The AM Forum
March 28, 2024, 03:57:53 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: RCA BTA-1R1, Pi networks, and remote tuning  (Read 5082 times)
0 Members and 1 Guest are viewing this topic.
K8DI
Contributing
Member
*
Offline Offline

Posts: 423


« on: October 11, 2021, 03:16:16 PM »

So, as an update, the BTA is assembled except for tubes, I've cleaned a lot of stuff, dust, etc. and tightened a bunch of hardware, replaced some that was missing, reconnected things, traced out a couple hanging wires and got them back in line. I've got the 240 line run and a cordset on the transmitter but I've resisted plugging it in just yet. I need to add an output connector so I can connect a dummy load, and test the HV interlocks. I'm trying to borrow a hipot tester to give the inductors a test before applying real power to them.

And then comes vacation..My wife and I leave tomorrow for New York and Broadway shows for a week...but I did stick a visit to New Hampshire on Friday morning in there to check out NearFest. never been, hope to see some of you there.

After vacation, I will fire it up.  And after it works, strip out the output network and do a new one.

I have:  a 15kV vacuum cap for Tune...a handful of used coils that are in there now...a gates roller....a gas cap for Load (Comet X-cap, 1.8kV/1600pF)...a couple vacuum relays to switch inductor taps.

I want: to figure out how to use a pair of servomotors to turn the caps.  The vacuum cap is many turns, the gas cap is 180 degrees. 

Anybody have any experience doing this? I remember a tuner someone built, but that was arduino/etc.  I would prefer something less digital, like a knob I turn inside the shack that turns a knob out in the garage.
Would love to hear if someone has invented this wheel before.

Ed

Logged

Ed, K8DI, warming the air with RF, and working on lighting the shack with thoriated tungsten and mercury vapor...
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #1 on: October 11, 2021, 07:15:07 PM »

I'll take a shot..

The 'pair of servos' in analog is a workable scheme, but there are many kinds of servos including those in which one side has no moving parts. Yours is complicated because you have to count turns on the vacuum cap and remember the position when the power is off.

By pair of servos, you may mean what some people call Selsyn motors. They are used in old radar units where one is turned by the antenna and one then turns the deflection yoke on the CRT, or where one is attached to a compass dial and one is attached to the ham or TV antenna rotor. These give a precise positioning around a 360 degree circle. Those are kinda low torque for big caps, even the larger motor sizes. They don't count turns, so a turns counter would be needed on both ends. They can skip a turn, if the one is held against the torque and the other is turned hard and is forced past the spot. They are used with amplifiers and other electronics in-between them. I just thought to get those out of the way because they are not too suited for driving caps.

A simple kind for vacuum caps is a DC gear-head motor with a limit switch to stop the motor in that direction, when it is at the number of turns. The limit switch can be attached to a gear reduction so that after X turns the switch operates.

A better kind of this does not use a gear reduction, but a stack of special 'washers', one keyed to the shaft and the rest free to rotate, all of them having a bent-over tab on their circumference, and there is one washer per desired turn, so that as one turn is completed, the 'driving' washer, keyed to the shaft, has its tab contact the tab of the next washer, and then after a second turn is completed, that washer's tab contacts the tab of the third washer, rotating it, and so on, until at the end of 20 (or whatever) turns, all 20 (or whatever) washers are now being forced to turn. The 20th washer's tab is special in that it has a part that sticks out farther and operates a microswitch to stop the motor in that direction. There is another microswitch near the last washer that is operated after all 20 turns are reversed and the last washer is finally driven the opposite way to stop the motor in the reverse direction. This works and can not forget its position. A simple momentary up/off/down switch on the control panel has only to apply the dc voltage of the polarity to move the motor the direction wanted.

An enhancement is, when the gearmotor is stopped, to use a spare set of contacts to momentarily put a low value resistor across it so that it brakes the motor.

This scheme is not very compatible with fast or variable speed tuning because the motor must stop right away when ordered. It should be OK for 1 turn per second though and then a slow speed for fine tuning.

The motor could also have a current sensor so that if something binds, it will trip and require intervention reset before anything gets broken.

I have searched for this 'stack of washers' rotation limiting switch assy to no avail, but perhaps others have seen this thing. It is real because I have two commercially made units here for myself, but I don't know the proper name for it.

The thing with vacuum caps is always the limiting switches. The issue with stepping motors and counting stepls is that the thing can forget its position and then how can it be zeroed?

So anyway, no processors, no digital junk, just analog.

There are better analog systems, such as used in the ARC-105 HF transmitter to automatically operate the tuning coil (attachment) though it is complicated. It's an example of a servo automatic tuning using an AC motor and a servo amplifier to drive it, from a signal from a phase discriminator.




* ARC-105 400W amp hookups20210526.pdf (684.88 KB - downloaded 177 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
w8khk
Member

Offline Offline

Posts: 1200


This ham got his ticket the old fashioned way.


