Solid State Heising

[KB3DKS]

 I haven't seen this discussed anywhere and maybe I'm missing the obvious.
If you use a solid state amp to drive a reversed audio output transformer configured in so called heising via a capacitor and mod choke for modulation what keeps the final plate voltage from back coupling and blowing out the output transistors in the driving amp when the plate supply is turned on?
A relay?

I have a couple of 100watt 70 volt tube type output trannys and a nice single rack space 100 watt solid state 70volt line amp. None of which I have a real use for.

Might give the VKII a little more Mod and some more RF headroom without the mudulator current draw.

Bill

[KA2DZT]

Bill,

May depend on how you connect the Heising set-up.  I'm thinking, if you apply B+ to the bottom of the secondary and the mod reactor with the cap between the top of the secondary and RF end of the reactor there may not be any initial current flow through the secondary.

I think, if you connect the top of secondary directly to the RF end of the reactor with the cap connected to the bottom of the secondary to ground, there would be an initial current flow as the cap charged up to the B+.

Probably something of greater concern would be the voltage insulation between the primary and secondary of an output xfmr.

Fred

[KA2DZT]

Bill,

Another thought,  there is a big step-down (looking back) with the output xfmr.  Looking back into the output transistors there should be a very low impedance which should absorb any voltage spike.  Speakers, in their movement, generate voltages back to the amp.  The amp absorbs these voltages to keep the speaker from moving beyond the applied voltage.  This is called the damping factor.  Something like dynamic braking.

Fred

[flintstone mop]

K1JJ definitely had that problem of a 'kick-back' to the solid state finals in his SS modulator....and that was a 200 watt audio amp.
Tom might check in here to answer. I could not find any links to the problem he had and posted.

Fred

[WA1GFZ]

the output voltage of the 70 volt line will be too low.

[W2PFY]

I have done this before with great results both in sound and percentage of modulation. My amplifier is a solid state PA amp with the 70 volt tap but, I chose to use the 16 ohm tap to prevent large voltages from reflecting back.

I was using a 60 watt audio transformer with a 120 watt amplifier and it would fully modulate an Apache.

[Steve - K4HX]

Quite a few have done it. The article below shows how. I modulated a Viking II this way about 15 years ago. It worked FB - up to 200% positive peak modulation. Then again, I was using a 400 watt amp.   



http://www.amwindow.org/tech/htm/obmod.htm

[WU2D]

I modulate my Command transmitter (a pair of 1625's in the final) with SS Modified Hiesing. I used a pair of Mosfets in Push Pull Class B and stuck diodes across the devices Drain to Source (Center Tap). No blow up!

[WD5JKO]

Bill,

   I once coupled a big SS Amp to my Viking I. I describe it here:

See Reply 8:
http://amfone.net/Amforum/index.php?topic=22341.msg162236;topicseen#msg162236

  I eventually used a Harmon Kardon audio output transformer to step up the audio. I did not have a 70v winding, and I recall using the 8 ohm tap. With such a big step down ratio, and the big Amp, the turn on/off transient was not an issue. With your 70V line it might. I might put a varistor across the 70V winding, maybe a V130L10A or similar device.

  In the schematic you see a series dual AC capacitor into the primary. This was because the DC resistance of the 8 ohm input was a tiny fraction of an ohm. With audio of my voice with asymmetry there was a DC component to it which would load the SS amp. The series capacitor fixed that. With your 70V winding tranny, I bet there is sufficient DC winding resistance to negate the need for a series capacitor.

  I wonder what the turns ratio is of your 70V line transformer? I hope it is at least 1:6, or better upward to 1:10 for headroom. So with 1:6 your 100V peak (70v rms) will make 600V peak which will 100% modulate 600v B+ to 1200v peak.

  Since I was experimenting with ultra modulation, the peak AC voltage on the modulated B+ line was way high for 600V wiring in a harness. I had to wire the modulated B+ separately and do so like the B+ to a big tube RF amplifier.

  This was the best sounding, and loudest AM signal ever from my station. I called it: "super modulation with progressive negative cycle unloading with carrier boost". 

   Those that say ultra modulation sounds like crap never heard this rig. I still got it on the shelf since dismantling it when I moved in 1996.

