Pie Network , Transmitter Tuning and Loading Questions

[Y2KEDDIE]

Pie Network Transmitter Tuning and Loading questions:
From reading texts I am aware that a pie network is needed to match the load (antenna) to the source (tube circuit).
I also am aware when C2 is unmeshed( minimum capacitance) , it’s reactance is highest and maximum current from the tank circuit is delivered to the load.
I’ve redrawn the pie network for my own study and visualize that as C2 meshes, it’s reactance decreases and shunts the output to ground, delivering less power to the load. Is this visualization / assumption correct?
I understand that operating the source with the correct plate voltage and plate current optimizes effiency. Is this why a range of capacitance is needed to get correct loading? Why not use minimum capacitance all the time and thus maximum loading all the time. Wouldn’t one want all the output one can get?
This brings up another question: Why do different Transmitters have different values for C2.The examples I am noting are from various tube circuits used in amateur radio homebrew circuits.
Typically I see (C1)365ufd variable is used for the plate tuning, but C2 ranges from 365 ufds to 1300ufds. If maximum loading is achieved with minimum (unmeshed) capacitance then why the extended value?
I’m guessing the higher value 1300+ ufd would give more range and resolution with different loads, but if the load is 50 to 75 ohms (or the same ) ,why would one need the higher value.
I intend on building a transmitter to experiment with and I intend only to use a dummy load fixed at 50 ohms purely resistive. The circuit I am using specifies 1200 ufd C2. Is this really necessary, can’t I achieve proper loading with 365ufd or even less?

 

[hevans1944]

I recommend that you borrow or purchase a copy of any edition of the ARRL Amateur Radio Handbook for information on how pi networks (and electronics in general) work. Those capacitors are RF tuning capacitors and are part of a resonant impedance matching network between the tube and the transmission line or antenna or dummy load. Everything about them is dependent on transmit frequency.

 

[duke37]

You do not want maximum power, that will give maximum smoke.
The output tuned circuit has to transform the load resistance (50Ω) to the optimum load on the final valve. It also has to keep the harmonics down to a reasonable level, a Q of about 12 is often used to achieve this.

Loads are rarely 50Ω resistive and C2 will vary to cancel reactance and to correct for different resistances.

As C2 is reduced, C1 will need to be changed to keep the circuit in resonance and the load resistance will need to be raised to provide the design output.

 

[Y2KEDDIE]

I understand the need for impedance matching.I read the ARRL text but still ask the same questions. I think I understandbut I'd just like confirmation. As I stated I plan to build a transmitter and use a 50 Ohm dummy load,so do I stil lneed the C2 capacitor and if I do what value?

I'm thinking most commercial transmitters specify a given plate current, and you wouldn't want to exceed that,hence the need for a loading control,but I never seen a specification on a homebrew transmitter. Most adjust for a dip in plate current and maximum output (minimum loading), hence the question: If I use minimum loading (capacitor unmeshed) why have it?

 

[duke37]

You must NOT use maximum output, you must keep within the valve limits.

The normal method of tuning is to set C1 to dip the anode current with C2 maximum, then reduce C2 a little and set C1 back to resonance. The current will then be a little more than previously. This is repeated until the current at dip is up to the design value. You have to be quick on the tuning or you may overload the valve. Big valves with carbon anodes can stand some overload for a few seconds.
It does not matter whether the transmitter is commercial or home brew.

The values of L1, C1 and C2 can be calculated but will need to be adjusted to cater for circuit variations.Do not forget the anode choke.

An EL84 may be run at 250V 40mA
A 6146 or 807 may run at 600V 100mA
An 813 may run at 2000V 100mA
These are all by memory but will not be far out. Look up the data for the valve you have.
Excess power can melt the glass envelope or anode or strip the cathode.
The output power will be about 50% of input power.

 

[hevans1944]

You still need C2 because its reactance is part of the tuned circuit, consisting of C1, C2 and L, that matches the output impedance of the tube to the load impedance. But since this is an "experiment" you are free to do as you please. Please let us know how that turns out for you.

The value of C2, in a properly designed pi-network, depends on the value of C1, L, the Q of the tuned circuit, and the transmit frequency as well as the two impedances you want to match. Many combinations of C1, C2 and L will work for a given combination of Q, frequency, and impedances to be matched.

