Need some help with sweep tube amp biasing

[loosecannon]

you know i have been considering that, and if it was my amp i would do it.

alas, it's not, and i know the person who owns this amp wants to see those final tubes driven hard.

looks like i'll be using the PAL321A biasing scheme.

I do think that using a relay to ground the cathodes on receive will work better than trying to switch the HV.
someone correct me if im wrong on that.
LC

 

[DoctorAudio]

If the Lazer power supply is the same as the Crusader, you would open the line here where the transformer winding feeds the 100uF electrolytics and connect the NO contact of the relay between the winding and the caps.

A 3-5 amp relay would be fine and you could even use 2 poles it it's DPDT to share the load. You can make the relay last longer by putting a .01uF 1KV - 3KV disc cap across the relay to debounce the contacts.
Even though this is a high voltage supply, there is only about 300 VAC here and is easier to switch than 900 Volts DC.

 

[nomadradio]

Has one small drawback. The tubes will continue to have HV after you unkey for as long as it takes the filter caps to discharge. Probably won't be a problem, seeing how many other amplifiers did it that way. That power supply really should have a separate bleeder resistor across each separate parallel pair of filter caps. The single bleeder they show from the HV to ground is not effective for two caps in series. Runs the risk that one filter will hold a charge after it's unplugged.

73

 

[loosecannon]

thanks docaudio.

good suggestion. yes it's about 275vac there, and that is exactly how i have the power supply set up. using only two 100uF caps since that's what was there.

as for the cap on the relay, you mean across the coil right? in parallel with the diode of course.

as you can see from the post i just made, i was thinking about switching the cathodes to ground in receive, but that does seem more complicated to achieve in this amp.

do you (or anyone else following along) see any problem with simply tapping in to the keying circuit right where the transmit relay gets its ground from?
asking because now there would be two relay coils getting their grounds through that keying transistor.
LC

 

[nomadradio]

So long as the second relay coil won't draw enough current to overload the keying transistor, hooking the second relay coil in parallel should work just fine. And a bigger keying transistor would fix that if it became a problem.

73

 

[loosecannon]

yep nomad, im already there. going to use a TIP regulator there.

i'll admit that all the good suggestions have me moving backwards a bit, even though i know i'll be creating a better amp in the end.

as it stands right now, im going to add a relay to switch the HV unless someone has a better suggestion.

I also have the bias set up exactly like the PAL321A schematic on cbtricks shows it, including the 10K resistor in series with pin 5 of the driver tube.
again, unless someone suggests otherwise.

thanks to both of you for all the effort.
LC

 

[nomadradio]

Once the standby-current problem is fixed, I would take the "if it ain't broke" attitude towards the grid bias. And if the performance comes up short, that would be a reason to tinker with it.

73

 

[DoctorAudio]

Loosecannon:
No across the switching contacts. The cap will suppress the arcing of the contacts a little, making them last longer. D&A is one of the well known sweep tube amp manufacturers that that uses debouncing caps on the high voltage relay contacts.

Of course debouncing caps aren't required and as you will be switching AC not DC, the arcing will be less anyway.

 

[DoctorAudio]

I just wanted to say I don't mean to step on toes, just trying to help. If I'm out of line here, lemme know.

 

[loosecannon]

oh heck no!

I just got busy with housework last night and never got any radio work done.
Honestly your help here has been invaluable, as well as nomad's and i think the combination of your ideas has led me to a good place.

as of right now, i am going to put the biasing back the way they had it in the Crusader 300 schematic, which is how this amp was set up before.

Im going to add in the relay to switch the HV off in receive, and i will be putting a cap across the contacts as you suggested.

Im going to add another bleeder resistor across the other 100uf cap, and im going to upgrade the keying transistor to a TIP42 because of the extra current being pulled through it.

for now, that's going to be it, and we'll see how it works.

the last thing i need is a meter for this amp, and man if they are not tough to find in the right size.

seriously, thank you for all the help, and you can bet i'll be hoping to pick your brain again when i build my own amp in the next few months. it'll be a 2 driving 5 6LF6 amp and i am currently gathering parts for it.
LC

 

[loosecannon]

ok im in the process of getting the amp ready for the new relay which should be here tomorrow, and have a question about the HV supply.

currently it is set up like this:

i am only using two 100uF caps, as that is what was there before i got the amp.

my question is about the two 470K resistors in parallel going to ground.

i was thinking about removing these and strapping one 470K across each 100uF cap.

thoughts on this?
LC

 

[nomadradio]

That's what we always do. Each separate resistor needs to be wired directly in parallel with one capacitor.

