Grounded Grid Class C

[KL7AJ]

In case a Class C grounded grid amplifier seems a little bizarre, we need to note that until around 1980, the Class C grounded grid ampl8fier was very prevalent in TV broadcast transmitters. Typically, one had a Class C grounded grid final amplifier that was SERIES modulated by a Class A tube in series with the plate supply. The Class C amplifier could run around 65% efficiency if well tuned, but, of course the modulator itself generally ran about 25% efficient...on a good day. Lots of oddball gyrations were employed to increase the modulator efficiency. One of the more common was to run the modulator in Class A during the video portion and run it in a switching mode (fully saturated) during the sync peak. Another trick was to run the modulator in class AB and then use PREDISTORTION to compensate for the non linearity of the modulator tube.

Interestingly, the concept of PREDISTORTION can be applied to a Class C CW transmitter to clean up the keyed waveform. Unlike old school key shaping, such as an RC time constant in the cathode, predistortion can be applied anywhere. You can use a predistorted SSB transmitter to drive a Class C "afterburner" and achieve extremely clean keying.

So,...there's always more than one way to skin an earthworm.

Eric

[KM1H]

You can also plate modulate a GG Class C amp but it also helps to apply a low level modulation to the driver stage as well to get the percentage up.

Carl

[WA7KKP]

I've worked around tv transmitters for 30+ years, and the only grounded grid amplifiers were class B, amplifying the MODULATED vsb tv signal. I have fond memories of the "stack" -- 7 5762 triodes in g-g parallel. These were common in most high powered television stations -- a cylindrical 'stack' on a squat base which had separate meters for each tube (grid and cathode current), and the lever switches for tuning, which was done by motors.

Most of the designs I've seen used a Class C grid modulated stage, with linear amps tuned (broadbanded) to cover the 4.5 MHz bandwidth of the vsb signal. RCA tv transmitters did that for years, until Harris came along with a solid-state modulator at the one watt level, amplified up through several stages to get to the output power level.

Since a G-G amplifier takes a fair amount of drive because of feedthrough power, there wasn't much demand for a Class C amp with 10 db or less of gain. Tetrodes were nearly 20 db gain (sometimes more), and the old triode amplifiers were somewheres in the middle of that range. Most engineers wanted the most for the least (circuitry), and that is why we have grounded cathode designs.

Gary WA7KKP

[KL7AJ]

Hi Gary.

I got to install one of Harris's first BT5L2 transmitters...the first one to use low level modulation and LINEAR (Class AB)amplifiers from thereon out. Such a pleasure to work on after that ancient Generous Electric transmitter. Just the "Vegetable Sideband" filter on that beast filled up a whole room, which ran at about 120 degrees because you burned up almost half the power in the VSB reject load! Until they went HDTV, Channel 2 in Fairbanks STILL operated a modified RCA transmitter (nodified from B&W to COLOR back in the 1950s). That had a room-sized VSB filter too!

That GE and RCA are two boat anchors I'll never miss!

Eric

[W8JI]

To be able to reach 100% modulation without significant distortion when plate or collector modulated, the stage cannot have any substantial feed-through power and the output power must closely follow the square of the anode voltage. This is why tetrodes, bipolar transistors, and grounded grid stages are problems even when class C, and need modulated on an earlier stage or an additional element in the output device to be clean.

In my Ranger for example, distortion is horrible if the screen is held fixed. The anode never comes close to reaching 0 volts on negative peaks, or twice B+ on positive 100% peaks, because the screen contributes nearly half of the modulation. If only one element is modulated, the screen would be the choice for least distortion. The waveform below is RF output compared to anode voltage, and line B is +40 volts. Even at +40 minimum anode voltage the combination of screen and plate modulation necessary in a tetrode results in over 100% negative peaks.



73 Tom

[KL7AJ]

The FCC rules for televison really were pretty good at "makiing a silk purse out of a sow's ear." So many compromises had to be made just to have a TV transmitter work at all, and the early FCC rules reflect this. Because of the typical "blowby" as you mention, the FCC set the maximum downward modulation "peak white" at 12.5% carrier power, instead of 0. This accommodated the grounded grid leakage issue, as well as the tetrode issue. However, not all TV transmitters used tetrodes...that old RCA rig "Killer" used some type of external anode glass triode...can't recall the type....some olde geezer in town might remember...HI! And from the earliest days, every transmitter I know of used some sort of low-level predistortion to compensate for the inherent non-linearities of the high power stages. Unfortunately (or not!) I missed out on some of the intermediate technology....my TV career jumped from maintaining one of the oldest transmitters on the planet to installing the very latest and greatest....using SAW filters and such...within 6 months. HI!

Eric

[AF6LJ]

The FCC rules for televison really were pretty good at "makiing a silk purse out of a sow's ear." So many compromises had to be made just to have a TV transmitter work at all, and the early FCC rules reflect this. Because of the typical "blowby" as you mention, the FCC set the maximum downward modulation "peak white" at 12.5% carrier power, instead of 0. This accommodated the grounded grid leakage issue, as well as the tetrode issue. However, not all TV transmitters used tetrodes...that old RCA rig "Killer" used some type of external anode glass triode...can't recall the type....some olde geezer in town might remember...HI! And from the earliest days, every transmitter I know of used some sort of low-level predistortion to compensate for the inherent non-linearities of the high power stages. Unfortunately (or not!) I missed out on some of the intermediate technology....my TV career jumped from maintaining one of the oldest transmitters on the planet to installing the very latest and greatest....using SAW filters and such...within 6 months. HI!

Eric

Talk about Whiplash Eric.

[W0BTU]

... predistortion can be applied anywhere. You can use a predistorted SSB transmitter to drive a Class C "afterburner" and achieve extremely clean keying.

That's a fascinating concept, Eric!

I can see how this could be done without a lot of trouble on CW. But has anyone ever done that on SSB?

[W8JI]

That's a fascinating concept, Eric!

I can see how this could be done without a lot of trouble on CW. But has anyone ever done that on SSB?

Mike,

Everything in engineering is a compromise of options. What is gained has to be weighed against the disadvantages.

When we look at this problem logically with an open mind, we see we might possibly gain only a few percent in efficiency on voice peaks. The price we pay for that is system cost and complexity. The system has to automatically compensate distortion in the PA with very high accuracy. It has to be automatic, with compensation determined real-time from the output envelope, to prevent the system from being extremely critical for PA transfer characteristics. These transfer characteristics change with PA operating voltages, drive level, loading, and tuning.

Why would anyone want to engage in all that complexity and cost just to gain a few percent efficiency at peaks?

73 Tom

[W0BTU]

Why would anyone want to engage in all that complexity and cost just to gain a few percent efficiency at peaks?

Somebody who's not in their right mind, maybe. :-)

I thought Eric might have been saying that it's been done on SSB. That sounded pretty far-fetched, but if it has been done, I would sure be interested in reading exactly how that feat was accomplished.