Thoughts of QRO...

[K5UJ]

Steve,

Thanks for taking the time to write these lengthy explanations.  A couple of questions:

What is meant by "plate to plate" in the context of impedance for example:  "...plate to plate impedance of 9500 ohms ..."

I am unfamiliar with this experssion.  It sounds like it means modulator plates to RF PA plate(s) but I have trouble understanding that since in the path there is a mod. transformer with a Z ratio.   Then there is a power supply choke in the path which I guess must have some impedance and the tube anode impedance. 

My other question has to do with the problem of getting the modulator voltage to zero.   I thought the modulator v. had to swing from B+ acrross zero to the negative value of B+ to cut off the carrier.

thanks,

Rob

[The Slab Bacon]

Rob,
      You are close but just off by 50%. In a perfect world the modulator output for a plate modulated transmiter has to produce an AC voltage symetrical with the input audio. And also must produce enough AC voltage to swing the plate voltage to the final amplifier from 0v (100% negative peak) to 2X the plate voltage (100% positive peak.

turns ratio is as / if not more important than the impedance ratio as the load impedance will change as the load on the final amp changes (how heavy or light you load the final)

the plate to plate impedance refers to the plate impedance (I x E) of the push-pull modulator tubes.

In chosing a mod transfoma, you want to pick a ratio that will give you the necessary voltage swing at the working impedances you have to deal with. Keeping in mind that load impedance presented to the modders is entirely dependant on the working impedance of the final that you present to it.

Impedance ratings of mod transformers are also good indicators of the voltages / current that it is designet to handle without going south.

I hope this helps     
     

[steve_qix]

Steve,

Thanks for taking the time to write these lengthy explanations.  A couple of questions:

What is meant by "plate to plate" in the context of impedance for example:  "...plate to plate impedance of 9500 ohms ..."

I am unfamiliar with this experssion.  It sounds like it means modulator plates to RF PA plate(s) but I have trouble understanding that since in the path there is a mod. transformer with a Z ratio.   Then there is a power supply choke in the path which I guess must have some impedance and the tube anode impedance. 

My other question has to do with the problem of getting the modulator voltage to zero.   I thought the modulator v. had to swing from B+ acrross zero to the negative value of B+ to cut off the carrier.

thanks,

Rob

Frank pretty much hit it!  The plate-to-plate impedance is the transformer primary impedance, which is dependant on the load on the secondary. So, a transformer with a 1.5:1 impedance ratio, if presented with 5000 ohms across the secondary, will present a 7500 ohm load across the primary (full primary, which is plate-to-plate), and this is the impedance the modulator tube are working into.

The actual impedance is not so much of concern as is the RATIO.  The voltage ratio (also the turns ratio - the same thing) is  the square root of the impedance ratio.

The actual impedance comes into play in 2 areas:

1) Can the modulator tubes deliver the required power into whatever primary impedance is presented to the modulator tubes?  Can the tubes pull the peak current required; do they have the plate dissipation; can they handle the voltage, etc....

2) Can the transformer operate at the impedance in question --> transformer voltage, current, capacitance and inductance all all factors here.

The voltage swing at the primary of the transformer will vary from approximately 0V using an ideal modulator triode, to approximately twice the B+ under the same conditions.  When the tube on one side of the transformer primary is turned on hard, and approaches 0V, the tube on the other side of the primary (turned off) will experience appximately twice the B+ at its plate.

At the secondary, the voltage across the secondary will depend on the turns ratio of the transformer, and will be an AC waveform, reversing polarity every half cycle.  The voltage across the secondary is added to, or subtracted from, the DC voltage that is delivered to the load, either through the transformer secondary itself, or coupled to the load through a capacitor/inductor combination (so-called "modified Heising").

Hope this also helps!

Regards,

Steve

[K5UJ]

Thanks Slab and Steve; I have to print this out and study it.  Slab I hope you are getting better. 

I forgot to mention before that the ART13 mod. trans. is 2:1 Z ratio, so the winding is around 1.4:1 & should not be too bad.  I think my natural voice is around 120 to 130 % assymetric which is all I care about.  The rig I'm thinking about is my TR-1, single 813 300 w. input, 1.5 KV RF and modulator, pair 811As for audio, 866As in power supply.  10 Meg audio input.  Once I get it going stock I want to temporarily wire in a mod reactor and capacitor as an experiment to see what that does for the audio.  I have a JT30 mic. around here somewhere (if I didn't screw up and give it to someone at some point in the past) that should be just the right Z for this rig.   going to hunt for it this weekend.