WWW
« Reply #2 on: October 11, 2021, 10:24:37 PM »

Pat has some good ideas here, and I would like to add a few other ideas to the mix.  First of all, I recall the tuning method used for the Collins ARC-65 sideband transmitter, an the associated antenna tuner.  That rig was used in the Boeing B-17 StratoFortress, and I had the opportunity to pull one out of a B-17 at the Davis Monthan Boneyard in Tucson, before the aircraft went to the shredder.  

Here are some pics of those two units:  https://aafradio.org/flightdeck/arc-65.htm
I still have the complete antenna tuning unit, and the final deck of the SSB transmitter.  They entail a rotary inductor, and motor-driven vacuum capacitors.  The antenna tuner had an auto-tune feature, with the directional coupler at the coax input to the tuner, and used a trailing wire out the tail of the aircraft for an end-fed Zepp antenna.  The vac caps were driven by a DC motor, geared down, and included a leadscrew running in parallel to the vac-cap shaft, driving a follower to actuate a limit microswitch at each end of the travel, preventing the motor from driving the vac cap shaft too far.  Circuitry in the tuner adjusted the vac cap and the rotary inductor for best match.

A similar vac-cap was included in the transmitter, geared down and limit switched in the same fashion.  If there is interest, I could photograph the devices to provide a detailed understanding of the concept.

Now, instead of trying to make something auto-tune, perhaps it might be better to have a capacitor that follows a knob in the shack, while looking at the plate meter and the forward/reflected power meter.  In addition to a limit protection, there is a rather simple way to accomplish the following action.  Bur for limits, instead of micro-switches, which could fail and compromise the expensive vac cap, consider an optical solution, whereby an LED and optical sensor could be used, where the limit situation could break the photon beam.  Very inexpensive, reliable, and easy to implement physically.  Limit lights could illuminate in the  shack, indicating the end status of either capacitor.

I have a CDR rotator and controller that I used at Davis Monthan.  It has a DC motor, and a pot inside that provides position feedback.  A separate pot is in the control panel.  The two pots form a balanced bridge, and provide a polarity sensitive DC offset when the pot in the rotator and the pot in the shack are not set to equal values.  This offset drives the DC motor, slowly if the offset is small, and faster if the offset is great.  This allowed me to position the antenna rotator in small increments, back and forth, at the turn of the knob.  A one-turn pot as in the rotator would not suffice for a vac cap, but I suggest a ten-turn precision linear pot like the ones used for the delay sweep in the Tek and HP scopes.  This could be belt driven from the vac cap shaft with a small timing belt, very cheap and available on amazon and ebay.  A ratio between the pulleys could allow the pot to cover more than ten turns, potentially the entire travel of the vac cap shaft.  Current limiting on the motor drive could control the torque to the point that it could not force the shaft beyond limits.   By the way, the limiting stack of bent washers that Patrick mentioned are used on all the Collins 75A and 32V PTOs, to limit the number of turns the PTO shaft is allowed to turn, effectively setting a virtual stop.  This would probably not be needed with some creative analog logic to drive the motors.

For the drive train, you might find that model airplane servos, or something similar, would give you the needed torque.  They use an internal pot and are limited to about one revolution, but using an external ten-turn pot instead of the internal pot would remove that limitation.  The servos are very easy to drive, with a square-wave oscillator with the ability to offset the balance; the more the offset, the faster the motor turns.  Test devices for the servos are very inexpensive, and include a pot to drive the servo in either direction, eliminating the need to build the variable-duty-cycle square wave generator.  Or, you could discard the electronic logic in the servo assembly and just use the motor and gear train to drive the vac cap shaft.

I hope this gives you some ideas with which to drive the vac caps gently and accurately from the operating position.
Logged

Rick / W8KHK  ex WB2HKX, WB4GNR
"Both politicians and diapers need to be changed often and for the same reason.”   Ronald Reagan

My smart?phone voicetext screws up homophones, but they are crystal clear from my 75 meter plate-modulated AM transmitter
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #3 on: October 11, 2021, 11:34:00 PM »

Very good on the automatic speed control built right in.

Here this guy did something totally analog with 10 turn pots. No overtravel protection or speed control mentioned so its very basic.
https://www.youtube.com/watch?v=OEJw3ZizoX4

just throwing it out there.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #4 on: October 12, 2021, 06:42:27 AM »

Some more ideas...

http://amfone.net/Amforum/index.php?topic=45862.0
Logged
KA3EKH
Member

Offline Offline

Posts: 772



WWW
« Reply #5 on: October 13, 2021, 09:35:02 AM »

