looks like the typical 10A jobber. Pretty much stock. It should work FBOM in the primary circuit.

[KA2DZT]

OK, great, one mishap averted  

how about the principle of simultaneously switching pa and mod B+, say
with a DPST vacuum relay ?

Also, the relays pictured below should _not_ be used for this kind of switching
with 1200-1700V...right ?

The larger blue relay (of the two pictured) will work FB to switch the primary of each xfmr.  It has 4 poles each having a 25 amp contact rating which is more than enough. This type relay should not be used to switch HV DC.

You need to find what the coil voltage rating is.  It wasn't shown on the label, but should be marked on the coil.

I'm not sure how you are doing the PTT.  Are you going to relay on the HV B+ or are you going to relay on the primaries of the two plate xfmrs, or what??  You may want to post a bit more detail of what you're planning.

Usually you will need two relays to have a step-start for the primary circuit.

Fred

[WD8BIL]

I'd forget about using relays that way and put 50K resistors in the cathode lines. Use smaller less expensive relays to short out the resistors during transmit.

see the red box in the schematic.

[KC2ZFA]

I'd forget about using relays that way and put 50K resistors in the cathode lines. Use smaller less expensive relays to short out the resistors during transmit.

see the red box in the schematic.

that's a another good idea...and since I'm using 812A/811A tubes I would have to adjust those 50k's for
appropriate resting current, to suffer a fraction of max tube dissipation, when excitation is off (and leave
the HV on at all times)...yes ?

To Fred: my main question is whether or not it is OK to switch the PA/MOD B+ simultaneously with, say,
a DPST vacuum relay.

As you guys can tell I'm a newbie and thus rightly doomed to reinventing the wheel 

[WBear2GCR]

The larger blue relay (of the two pictured) will work FB to switch the primary of each xfmr.  It has 4 poles each having a 25 amp contact rating which is more than enough. This type relay should not be used to switch HV DC.

You need to find what the coil voltage rating is.  It wasn't shown on the label, but should be marked on the coil.

Fred

That's a P&B relay - the coil voltage and contact rating is in white on the end of the blue plastic base moulding... I love those relays. Will take/buy any and all that someone doesn't want... I use them for slow start and power on for my big audio amps.

I think it is a 36vdc coil, says the voltage and resistance on the side, but I had difficulty reading it on my larftop compooter... run it with more than 36 volts...

 

                        _-_-bear

[WA1GFZ]

Use Zener diodes or diode junctions in series to set your resting current rather than adjusting the 50 K resistor. When you select the 50 K resistor you want to force the tubes into cut off without sending the relay contact voltage above break down or tube cathode voltage too high.  The doides would go between the relay and heater winding center tap. Good to use caps across the diodes to eliminate any diode noise.

[WD8BIL]

Yes, as Frank said, use zeners or a bias supply to set idoling current. The 50Ks are there only to drive the tubes to cutoff during receive.

[KC2ZFA]

Yes, as Frank said, use zeners or a bias supply to set idoling current. The 50Ks are there only to drive the tubes to cutoff during receive.

OK. But if I'm getting the operating bias from a combination of grid leak and fixed
bias (fixed bias which is enough to cut-off the tubes when drive is out) then I don't
really need the cathode bias in order to just leave the PA HV on all the time and
only switch the modulator HV on/off during PTT. Yes ?

[WA1GFZ]

Say you have 1ma flowing through the 50K resistor when the relay contacts open. E=I*R so there is 50 volts across the resistor placing positive 50 volts on the cathode of the tube or tubes. Say you have grid leak bias. Now the cathode is 50 volts and the grid is zero. Same as negative 50 volts bias since the grid has less voltage on it than the cathode. This way you can use low voltage relays as Bud suggested. Since the tubes are in cut off you can leave the high voltage on as long as you respect the possible safety hazzard. You want the rig built into a cabinet or rack so the cat doesn't get fried when it comes in the shack.

[KA2DZT]

To Fred: my main question is whether or not it is OK to switch the PA/MOD B+ simultaneously with, say,
a DPST vacuum relay.

As you guys can tell I'm a newbie and thus rightly doomed to reinventing the wheel 
[/quote]

IMO you can use a vacuum relay (DPST) to switch the B+ of the two supplies, provided the relay is designed to handle the HV (I think you said 1250VDC).  The problem with switching HVDC is the arcing (at the relay contacts) that may occur.  Especially, when dropping the B+ from the mod xfmr.

