Need some help with sweep tube amp biasing

[loosecannon]

Hi all.

Im working on an old amp (lazer 3000)for which there is no schematic, so im having to use similar schematics to return this beast back to "stock" form. Not only is there no specific schematic for this amp, it has had a few different sets of hands in it during its long life.

I have everything pretty much figured out, but the biasing scheme used is causing some head scratching, so i was hoping that some of the more experienced folks on this forum could help lead me in the right direction.
any and all pics of any of these amps are appreciated if you have them.

Here are the schematics im working with, starting with the one that is closest to this amp.
(pretty much identical except for the receive preamp which im not worried about right now)

the first one to look at is the Crusader 300. (PDF attached at the bottom)
here is a screenshot of the specific area i am looking at:

starting at the top you can see the DC bias cap, the 47uf with the positive to ground. from there it goes through a 330 ohm resistor, another cap with positive to ground at 100uf, and on to pin 5 of the tube (6LF6's).

the part i want to know about is that 350 ohm 10 watt resistor in the middle of the pic, that goes to a 22uH coil with the other end of the 22uH coil connecting to the RF input to the driver tube.
notice that there is a line that seems to short out this coil. WTF is up with that is one question i have.

the next pic is from a PAL 321 that shows the same sort of biasing, but without the line that seems to short out the coil. here is a link to the whole schematic: http://www.cbtricks.com/Amp/pal/321/graphics/pal_321_sch.pdf

The above pic is very similar to the first, but without the "shorting line" of course, and at the top there is a 1K instead of the 330 ohm resistor, and a 120 ohm resistor is used instead of the 350 ohm resistor.
i am curious as to the effect of the resistance changes as well.


Ok, so those two are pretty similar, and im very thankful for any insight as to what is going on with them.


Now on to the schematic for the PAL 321A linked here: http://www.cbtricks.com/Amp/pal/321/graphics/pal_321a_sch.pdf

in this pic from the "A" version, the same 120 ohm 10 watt resistor is shown, but this time, it goes straight to ground. no 22uH coil, and not connected to the RF input of the driver tube.
Im showing the top of the schematic here, but if you follow the letter "D" to the bottom in the full schematic, you can see where the biasing starts.

this one also has a 10K resistor in series with the run to the driver tube. not sure if that's related or not, but again, im hopeful for any insight as to what's going on and why.


Sorry for the long drawn out thread and slow road to home, but it's late and i've been staring at these things too long LOL.

thank you to anyone who can help me with this, and just to put it out there, im open to using a bit of a different biasing scheme if someone has a better way.
LC

 

[DoctorAudio]

What's going on (at least on the Crusader and the earlier 321) is when the amp is in operate mode, the high voltage is always on (no TX switching) and the cathodes aren't switched off either. In essence, the tubes are always "on". What they are doing is applying a fairly high negative bias to all of the control grids when the amp is unkeyed to drive the idling bias down to a reasonable level. (supposedly, this is why a Crusader 300 is a tube eater)
When the amp is keyed, the second 350 ohm resistor is grounded through the first 22uH choke, through the transmit relay then through the second 22uH choke at the transceiver input coax jack. This effectively lowers the control grid bias to an "operating" level during transmit mode. The only reason the values are different on the 321 is either the floating voltage is higher/lower when unkeyed or the designers opinion on what voltages are correct differs from the designer of the Crusader.

On the 321A schematic, PAL realized the error of their ways and finally decided to switch off the HV supply when unkeyed instead of changing the grid bias to shut off the tubes. This is why the 120 ohm goes straight to ground as the negative bias is fixed at the operating voltage at all times.

The 10K on the control grid of the driver of the 321A is just there for "grid leak" which raises the static bias on the control grid to a higher level than just the fixed bias, varying with drive power. Kind of a "cushion" so to speak.

 

[loosecannon]

wow thank you for the great response!

so it sounds to me like the superior method is to eliminate that 22uH choke and just follow the PAL 321A setup where the 120 ohm resistor goes to ground and there is a 10K in series with the line to pin 5 of the driver tube.