You may want to consider running the plate supply at reduced voltage, maybe 120 instead of 240 until you work out your output tank. At that power level things will go really bad in a hurry if you don’t have your input and output tank right. Those transmitters worked the output tubes hard in normal operation and if your tanks are not right you can damage the tubes.
I used a network analyzer and built the input and output tanks up with just filament and bias and no HV to get in the ballpark but imagine if you don’t have a network analyzer you can use something like a grid dip meter to get the tanks close before applying HV but applying the full plate without first being close can be an issue.
You can isolate the 240 volt input to T502 and still operate all the other 240 volt loads in the transmitter and I would use a autotransformer to slowly apply HV while you are trying to work out the output tank. And if you planning on operating above 160 you may also soon discover the tank for the 6146 driver leaves a lot to be desired and you may end up with having equal amounts of drive for not only the carrier but also the first and second harmonics.
I had the benefit of having both network analyzer and spectrum analyzer so was not that bad building up a tank that was on channel but if not careful and if you just go by output power you may end up generating a second harmonic or first lower like if your planning on 3885 you may find that the driver has equal drive to the PA on 3885 and 1942
RCA installed a second harmonic trap in the PA just for this sort of problem, if you were in the low end of the broadcast band you needed to have that trap for the second harmonic. On my 1M I was at 1885 and had to use the trap to cut down on the 3770 second harmonic to get that below 60 dB from the carrier. If you don’t take harmonics into account at that power level and are only attenuating them twenty or thirty dB you can end up having well over a watt or more on that harmonics.
A last thought also is that instead of a high pot tester you can remove all the PA and modulator tubes and apply full HV at that point and if anything is week it will flash over at that time. If everything is good it will not damage the transmitter doing this although you may want to also pull the 6146 driver. Also, at that time you may want to look at C503 and C504 being they often become leaky and if they short they will kill one of the modulator tubes. RCA was big on those direct coupled audio drivers and that often-caused issues in broadcast service.
And what ever you decide to do DON’T disable the interlocks or the HV discharge interlocks for when you open the transmitter! Even at reduced power that transmitter will kill you if you don’t respect it!

Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #6 on: October 13, 2021, 10:01:45 PM »

I love that one. Please post as you get the tuning set up!


An aside which is not the same as multi-turn pi network tuning of the BTA-1R1, but tunes by shorted loop 'variometer' style tuning of the LC tank inductance. It tunes itself to resonance automatically on any band, assuming the load is correct.

It's not what you are working to accomplish but it is different in a refreshing and 100% lowest-possible-tech analog way. Eitel and McCullough were no doubt out to prove a few things for the World's Fair. The article about station W6USA is in RADIO magazine, June 1939, on worldradiohistory.org.
https://worldradiohistory.com/Archive-Radio/30s/Radio-1939-06.pdf


Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
K8DI
Contributing
Member
*
Offline Offline

Posts: 423


« Reply #7 on: October 19, 2021, 09:48:22 AM »

Just as a followup, after walking around and around at NearFest, where I talked to everyone I saw with either servomotors or motorized tuning setups (well, at least those I recognized as motorized). I came away with some ideas. I'm going to use a stepper and an encoder as a first attempt. I've been told if I use the right driver board, I can current limit the motor and it will shut off on the vacuum cap endpoints when the current climbs. I will have to figure out the details of how to tell where the thing is in the range, but it is a first generation start, that I can manage to actually build the hardware/mounting, and will go from there....  I will also help myself by limiting the first output network build to 80 and 40, which simplifies things...

Ed
Logged

Ed, K8DI, warming the air with RF, and working on lighting the shack with thoriated tungsten and mercury vapor...
KC2ZFA
Member

Offline Offline

Posts: 441



« Reply #8 on: October 20, 2021, 10:43:48 AM »

you may also want to look into the stepper motor stuff shown in https://www.trueladderline.com

Peter
Logged
W1RKW
Contributing
Member
*
Offline Offline

Posts: 4410



« Reply #9 on: October 20, 2021, 05:15:11 PM »

Ed,
I was working the Arduino remote tuner.  I mothballed the project because of covid. My shack turned into a work from home work space.  I had it working but just needed to fine tune it.  Hope to pick things up this season and hope to integrate the GFZ audio board into the 813 rig as well now that I'm back at the office almost full time.
Logged

Bob
W1RKW
Home of GORT. A buddy of mine named the 813 rig GORT.
His fear was when I turned it on for the first time life on earth would come to a stand still.
steve_qix
Contributing
Member
*
Offline Offline

Posts: 2599


Bap!


WWW
« Reply #10 on: October 25, 2021, 11:39:50 AM »

Steppers would be the way I would pursue this.  Precision control, lots of torque and control can be facilitated over a network connection - a real advantage.
Logged

High Power, Broadcast Audio and Low Cost?  Check out the class E web site at: http://www.classeradio.org
w9jsw
Two shots of Whisky
Contributing
Member
*
Offline Offline

Posts: 592



« Reply #11 on: October 25, 2021, 07:40:40 PM »

https://danielwilczak101.medium.com/control-a-stepper-motor-using-python-and-a-raspberry-pi-11f67d5a8d6d

Easy tutorial example on how to drive one from a Raspberry Pi.
Logged
K8DI
Contributing
Member
*
Offline Offline

Posts: 423


« Reply #12 on: October 26, 2021, 09:15:05 AM »

Steppers would be the way I would pursue this.  Precision control, lots of torque and control can be facilitated over a network connection - a real advantage.

Can you tell me how the control over network is implemented with steppers? I use networked control for relays, but have no experience with steppers.

Ed
Logged

Ed, K8DI, warming the air with RF, and working on lighting the shack with thoriated tungsten and mercury vapor...
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.062 seconds with 19 queries.