screen voltage vs output...4-400

[KL7OF]

Playing with Mcmartin BA1K transmitter......The rig has been converted to 4-400s from 4-500s..the conversion involved changing the filament transformer only....the rig has separate supplies for the modulator screens and the PA screens. The screen supplies put out 800volts...The transmitter was run with 800volts on the screens of the 4-400s by the previous owner....When I first got it up and running, (with 800 volts on the screens) It put out rated power on the 500 watt tap (2500 plate volts)...I wanted to bring the screen voltage down to the 500 volt range as required by class "C" 4-400s....As soon as I drop the screen voltage, the out put power goes down as well....seems reasonable but, these tubes are supposed to only have 500 volts on the screens and produce 500 watts with 2500 plate volts..They won't give anywhere near rated power with 500volts on the screens....I have varied the bias voltage during these experiments, but no joy....I have tried different tubes and they all act about the same...Unless I piss beat the screens with 800 volts, I can't get close to rated output...I have talked with another ham that has a BA1K and he is running 800 volts on the screens. his tx works well..I don't want to damage the tubes so I'd like to run the recommended screen voltage..Any of you running class "C" 4-400s ???If so what voltage is on the screens?
   Any ideas or suggestions would be appreciated....Thanks...Steve KL7OF

[WQ9E]

Steve,

The Johnson Desk KW and 500 both use a 4-400 final.  The Desk KW runs the plates at 2500 and the screens at 400 with 90 volts of fixed bias at the 1 KW input level for the pair of 4-400 finals.  The 500 runs a lower plate voltage (2000) with higher screen voltage (440) with its single 4-400 final at 500 watts input.

Hope these data points help.  I load my Desk KW to 450 watts in the low power position and it develops about 330 watts carrier output on 75 run this way, it will run close to 750 watt carrier output in the high power position.  My 500 runs right about 350 watts carrier output on 75 at 500 watts input.

[N2DTS]

I would say you can not have 800 volts on the actual screens, there has to be something in line, otherwise the screen current would be way high and the screens would melt.

I would think you could get good power out even with a low screen voltage, like 300 volts...

[n2bc]

I had an RCA BTA1R1 for a while, 4-400s modulating 4-400s.

It has a separate mod screen supply, it runs the 4-400 modulators as follows:
1KW:  Screen = 750V; bias = -145V; Anode = 3100V
500W: Screen = 750V; bias = -150V; Anode = 2200V

The PA 4-400s (separate anode supply, primary tapped for 1KW & 500W)
1KW: Screen = 500V vis anode supply dropping R (50K); bias = -400V; Anode = 3000V
500W: Screen = 340V via anode supply dropping R (50K) /w relay switched adj tap; Anode = 2150V

My HV Power transformer did not have the 500V tap. For low power I simply ran the HV power transformer on 120VAC, halving the voltage PA anode and PA screen voltage, then adjusting for 1/2 the current in the PI net so as to maintain the proper LC ration.  I adjusted the modulator low power bias to keep the mod anode and screen current 'happy' there was tons of audio to play with.

Another mod I made was the addition of a dedicated screen current meter for the RF PA and Modulators.

If you would like I can send you a PDF of the 1R1 manual & you can see all the stock setups.

Have fun & be careful in there!    73, Bill  N2BC

[W2VW]

Can the plate load be increased?

[KL7OF]

Can the plate load be increased?

Running the plate at 2500V @ 330 ma....Haven"t been able to load it much more ...Using a bamboo radio for an RF driver.. about 35 watts drive available....
We are using pulls of ughknown quality...It would be nice to have some new tubes..

The best performing tubes we have will give 500 watts out at 2500 V anode only if the screens are pumped up to 700-800 volts...When I lower the screens to 500 or lower, the power out falls waaay off.

[W2VW]

Unless I'm missing something it looks like plate loading should be increased and screen voltage reduced. Re-dip tuning.

Flat tubes would show up as lack of emission via poor positive peak modulation.

[N2DTS]

Well, get to the bottom of it.
In a safe manner, measure the screen current and voltage at the screens.