Jim
WD5JKO

[Bill, KD0HG]

I agree with Jim. Been using the 3-diode Ultramod design in the big rig for years. Sounds FB.  IMO the main benefit of the Ultramod design is the prevention of splatter by preventing >100% on the negative side, an important thing if running QRO on a crowded band.

It's not the Ultramod circuit that makes someone occasionally sound like crap, the problem is on the receiver end. I've experimented some with modern broadcast transmitters and asymmetrical modulation and you can hear distortion begin creeping up on a simple diode detector receiver at about 120%-125%. At that point it's not bad at all, but you can begin to hear it. Beyond that point, it's a matter of discretion playing the inevitable "loudness vs. quality" game...No free lunch.

Bill

[Bill, KD0HG]

For the noobs to AM (and my own dead brain cells), here's a breakdown of how much audio power you need to plate modulate a 100 watt transmitter, exclusive of losses:

MOD %       AVG INPUT     PEP INPUT     PEAK VOLTAGE        AUDIO PWR
                  TO FINAL       TO FINAL         ON FINAL             REQUIRED

0%              100 watts           100 watts            1,000 V           0 watts
100%           150 watts           400 watts            2,000 V           50 watts
150%           212 watts           600 watts            2,400 V           112 watts
200%           300 watts           900 watts            3,000 V           200 watts

Voltages and tube dissipation start to get hairy after a while.

Bill

[KB3DKS]

A lot of good info here!
I think my original question has been well answered. The back voltage that the SS amp see will likely not be a serious problem. I forgot about the stepdown ratio there. The "ultramod" enhancement is going a little beyond but interesting never the less. Amjust trying to squeeze a little more mod out and gain some RF headroom by taking the modulator load off the plate transformer. As it is the Iron barely gets warm even after an old buzzard so the possibility of adding another 6146 or loafing with a pair of 4d32s with only minor physical modding may be possible.
The 125% mod with a Valiant RF level would work out well for me with the indoor ant.
  The 70v line transformers I have came from a Fannon PA amp with 4 6L6s in PPP at about 425v.
There are windings for 4-8-24v and 70v so finding a good ratio should be easy. The solid state amp could also be bridged.
 There is also a large Bell Tower amp 70V transformer that had 2 6550s at about 475v but it is only 70V Line. A nice potted StanCor unit as large as a DX100 plate transformer.
  How would you go about implementing the 3 diode circuit when there is no DC thru the modulation transformer as in this application?

Bill

[Steve - K4HX]

Put the three diode circuit across the modulation reactor.

[KB3DKS]

Put the three diode circuit across the modulation reactor.

Ah! suspected as much. Now to find the right reactors. I have a KW one but it weighs about as much as the VKII an am currently unable to even pick it up.
Bill

[Steve - K4HX]

LOL! That would be funny. I used a 250 watt broadcast tranny with a Globe Champ once. The mod tranny weighed about as much the the Champ.

I've never done it, but some have series up several smaller power supply type chokes to make a mod reactor.

Put the three diode circuit across the modulation reactor.

Ah! suspected as much. Now to find the right reactors. I have a KW one but it weighs about as much as the VKII an am currently unable to even pick it up.
Bill

[KA2DZT]

Bill,

The diode neg peak limiter gets connected to RF terminal of the mod reactor.  You need high speed diodes.  Ordinary PS diodes are usually not fast enough.  Even with 600-800 volts on the plate you need a PIV of at least 4KV or better.

I use a circuit that uses only one diode which works FB.  You don't need the load resistor on the mod xfmr or the diode that's connected to it.  Also, you don't need the steering diode.

Build a small DC supply of about 10% of your plate voltage with the usual filter caps.  Use a RCRC filter. A backwards connected filament xfmr will work.  After your last filter cap add another 3K 2W resistor with maybe a .1 cap to ground.  Now connect your diode to the 3K resistor.  The cathode end goes to the RF end of the mod reactor.

Any load on the mod xfmr, everyone thinks is needed, is presented by the internal impedance of the small power supply.

You will need to use a string of diodes to get the 4KV PIV.  I also put .5meg equalizing resistors across each diode (no caps) in the string.  Many folks will say you don't need the resistors but I use them regardless.

Fred