73

 

[loosecannon]

Success!!!

well, mostly anyway.

Still a couple of minor bugs to work out, but the bias is put back to the way the Crusader 300 schematic shows.
(Dmans, i can't thank you enough for pointing out that tiny dot mistake on the schematic, that really helped me)

the HV relay is in and working as it should.

the resistors are strapped across the 100uF caps as suggested. (i went with 240K since i had some. someone let me know if they dont like those values)

With about a 1.5 watt deadkey it will key almost 100 watts on AM. the radio im using only swings up to about 5 watts, so swing is unimpressive at somewhere over 200 watts.

on SSB with about 10 watts in she will do 300 PEP or so.

I've only done a few quick keyups because of the issues i'll mention below.

first, input SWR is about 2.5, so im going to try adding a 463 trimmer cap in parallel with the input coil. (the .75uH coil shown in the schematic). my plan is to put it before the coil.

open to any suggestions on that.

next issue is the TUNE control sits perfectly fully UNmeshed when at max, which im guessing means its really not at max.

again, open to all input on changing this, spreading or compressing the coil or changing the series cap value etc...


the LOAD control is one that will only rotate half a turn, and it sits just about fully unmeshed when at "max" (again might not be true max).

there is something else i notice about the LOAD control and that is that it's peak is REALLYYYY tight.
like by the time the needle of the wattmeter has moved up, you are past the peak already.
very touchy.


DRIVE control sits pretty much perfectly half way between meshed and unmeshed so that seems ok.

SWR into the antenna seems to be a bit over 2.2. (i figure this has something to do with the fully unmeshed output tank circuit)

with amp off this antenna has an SWR of about 1.3.


I haven't tried it into a dummy load yet, but i can if someone thinks that info will be helpful.

thanks for all the help so far and any future assistance.
LC

 

[nomadradio]

The Plate Tune control shows its peak at minimum capacitance. The thing to remember here is "too much" when a control goes all the way to minimum. Can't make the capacitor any smaller, so the coil attached to it is the next target. Simply spreading the coil's turns farther apart will reduce the coil's inductance. Maybe enough to restore the peak position of the Plate Tune back between its extremes. And if that's not enough, clamping two adjacent turns together and lap-soldering them together will reduce the functional length of the coil by one whole turn. Squeezing the remaining turns as close as they'll come without touching is usually next after 'losing' a whole turn of the coil. Reducing the coil one full turn will usually become an "overshoot" correction. You'll know you reduced the coil too much if the Plate Tune now shows a peak fully meshed, at its max capacitance position. Squeezing the remaining turns back close together will compensate for the overshoot to some degree. Maybe enough maybe not. Shorting across one half of a turn with a piece of bare wire is not unheard of.

If you have ever heard the term "cut and try", that's what is usually needed for this issue.

73

 

[loosecannon]

still working on the tuning.

spreading the coil that connects to the TUNE control didn't really change anything so i shorted a turn, which still didn't change anything that much.

this would be the lower coil in the diagram below:

it has five turns. (four after i shorted one)

the upper coil that connects to the coax going out to the antenna shows 4 turns in the schematic, and i count 7 on mine.

there are also two other differences i noted. my amp does not have the two 120pF caps in series as shown. they're just not there.
the other difference is that DC blocking cap is shown as 2000pF and mine has 4000pF there. not really sure what difference this makes.


The last issue i am having is with the input tuning. i have tried trimmers that start at 10pf and can go up to 400pf, and never found the dip, although anything over about 70 just caused the relay to chatter.

to me this says i might have to change that .75uH coil going to the cathode of the driver tube, but not sure which direction to go.

doing so will be a bummer since it will involve taking the board up again which is a PITA.

I did notice that on the LOW setting, input SWR was only about 1.5 and that makes me wonder if it has something to do with that 47 ohm resistor used for high and low power since the cap is in parallel with it when in high power mode. i did use a non-inductive resistor for this if that matters.

I know i just need to fiddle with it more to get it right, but the insights from you guys who have done this many times is super helpful so please do let me know if you see anything i am missing.

again, thanks for helping me get this far.
LC