[Amps] GS35B on 144 Mhz

TexasRF@aol.com TexasRF at aol.com
Mon Jan 23 13:58:18 EST 2006


In a message dated 1/23/2006 12:53:00 A.M. Central Standard Time,  
vk4tzl at bigpond.net.au writes:

Hi Guys,

I'm just in the final stages of tuning up a GS35B on  144 MHx. The 
original design was down by DJ5RE. It is featured on the N2DX  QRO Russia 
Site as a project by DG2KBC.

It has all gone together  quite well, till I got to running a bit of 
power into it. The Input side  of things has tuned up well, with a low 
SWR. However, I'm am having  trouble with the Anode Tuning. As I am a 
Ham, I changed things...instead  of using a Flapper for Anode Tuning, I 
tracked down some very nicely made  brass Anode and RF Output threaded 
Rods with Brass Disks on the end. I now  have an 80mm Diameter Brass Disc 
that approaches the side of the GS35B  Anode cooler from the front panel 
at right angles to the  Cooler.

The threaded rod runs in a brass sleeve, that is firmly screwed  to the 
front panel, that can be locked with set scew when you get the  tuning 
right. In my mind, it achieves the same result as the brass flapper  that 
does exactly the same thing, but much more crudely.

My  problem...on initial tune up using my h/held with about 2.5 watts 
output,  I could not get any hint of tuning...more a slight increase in 
power as I  wound the Anode Tuning Disc away from the Anode Cooler. I 
figure that the  plate line may to too long ( 240mm * 60 wide * 1.6mm 
thick ) I shortened  the length by around 12mm and found that I could now 
get reasonable power  out...around 60w with 2 w input, but the tuning 
still is at max distance  from the tube with no "peak".  I made up 
another Anode Line, shorter  again at 214mm. Result....the same...Moving 
the Disc away from the Anode  till I run out of Thread just sees the 
Power increase.

Getting  brave, I hit it with 25W....now I see around 300W Output, about 
300ma  Anode I with 100ma Standing bias and 2800v on the Anode. It seems 
stable,  but its got me stumped why I cannot get the Anode to tune with 
the Disc in  nice and close ( which is where I would assume it would be, 
as way out  from the Tube it cannot be having much effect ) I can't 
believe that the  Anode Line is still too long...maybe it is. Anyone got 
a simple way of  proving this ?

At the moment, thinking about it, I could remove the  Anode Tuning Disc 
altogether and get max power out...doesn't sound right  does it ?

This is my first Large Russian Triode, so I'm moving very  carefully lest 
I let loose a small mushroom shaped cloud and really scare  myself :)

The Amp is protected by a G3SEK Board and all DC Conditions  look OK. I 
will be running more Anode Volts when I get then thing  behaving, but at 
present it is spread all over the bence with a different  Anode Supply 
running it.

Any thoughts  ??

73

Glenn
VK4TZL





Hi Glenn, yes I have some thoughts about your amplifier:
 
When you look at the documentation from DG2KBC you see that he is reporting  
only 36% efficiency while other builders report 60% or more.
 
This is a very strong indication that something is wrong with that  
particular design! The plate tuning capacitor is connected in series with a thin  and 
fairly narrow strip of metal to the floor of the plate cavity. The strip of  
metal is almost exactly parallel with the axis of the tube and quite close;  
allowing very heavy magnetic coupling into the rf circuit. The strip basically  
will act as a link coupling loop that is connected in series with the plate  
tuning capacitor to ground. In essence this will shunt an inductance across the  
plate circuit and raise the resonant frequency of circuit. To resonate the  
circuit, the plate resonator has to be lengthened and in this case to about 10  
inches. Whatever losses may be present in the above described "loop"  will  
load the plate circuit and will degrade efficiency.
 
Your method of plate tuning avoids all of the problems mentioned above but  
due to the missing inductive loading of the original design your resonant  
frequency will be much lower. 
To compensate for this, you have to shorten the plate resonator or you  could 
increase the width of the resonator or do both.
 
The fact that your output power increases with a shorter resonator lends  
credibility to scenario presented here.
 
I encourage you to continue your mods to the plate resonator until you have  
passed the desired frequency of resonance and require addition of tune C to 
peak  the output power. I think you are exactly on the right track already.
 
You can also do the resonance tests with amplifier off by feeding a low  
power signal through an swr meter into the rf output connector. As you vary the  
frequency of the low power signal you can see a definite dip in the swr as you  
cross resonance.
 
Hope this makes sense and is helpful.
 
73/K5GW
 
 



More information about the Amps mailing list