GS-35B 6 meter amplifier

(NOTE:  don't ask for a schematic, I don't have one. If you need a cookbook, look someplace else. Don't ask me to build one for you, I won't do it.)

This project was conceived because of a need for more power for my EME operations. Initially, I was going to make this a "showcase" amplifier using a 4CX1600B but as I got into the design it became obvious that wasn't going to happen. The 4CX1600B design was discarded and was replaced with a GS-35B. One of the major factors influencing this amplifiers design was the need to fit it physically into an existing spot on my operating desk. The odd chassis size and shape are a direct result of this requirement. Unlike most of my amp projects where the electrical design drives the mechanical, this effort required the reverse.

Web research showed that I would be drive limited with the GS-35B tube because of it's relative low gain. I wanted to have  the full 1500 watts output without straining the exciter in key down digital mode. With only 3200 VDC available for the anode supply, I had to be very careful about efficiency both in the cathode circuit and the anode circuit.  Most builders seem to favor a "T" match for the cathode network. Since inductors are much less efficient than capacitors, I used a Pi network. The single inductor is small and efficient and the two capacitors are made up from parallel silver micas.  The wisdom of this choice seems to be obvious since I only need 75 watts of drive to obtain 1500 watts output with only 3200 VDC on the anode.

In the photo of the cathode compartment below the Pi network is just to the left of the tube base. The heat sink above the tube base is the cathode bias regulator. The circuit just below the filament transformer is the filament in-rush current limiter.

A side view shows the anode compartment construction.  I wanted to use a "squeeze coil" tuning system inspired by WA7TZY's excellent 8877 amp but did not have the equipment necessary to fabricate the parts needed. I settled on a "flapper" cap tune system. The required anode tune "C" was very small and the parts were easily fabricated with the equipment I have. Like Fred's 8877 amp, I used a "L" network and placed the anode coupling capacitor between the anode inductor and the loading capacitor. This placement allows the RF choke to be smaller and easier to implement. The chimney is a polystyrene sewer pipe fitting with a ring of card stock paper glued to the top rim.  No need for teflon or ceramic if you pay attention to how the material is placed with respect to the chassis and high RF voltage components.

Looking at the anode compartment from the top:

A look at the rear shows the ventilation holes are in the back instead of the top. Due to the placement of the chassis in my operating desk, air flow out of the top of the chassis would be blocked. This made it necessary to vent the air out of the rear.  Not my first choice but it worked out OK. With 1500 out key down the temperature rise is acceptable. No thermal drift is apparent at full power.

The simple front panel:

This amp project was almost boring. There were no surprises. Everything worked out as planned and the on-the-air performance is everything I had hoped for.

Power out = 1500 Watts (key down, no PEP bullshit)
Drive power  = 75 watts
Anode voltage = 3200VDC
Anode current = 820 ma.
Grid current = no idea. No need to measure this unless you are using a marginal tube like an 8877!