I think that's what im going to do.
Thank goodness for people like you and Nomad who understand this stuff since no one teaches these concepts anymore.
LC

 

[DoctorAudio]

It is the superior way but remember to check the factory Lazer circuit to make sure either the high voltage is switched off or the cathodes are lifted from ground (through the cathode chokes) when the amp is not keyed. The tubes need to be switched "off" one way or another in order for the fixed bias circuit like the 321A has to work correctly.

 

[Dmans]

LC,
Check the "sticky notes" I added to the Crusader Amp schematic. Not disagreeing with Doctor Audio, but I don't see the 22 uH coil as being shorted during TX or RX. Looks like this Amp is chock full of indicator LED's???

73
David

 

[loosecannon]

DoctorAudio,

again thanks for the response. The lazer 3000 is identical to the crusader 300, or it was until i made the bias circuit like the PAL321A a few hours ago.

the input to the cathodes of the tubes is the same, and it looks to be the same as the PAL321A as well.

your response has me a bit concerned only because im not sure exactly what "lifts the cathodes from ground" when the amp is unkeyed.
This must mean lifted from RF ground? because i can't see anything doing any DC switching.

this is the part that always gets me, what the RF is doing that isn't apparent (to me at least) by looking at the schematic.

I guess what im saying is that right now i have, in essence, a crusader 300 amp in front of me that i have altered the bias circuit to match what is in the PAL321A schematic.

Can you tell me if what i have will work?

Dmans,

The line i was talking about starts in between the resistor and the coil and connects to the other side of the coil. just seemed strange to me.
yes, each switch has an associated LED just to make things more complicated.


And just to put it out there for everyone, WTF is up with using PC boards like this???
so much easier to work on these things when you can just flip the amp over and work on the circuitry.
with this stupid setup, i have to put it all together, test it all out, and then if something isn't right, i have to take the whole damn thing apart again.
i'll post up some pics just so there are some available online for future searchers.
LC

 

[Dmans]

The line i was talking about starts in between the resistor and the coil and connects to the other side of the coil.

LC,
It looks to me like the "non-resistor side" of the coil connects to the anode of the LED ( I have labeled as "RF Indicator") then to the RF input via the TX relay and the cathode of the driver tube. There is a .01uFd cap that is attached to the cathode of the same LED. There is a "dot" above that capacitor (normally signifying a connection) that I believe is a mistake in the schematic but more likely than not necessary to cause the LED to act as an RF input indicator.

I understand your frustration with the PC board construction. Makes me like the "Old School" (under chassis terminal strips) assembly much better.

73
David

 

[loosecannon]

Dmans, thank you for bringing up the possible mistake in the schematic.
for those of us trying to bring these old mutants back to life, mistakes on a schematic can be a death knell.
LC

 

[nomadradio]

The tricky part here is L2, a RF choke between the center pin of the radio socket and ground. When the relay closes, this puts a DC ground onto one end of the 22 uH choke. The other end of the choke causes the two 330-ohm resistors to divide the negative bias voltage seen by the tubes to be divided in half. Removing this ground connection in receive mode doubles the negative bias voltage on the tubes' control grids. May or may not be enough negative bias to shut the tubes down, but it is intended to put the tubes on standby.

Not the clearest schemo I ever saw.

73

 

[DoctorAudio]

Loosecannon,
I see your confusion. I was speaking partially in general terms. Some amps (Maco) use a relay contact to lift the cathode chokes from ground during receive to turn the tubes off . Other amps (Palomar) use a relay contact to switch off the high voltage during receive. Both of the methods work great as the tube is completely off while receiving.
Part of the confusion comes from me not looking at the 321A schematic closely enough yesterday.
I have worked on hundreds of PAL 321 and 321A over the years, the majority were 321A's. I can tell you that the "A" version DOES switch the high voltage on and off when keying which is why I mentioned they discovered the error in their ways. I see now that the schematic on CBTricks is incorrect and does not show any high voltage switching. If the amp were ran as shown in the schematic, the tubes would probably die in a day as they would be dissipating several watts, even while not being keyed.