I am not sure you could run an actual 800 volts on the screens without them melting, not in class C.
In AB1, I think you run the screen voltage much higher, so 800 volts might be normal.

How are the screens in the RF deck modulated?
I would think in a broadcast transmitter, they would use a mod transformer winding.

[kb3ouk]

Where are you measuring the 800 volts, at the supply or the tube? I have to wonder if there might be a dropping resistor between the supply and the tube to limit the screen current. So with the supply at 800 volts you have the proper screen voltage after the resistor, but lowering the supply voltage causes the voltage after the resistor to drop below 500 volts.

[W1ITT]

The power in a tetrode varies as the "three-halves power" of the change in the screen voltage.  That means that the amount of change is raised to the 1.5 exponent.  This is the "three halves power law" that you may have read about.  Which tells us that variations in screen voltage are rather hot controls for the tube.  In tube type broadcast transmitters the most common way to adjust power was by running screen voltage up and down.
The screen dissipation in a 4-400 is 35 watts, so  be very cautious about goosing it by merely cranking the screen voltage up.  All of the math assumes a tube that hasn't lost most of its emission.  And as someone pointed out previously, look for dropping resistors in the screen lead.

[WU2D]

Way out in left field but still on the screen grid control thought. I saw this effect when I was fooling around with Hartley type regen receivers like the Novice Special. The regeneration control is the screen voltage. Varying the plate voltage even by 100V had little or no effect on the feedback or gain. Thus is the beauty of electron coupling as a buffer.   

[KA8WTK]

  I may have missed it, but I don't see mention of how much screen current is being drawn or what grid drive is. With my single 4-400 I adjust the screen supply via a variac to keep the screen current in line as I load it up.
  Conditions are: Grid - 12 ma, Screen - 30 ma., Plate current - 250 ma., Plate Voltage - 2000 volts, 365 watts output
  I could be wrong, but for two tubes at 2500 volts, shouldn't you see 56 ma. total from the screen supply, about 24 ma. grid drive and load that rig to about 550 ma. for maximum output? These figures are double the single tube values for a plate modulated 4-400A in the Eimac datasheet.

[KD6VXI]

Hey,

For more drive,  some plasdick radios will put out close to 100 watts carrier in FM mode.

Might help with more grid drive.....

--Shane
KD6VXI

[N8ETQ]

   Shorted/Blown screen bypass cap(s)

  That would be interesting:

"  I may have missed it, but I don't see mention of how much screen current is being drawn"

Just sayin'

/Dan

[kb3ouk]

Hey,

For more drive,  some plasdick radios will put out close to 100 watts carrier in FM mode.

Might help with more grid drive.....

--Shane
KD6VXI

Even in AM it should put out close to 100 watts if you aren't modulating it. My Icom 718 will do close to 100 watts of carrier on AM, I have the internal carrier adjustment turned up all the way though too, but it will still do it, just won't modulate at 100 watts carrier.

[KL7OF]

This transmitter has  800 volt screen supplies...One 220volt input for two 800 volt outputs...one each for the modulators and the finals...It originally had 4-500s ......NOW it has 4-400s    I want to run the modulators in class AB1 or AB2 so I can use the supply at around 700 750 volts at the tube pin..got that part working...The final is ClassC....I will supply the screen thru a 50K resistor off the B+ on the choke side of the Mod transformer and get it down around 500 volts......I have measured the voltages at the tube pins...I was not able to get the screen current section of the Mcmartian "test meter" to function The original  screen circuit had 800 volts fed to 2 sliders on  a low ohm 200 W resistor with the ends of the resistor going to the screen pins...It wasn't a voltage dropping resistor , more for balancing the tubes...  I was using a Yaesu 901 with the output going direct to the grids of the finals thru a 1000 pf doorknob....By the way, I ran this transmitter with 775 volts on the screens and they didn't melt...I know of another Mcmartian BA1k that is being used successfully  on the ham bands with 775 +on the screens class C.....I prefer 500 volts...

[VE3AJM]

Will/did that transceiver driving the final tubes grids provide you with enough grid drive? ie. 20 - 25 ma.

Al VE3AJM