So what frequency do you intend to use and what type of tube? A schematic of the circuit you plan to "experiment" with would allow us to provide a more considered opinion.

 

[Y2KEDDIE]

I was planning on building a circuit using a 6T9 tube ( AF4KBA Directory- Home brew from 1969 -1992 ARRL Handbook). I was going to start with 40 meter operation. The circuit shows the loading capacitor to be 1000ufd. I was hoping to get by with 365mfd.

I was hoping there was a general rule of thumb on selecting values. I guess I will have to pick different values and plug into the formula, working backwards to calculate Loading Cap for load impedance of 50 Ohms. ( I have and will use a Bird, 50 Ohm fixed dummy load).

I have looked at many home brew transmitter circuits but have not found any specifying a optimum plate current. I guess most are limited by the power supply which probably does not provide enough current to damage the tubes used. I will consult the tube manual. I imagine the frequency used is a factor along with plate voltage and current.

Looking at a few commercial amatuer transceivers I noticed not a lot of capacitance in the loading controls. I was thinking that this is because these transmitters are designed for a limited load 50 -75 Ohms, requiring an" antenna tuner" to load into a greater range of impedance.

If I decide to use a different tube, it will have a different Rp and maybe using more capacitance for loading will give more flexibility, If I experiment with different circuits maybe I should start with a 1200 or even 2000ufd loading capacitor.

Next: how to specify a air variable capacitor and where to procure.

Also: In the ARRL Hints and Kinks there was a modulator circuit for the 6T9 transmitter using a Stancor 2871 modulation transformer. Does anyone know where I can find one?.

 

[davenn]

I was planning on building a circuit using a 6T9 tube ( AF4KBA Directory- Home brew from 1969 -1992 ARRL Handbook).

scan a pic of the circuit .... not everyone has ARRL manual and those that do may not have that year


nice to hear you have your amateur radio licence, what is your callsign ?

cheers
Dave
VK2TDN

 

[duke37]

From what I see, the 6T9 will be similar to an EL84 audio valve or 5763 or 6BW6 transmitting valves. You should run at 250V and 40mA. The anode load will be about 6k.

C2 will be much less than 1000pF at 7Mhz. You can find the equations and calculate it. The voltage will be fairly low so an air spaced capacitor as used in valve receivers will be suitable. These are usually dual section but you may need only one section.

 

[hevans1944]

... nice to hear you have your amateur radio licence, what is your callsign ? ...

@Y2KEDDIE may not have an amateur radio license yet, or even be interested in obtaining one. However he is a seasoned, experienced, and well-qualified electronics technician. As long as he keeps his "transmitter experiment" confined to exciting a non-radiating dummy load, i.e., the Bird watt meter with 50Ω dummy load, there is no need for a license in the United States of America.

There are many RF design engineers that do not have an amateur radio license or any other kind of license or even a college degree, all of which is perfectly legal here in the States. I can relate to @Y2KEDDIE because from 1967, when my Novice license expired (non-renewable back then), until 2013 when I obtained an Amateur Extra class license, I was fully immersed in technology, some of it involving industrial RF power applications, mostly in unlicensed ISM bands. Through all those years I told myself I didn't have time for the distractions of an amateur radio hobby. However the ham radio bug had already bit me. I missed those late nights listening for faint CW signals on 80 m and trying to contact those stations with my homebrew 6146B rig.

I finally decided to do something about it since I wasn't getting any younger. The FCC had removed the Morse code obstacle years ago, and now publishes both the questions and answers in a pool that is used for actual Technician, General, and Extra class license exams. There are study books and online courses available whose sole purpose is to allow anyone to learn enough to pass the exams. With license in hand, there is a whole world open to experimenters to actually communicate with each other via something other than the Internet and cell phones.