In conclusion, the only way to convert your Lazer so it won't be a tube eater is to incorporate a way to switch off the high voltage during receive. You would either have to add another relay along with the stock transmit relay or change out the stock relay to a 3PDT and rewire it accordingly. If you did want to tackle it, I could provide you with info on where to tap in to switch the high voltage safely.

 

[Shockwave]

I like to apply bias to the same element as the drive. These sweep tube designs that introduce a control grid or small screen voltage through unregulated resistor dividers, tend to exhibit more thermal drift in power output as a result of unstable bias with fluctuating tube currents, that can even go negative in some cases.

The TL431 three terminal shunt regulator, buffered with a power transistor makes a perfect regulated bias replacement for many amplifiers. You can precisely set the idling current with this device and there is no change in voltage or current. Use a cutoff resistor to limit the voltage drop on the cathode when the bias relay opens. Just large enough to stop the tubes from drawing any current in standby. Then bring the control grid to DC ground like the other grids.

I'm not saying you need to do this mod on that amp, but if you notice bias drift the longer the amp is keyed (often noted as increasing carrier on watt meter) or a B+ relay keeps wearing out, this is one alternative that fixes those problems while prolonging stable, end of life tube use. Doing things like grounding all of the grids or using shunt regulation, allows you to get the full life use from tubes that are beginning to develop more secondary emissions near end of life.

 

[Dmans]

if you notice bias drift the longer the amp is keyed (often noted as increasing carrier on watt meter)

...........as does the Palomar 300A!

73
David

 

[loosecannon]

once again, thank you all for your participation in this thread.

Dmans, i see what you mean about that "shorting line" now and i agree with you. that dot is a mistake. thanks, that makes a lot more sense.

Nomad, thanks for the explanation of how they are attempting to shut the tubes down when the mic is unkeyed. I gather that you don't have a lot of confidence in their method?
what do you think of the way the biasing is done in the PAL321A schematic?

DoctorAudio,

I do think im going to keep the biasing like the PAL321A shows, and i think adding an extra relay to shut off the HV DC in receive mode sounds like the way to go.
I think i can figure out how to do it by tapping in to the keying circuit just like the TX relay, but i would very much appreciate seeing how you would go about it, as well as what the specs of this relay would need to be. I measured 850 volts DC out of the HV supply.

also, if i keep the PAL321A style biasing scheme, i think i will have to remove that 22uH choke that goes from the center pin of the input SO-239 to ground. can you confirm that for me?

Shockwave, your method sounds interesting, but i think i would need to be spoonfed the directions in order to make it work.

thank you all for taking the time to explain all this stuff to me, and who knows maybe there are some sandbaggers following along who will also benefit from your knowledge.
LC

 

[loosecannon]

Hey Nomad,

Seeing 399's new thread on the kris boomer made me look at the schematic i drew up for the kris power pump a while back with your help.

I think the diode idea you suggested to me recently regarding the second diode in the bias line is shown in that schematic, as well as a set of relay contacts that would turn off the cathodes when the mic is unkeyed.

can you take a quick look at that schematic and confirm that for me?
also can i just use this biasing scheme on the lazer 3000? it seemed to work very well on the kris.
LC

 

[nomadradio]

Yep, it's there. Only works this way with a single-stage amplifier. Gets trickier when the grid-leak voltage from the final tubes is also being fed to the grid of the driver tube.

Since the rectified negative DC from the final grids will be greater than the fixed negative bias on the driver grid, the size of the filter cap is less important. Using a small capacitance is less likely to result in 'hum' modulation.

A single-stage amplifier is just simpler all around. Yet another good reason to just lose the driver tube in these old amplifiers.

73