There are ways to overcome these arcing problems.  One way is by using a number of HV disc caps from the relay contacts to ground.  I use 2 sets of two .0022 6KV disc caps to ground (one set at the relay contacts and the other set on B+ terminal of the mod xfmr).  I also use a HV cap from a microwave oven (.8ufd 2200vac, good for about 5-6KVDC) from the B+ terminal on the mod xfmr to ground.  This should eliminate most any arcing.  When you open the relay contacts these caps should absorb the counter EMF from the collapsing magnetic field in the mod xfmr.  The discs will handle the faster rise time arcs and the .8ufd cap should handle everthing else.  I also use a .01ufd 2KV disc cap on the PS side of the relay.

Having said all of this, my plate supply is about 700 VDC.  Using a 1250VDC plate supply you may need caps with a higher voltage rating.

Since you're using two plate supplies you may need two sets of everthing I outlined above.

Another thing you need to watch,  unloading the plate supplies will cause a rise in the plate voltage on the PS filter caps.  These caps need to have a high enough voltage rating to handle this higher voltage.  This is true regardless of what method you use for putting the xmtr in standby.

One way to eliminate all of these problems, I mentioned above, is to shut off the HV supplies on standby.  You do this by using a step-start relay circuit on the primaries of the HV xfmrs.  The step-start relays are controlled by your PTT circuit.

As you can see, you need to do a little designing on your PTT and control circuits.

Hope some of this helps,  I'm sure you'll get more hints and advice on my methods and other schemes.

Keep in mind, building a xmtr from scratch is not always as simple as it looks.  But, you'll be an expert by the time you're done.

Fred

[KC2ZFA]

Another thing you need to watch,  unloading the plate supplies will cause a rise in the plate voltage on the PS filter caps.  These caps need to have a high enough voltage rating to handle this higher voltage.  This is true regardless of what method you use for putting the xmtr in standby.

both supplies are SS full-wave rectification, choke input, with caps at 2.25 kV
and 56k bleeder. That should keep things tame if I'm to switch the high voltages.
One of the rectifier boards is shown below.

Thanks for the info guys. I'll draw the cathode resistor stuff as I understand it
from your comments and submit a schematic to see if I got what you told me.

[WD8BIL]

Yes, the bias for a Class C final is set to cutoff and just leaving it that way can work. But what you really need to do here is drive the stage deep into cutoff with no drive present. This will insure no surprises when you un-key such as unwanted oscillations. For the cost of a few resistors and a small relay it's a lot of protection!

[WBear2GCR]

I am not seeing any really good reason to key the B+ supplies... why would you want to do this??

You may want to delay the turn on of the B+ supply WRT the filaments, but that is different.

If you key the B+ supplies you will have to deal with the no load B+ voltage vs the loaded B+ voltage. Even if you swamp the B+ when it is not connected to the load with a power resistor, there will be a nasty spike during the time it takes the relay to switch.

One of the many "tricks" - suggested to and tried successfully by John KC2FXE on his version of the "JJ" 813 rig -  is to use a cap in parallel with the 50kohm (or whatever value you use) resistor that permits it to charge up at the right time constant, to prevent a "hard switch" that causes an arc on the secondary of the mod iron. Worked like a charm.

               _-_-bear

[KC2ZFA]

I hear you Bear. So the best solution is

a) all filaments, biases, and cathode resistors on.

b) soft-start of HV with some sort of inrush limiting (meaning HV will appear
    on the PA and mod).

and then it's fun-time with the PTT keying the receiver mute, antenna, exciter,
and cathode resistors in the right order.

do I feel lucky ?  now do I ?

[WBear2GCR]

The need for soft start on the initial turn on depends a whole lot on how much juice the thing draws, but sure, a series resistor with the primary that is then shorted out with a relay is OK FB OM... If you get fancy a case can be made for fault detecting circuitry that will dump the primary in the event of an over current or voltage... often folks use a bit of thin wire between two ceramic standoffs for fusing in HV places, afaik. Others may have more to add on this...

                      _-_-bear

[W1RKW]

In my 813 rig I use a variac to bring up the 'juice' rather than step start it then short out the variac with a switch.  Granted a step start is nice and automatic but as a knob twister and switch flipper it doesn't bother me to have the TX 100% manual operation. And having the variac I still have the option of running the rig at a lower VAC.

[WBear2GCR]

There ya go!

Plus you get a BIG knob to turn... !

               _-_-

[Opcom]

Yes, as Frank said, use zeners or a bias supply to set idoling current.

Idoling Current: The amount of plate current necessary before candles and incense can be lit in front of the transmitter.

[KC2ZFA]

Idoling Current: The amount of plate current necessary before candles and incense can be lit in front of the transmitter.

no need for idols. I will hang this

from the plate tune knob 

[WD8BIL]

Now that's funny! I didn't catch that, obviously!

[KC2ZFA]

would the pictured relay be able to switch 1200-1700 volts (provided
appropriate caps to ground are used for the spikes) ?

Peter