I think there will always be room for discovery and experimentation in RF communications, especially in new digital modulation techniques. For example, most people don't realize that it is possible to recover modulated signals well below the noise floor of a receiver. NASA does this all the time with deep-space probes, giant dish antennas with LNA front-ends, and sophisticated signal processing software. Here on earth Hams take advantage of the Global Positioning System to obtain supremely accurate clocks for coherent transmission and reception of digitally modulated signals, even using the Moon as a passive reflector (EME communications). I would encourage @Y2KEDDIE to pursue at least a Technician ham radio license, if he does not already possess an amateur radio (Ham) license. A Ham radio license is easy and inexpensive to obtain. Affording a rig to play with is another story, but many Hams start out by building their own.

73 de AC8NS
Hop

 

[Y2KEDDIE]

Thank you, hevans (hop), for your kindwords and true Elmer spirit. I agree with you 100%!
From the Amateur's Code (as found in the front pages of the ARRLHandbook): #4 The Amatuer is friendly....Slow and patient sending , when requested, friendly advice and counsel to the begginer, kindly assistance, cooperation and considerration for the interests of others; these are the marks of the amatuer spirit. -Paul M.Segal.

I have no intentions of using my experimental MOPA on the air. It is to be used with a dummy load as I stated in previous thread. I had an amatuer and commercial license for 20+ years,operated a land mobile repair business,and also served in the military as a 71M20 field radio repairman. That was over 40+years ago. I changed career paths a long time ago, but have rekindled an old interest. I have forgotton much and wish to re-educate myself. I enjoy discussing electronics and learning from different points of view. At this time,I have no interest in pursing Ham radio. I hope this offends no one.

 

[hevans1944]

Thank you, @Y2KEDDIE for your reply. No offense taken for not wanting to pursue Ham radio at this time.

To be honest, except for attending monthly DARA meetings and volunteering twice now at Hamvention® I haven't made much use of my new Ham license... a few on-the-air contacts on 80 m and 40 m, but now my rig sits idle most of the time awaiting a decent antenna and feed line. My step-daughter and her husband-to-be gifted me for Christmas with a 6 m monopole they found at Radio Shack. So "real soon now" I plan to stick up a pole on the garage on which to mount the antenna and see how it works. I also have a 2 m BaoFeng HT that I have used twice as an amateur radio volunteer for the USAF Marathon held at Wright-Patterson AFB. I did make a 2 m repeater contact with another ham last year, but quickly grew bored with the conversation as it appeared we had nothing in common other than Ham radio licenses. DARA is heavily into digital communications, hosting repeaters for the proprietary D-STAR system, which doesn't interest me at all at this time. Nor do I have any interest in operating from the DARA W8BI club station although they have some fantastic antennas there. I want to re-acquire my Morse code abilities and have CW rag-chews at low power. For me, "low power" will be ten watts or less using the KX3 barefoot or 100 watts if I fire up the KXPA amplifier, still in its original shipping box.

Which brings us back to your original query about the output loading capacitance... I took a look at my KX3 automagical antenna tuner schematic and discovered it isn't a simple pi network. Lots of relays switching inductors and capacitors in and out between the FET final and the antenna. Nor does it use very large capacitors. But if it's metal, it will load it. Maybe because the output impedance of the "final" is relatively low, compared to a vacuum tube plate circuit, it doesn't require a lot of capacitance or inductance to match a wide range of load impedances. That's Just a guess though. It's been many years since I have studied impedance matching networks.

If I were you, I would try a broadcast-band superhet receiver tuning capacitor. These are usually in two sections, with one of the sections typically in the 365 μF maximum range. The capacitor plates are rather closely spaced, but if you keep the power down to 5 watts or less that shouldn't be a problem as 5 W across 50 Ω is only about 15 V or so. If 365 μF turns out not to be enough to match a 50 Ω load, you can parallel it with some silver mica capacitors to bring it into range. IIRC, after dipping the plate current and reducing the loading capacitance, then re-dipping the plate current and reducing the loading capacitance, etc.. you eventually reach a minimum loading capacitance that will result in a reduced plate current at the dip if the capacitance is decreased further, assuming the tube hasn't maxed out on its power delivering capability. This memory may be faulty because it has been a long time since I've loaded up a final amplifier.

Oh, and thanks for keeping the memory of vacuum tubes alive! Stay out of the high voltage though.

 

[Y2KEDDIE]

Yes, be advised my junk box. (boxes) are 50 years old. I have lots of tubes and transformers. I have almost everything one could need except, (LOL) , those few critical components that, (today) are extremely expensive!