73
Rob

[steve_qix]

Thanks Slab and Steve; I have to print this out and study it.  Slab I hope you are getting better. 

I forgot to mention before that the ART13 mod. trans. is 2:1 Z ratio, so the winding is around 1.4:1 & should not be too bad.  I think my natural voice is around 120 to 130 % assymetric which is all I care about.  The rig I'm thinking about is my TR-1, single 813 300 w. input, 1.5 KV RF and modulator, pair 811As for audio, 866As in power supply.  10 Meg audio input.  Once I get it going stock I want to temporarily wire in a mod reactor and capacitor as an experiment to see what that does for the audio.  I have a JT30 mic. around here somewhere (if I didn't screw up and give it to someone at some point in the past) that should be just the right Z for this rig.   going to hunt for it this weekend.

73
Rob

Sounds reasonable.  Strong recommendation: Use a GOOD audio driver - preferably something that's a source follower (the best) or cathode follower (with high current capability tubes and feedback).  The driver will, above any other factor, determine the quality of the audio.

I have actually achieved outstanding audio with a viking II using the stock modulation transformer and 807s (triode connected) using a source follower audio driver, and about 10 dB of feedback around the whole modulator.  And the viking II modulation transfomer is not that good!

The idea with the driver is to provide a perfect voltage source - a voltage source that will deliver whatever voltage is asked of it, regardless of the load.  A grid is a non-linear load (it resembles a diode curve), and because of this, ANY resistance in the driver will cause a non-linear driver slope (the slope should be absolutely linear).  This cannot be achieved in any practical way with tubes - they just have too much internal resistance.  Feedback helps lower the internal resistance, but you won't even come close to what you get with a MOSFET source follower.  Anyway, a suggestion....

Do whatever you can to eliminate the driver transformer.  With only 1 transformer in the system (the modulation transformer), your success with negative feedback will be GREATLY increased.

Regards,

Steve

[K5UJ]

I did some checking.  I am going to try to get back to the original topic but at the same time get into the "source follower" thing if I can.  I looked at the schematic for the TR-1.  The 811As are class B which is good.  They are driven with a phase splitting (for p.p.) transformer which as I understand it, is not good from a fidelity point of view.  I'm not interested in doing any radical mod to the rig because it is rare (adding a reactor and cap to ground are non-invasive) but all of this is possible with homebrew.  Getting back to QRO, I studied the schematic I have for a Sparta 701 1 KW bc rig to see how they did the audio.  The 701 runs pair 4-500A modulated by a pair.  I have been able to slowly learn and understand more about it.  When I first saw the schematic about 18 months ago it looked incredibly complex and discouraging.  Not as much now.  The modulators are directly driven by a pair of p.p. 12BY7As that have their grids connected to the secondaries of a driver transformer.  Each secondary winding (the opposite ends from the 12BY7A grid connections) is connected to the plates of the 4-500As through 5 Meg resistors.  this must be the source follower method.  

The 811As in the TR1 are driven via a transformer by a single 6V6 connected to one end of the driver transformer primary (low v. B+ on the other end).  The secondary is the typical CT secondary with the ends connected to the 811A grids. This is kind of interesting:  the 6V6 plate is also fed back to the cathode of its driver, one side of a 6SL7GT via a 390 K resistor and .05 uF 600 v. cap in series.  The plate of the 6SL7GT is connected directly to a pair of 6V6 grids via a .05 uF cap.  I wonder if this is a cathode follower?  I'm trying to not take this thread over but it looks like I have, sorry about that  

Rob

[Steve - K4HX]

That description does not sound like a cathode follower (source follower would use FETs). It's pretty likely the modulators in the Bauer were running Class AB1 so no driving power was needed, just enough voltage swing. A cathode follower will provide power to the following grids, suitable for Class B modulators.

[DMOD]

Anyone know which BC rigs used a pair of 4-400 modded by a pair?
Probably a good place to start.

Yes, the Sintronic AM1 used that tube circuitry and configuration.

I still have the schematics.

Wokred on many an AM1 in the 60's, and many were used as backup transmitters in the 70's.

Phil - AC0OB

[K5UJ]

That description does not sound like a cathode follower (source follower would use FETs). It's pretty likely the modulators in the Bauer were running Class AB1 so no driving power was needed, just enough voltage swing. A cathode follower will provide power to the following grids, suitable for Class B modulators.

Okay, I think I was getting s.s. and vacuum tube circuits confused without realizing it.   Still, the driver plate feedback to the cathode on the preamp tube in the TR-1 looks like some kind of control method of some sort.  I'll just have to learn more about it.  Thanks

R.