My military experience was with the R-390 receivers and T-195 transmitters. I once worked on a GRC-106 and if memory recalls: that transceiver had automatic tuning and loading similar to what you describe of your KX3. There was a motor driven turret with coils for each band segment and lot's of relay switched capacitors.

I like the idea of ganged 365 uf broadcast capacitors. I have an Atwater Kent Model 70 TRF receiver that I thought of cannibalizing and paralleling the three capacitor sections. I just don't have the heart to destroy something that's working and of antique value.

I was also thinking of switching silver mica capacitors in parallel as you suggested. I seen this approach for multi band transmitters for 160 - 10 meter coverage. My original thoughts were to build a single band (40 meter) transmitter hoping I could get away with a smaller value loading capacitor rather than the 1000ufd specified, especially with a fixed 50 Ohm resistive load.

Another thought was to be able to control power output by varying the power supply voltage with a variac once loaded with a fixed. capacitor.

 

[hevans1944]

... I like the idea of ganged 365 uf broadcast capacitors. I have an Atwater Kent Model 70 TRF receiver that I thought of cannibalizing and paralleling the three capacitor sections. I just don't have the heart to destroy something that's working and of antique value.

Yikes! Please don't kill a working Atwater Kent! I know of a fellow here in the Dayton area who restores old radios. He would have a heart attack at the mere thought of cannibalizing a working antique radio to repair or restore a non-working radio. He recently retired from Mendelson's Electronics, a surplus store here in Dayton, and I invited him over to take as many of my "spare" vacuum tubes as he wanted... gratis. Cleaned me out of everything except a few transmitting tubes I couldn't bear to part with, even if he offered to pay for them... which he didn't, not needing them of course.

I was also thinking of switching silver mica capacitors in parallel as you suggested. I seen this approach for multi band transmitters for 160 - 10 meter coverage. My original thoughts were to build a single band (40 meter) transmitter hoping I could get away with a smaller value loading capacitor rather than the 1000ufd specified, especially with a fixed 50 Ohm resistive load.

Well, play around with it. Since the load won't be changing, you can hook up whatever will get the job done then measure the capacitance of the final result to allow you to substitute whatever fixed or variable capacitance you need.

Another thought was to be able to control power output by varying the power supply voltage with a variac once loaded with a fixed. capacitor.

That could work, although it may shift the load line of the power output stage somewhat. After all, AM transmitters IIRC stick an audio power transformer in series with the HV to the final to create amplitude modulation. Took as much AF power input as the RF power input to get 100% modulation though. Hams that did this homebrew fashion needed deep pockets to put out a full gallon (1 KW) AM signal. I had a now-deceased Ham cousin who did this (he hated SSB).

Another once-popular way to vary power output was to vary the screen voltage on the final. This is how the high voltage push-pull power oscillator on a particle accelerator I used to operate and maintain was controlled. The oscillator drove a Cockcroft-Walton voltage multiplier up to 1.7 MV terminal potential. That voltage was sensed with a "field mill" generating voltmeter, mounted several inches away in a tank filled with sulfur hexafluoride gas at 125 psig. The voltmeter output then provided negative feedback to control the accelerator terminal voltage, from about 115 kV to 1.7 MV by controlling the screen voltage of the oscillator tubes.

 

[davenn]

If I were you, I would try a broadcast-band superhet receiver tuning capacitor. These are usually in two sections, with one of the sections typically in the 365 μF maximum range. The capacitor plates are rather closely spaced, but if you keep the power down to 5 watts or less that shouldn't be a problem as 5 W across 50 Ω is only about 15 V or so. If 365 μF......

365 pF I think you mean

uF values are too large for any radio tuning stages

y2keddie .... ohhh just noticed you were doing the same error throughout your posts

broadcast tuning caps are all in the around 200 - 400pF range for max capacitance

 

[hevans1944]

Ooops! Well, @Y2KEDDIE and I know that it's 365 μμF or 365 puffs. Thanks for catching that, Dave.

 

[Y2KEDDIE]

Yes, mmfd or pfd. Schematic was wrong too,lol. That would be a lot of plates!