Using SS PA amps as modulators?

[wb1ead]

Gud Morning all..I may have asked this question once upon a time but can't remember the answer..I have a few rescued PA amps from a local American Legion hall..one a 20 watter the other 2 are 25/30 watters..so considering that there are a number of well I guess PW transmitters out there like the AT-1/DX-20/Johnson Adventurer/Eico 720 and I'm sure many others..has anyone considered using a PA amp to either screen mod or perhaps plate mod one of these rigs?..didn't see a headphone jack so I guess it won't be Hi impedence output but with the right audio transformer or a mod are these valuable?..why do I ask?..well I've dragged these suckers to various fests including Nearfest with not a one even looking/inquiring 'bout them..my price is very very low but no takers which begs me to ask "are these useful/valuable as modulators ?"..thanks if you reply!   73 de DAVE WB1EAD...tired of draggin' dem around...

[N1BCG]

Yes they are.

My first homebrew xmtr used an 807 as a PA and a vacuum tube Bogen VP-40 amplifier that was given to me by the maintenance guy at my grade school. I located a Thordarsen multi-tap transformer with a variety of output impedances, reverse connected it so that they amp has the 8 Ohm windings and the 807 got the 5K primary. It worked GREAT!

Those amps, assuming the power rating is realistic, can fully plate modulate a rig operating at 2x the AF Wattage. Just gotta locate an old output transformer that can be connected backward...

[w4bfs]

unless the transformer has a gapped core, it needs to have a low net dc current in the windings to avoid saturation and consequent distortion .... use of a higher wattage xfmr can help offset this .... best way yet is to do the shunt feed or commonly called heising method ....if the audio pa has a 70V output this is raising the output impedance in the right direction

[KK4YY]

I'm currently using a Bogen PA amp on the receiver side. I have, in the past, used the same amp to drive a pair of modulator tubes on a transmitter. Neither of these are new ideas. I'm currently using the audio amplifier in an old Pioneer stereo receiver from the 1980's to drive my modulator tubes.

I haven't used a consumer amp to directly modulate a transmitter, only as a driver. But, with certain caveats, that too can be done.


Don

[MikeKE0ZUinkcmo]

I've used this Rube Goldburg modulator with ARC-5 transmitters, with good success.





I then modified the stereo amp by paralleling the output stages to make a 100Watt source with high and low gain inputs so I can use it with any audio source.

[wb1ead]

Thanks guys!..lot of gud responses and that is what I thought..you cud press them into service as a modulator..heck the freq response really is tailored for speech although Hi Fi in nature..yes Beefus they all have the 70 volt tap as well as individual taps for 4/8/16 ohms..proper audio transformer remounted inside in place of what's now wud do it..nice set-up there Mike!
                          OK I'll keep "draggin the line"..thanks again guys!   73 de DAVE WB1EAD
PS: if you see me in the "lower" parking lot at Nearfest..say hi!!

 

[WA4WAX]

Here is one approach if you want to screen mod.  8 to 15 watts from the SS amp, along with the right backwards transformer.

Page 38 in the magazine, page 40 on the pdf reader.

Have fun.

http://www.electronicsandbooks.com/eab3/manual/Magazine/C/CQ%20Amateur%20Radio%20US/1952/07%20July%201952.pdf

[kb2vxa]

If you can scrounge an output transformer from a high power hi-fi amp you'd have pretty much the ideal mod iron for your project. A Williamson type is closer to ideal since it has a dual tapped primary as well as a multi tapped secondary giving you more impedance matching options.

If you use a stereo amp many people make the mistake of thinking that paralleling the amps will deliver twice the power, operated in phase buys you nothing. Run 180 degrees out of phase will double the power, think of it as batteries in series vs. parallel or phasing transformers in a buck/boost configuration. First, eliminating the feedback in the phono inputs eliminates the RIAA EQ curve. Then inverting the phase of one input I'll leave up to your imagination since there are several ways, then again substituting an op amp with + and - outputs for the preamps may be easiest. Lastly, using the two high or + outputs (the low - are usually grounded) puts the amps in push-pull configuration.

[KK4YY]

As previously mentioned in this thread, a gapped core transformer would be needed to use as a modulation transformer unless the unbalanced DC can be removed from the secondary (via the "heising" circuit). A scrounged audio output transformer would probably not have a gapped core unless it was used in a single-ended class A audio amplifier (and they can get pricey).

For a relatively low power transmitter, the modified heising circuit shouldn't be to difficult to implement. Along with the scrounged OPT you'll need a good sized choke (which is gapped) to carry the DC.

So, you're looking at two chunks of iron and a nice oil filled capacitor to fit in there somewhere. But, if you do all that right, it can really sound quite good.

[WA1LGQ]

 This system works great on my Viking 1. I use a Randall rackmount 120 watt PA amp into a shunt feed  Heising type setup. Chokes, coupling cap 4 uf and a large old Olson tube output transformer. Remember Olson?  Actually I could use a 20 watter on a pseuo-military transmitter that I am working on. Maybe you will have it at Nearfest?
Larry             

[wb1ead]

OK guys..gud info all around..WA4WAX interesting and fairly simple in that screen mod PDF you supplied..that is what I think most of those earlier CW only rigs were mentioning in their op manuals to do..screen mod..however considering the RMS output of other PA amps including the 3 we have..plate mod is also doable with a little extra effort..heck most of the rigs mentioned are one bottle RF types anyway..tnx to info and suggestions from KB2VXA and KK4YY..no I'm not into any of those rigs or need a modulation amp but all this info I'm sure will be of service to others..Larry WA1LGQ yes I'm "draggin" them down to Nearfest unless the weather is bad enuf..99% sure of Friday..for a Lincoln there's an MPA-20..for two MPA-25's one working the other a parts mule maybe 2 Lincolns wud be fair..and in all fairness I did'nt want this to be a posting for the "For Sale" section so I'm sorta saying yah but hopefully I don't overstep my bounds....Hopefully someone will post a "Are ya going to Nearfest ?" soon..THANK YOU GUYS!  73 de DAVE WB1EAD

[kb2vxa]

Speaking of Moose country......

[wb1ead]

Ya know Warren when I saw that pic one thought crossed my mind..the snow pack really DID get that high?..hard as heck to believe so I'm assuming it was "man made"..in 45 years in our current home we've only seen one in the backyard and he/she was a youngster..but go only 30 miles or so north and they are plentiful...73 de DAVE WB1EAD

[kb2vxa]

Something I missed back there a way is just a bit confusing: "best way yet is to do the shunt feed or commonly called heising method ....if the audio pa has a 70V output this is raising the output impedance in the right direction" Is that the amp output or the transformer used for mod iron? Either way I don't understand it, solid state amps don't have 70V (600 ohm) outputs or any specific impedance outputs. Then again the lower the impedance load the higher the power output which is why ss outputs hate shorts and shoot smoke at them.

Modified Heising (Heising doesn't use a transformer) is easy to build using an oil filled cap like the one found in a Johnson Viking transmitter HV power supply filled with lovely carcinogenic oil and a swinging choke used in a choke input filter. You need not get fussy over the BMF transformer, there is no need to match impedance accurately, a solid state output has no specific impedance and tolerates anything from about 3 ohms to no load (open circuit).

[KK4YY]

...there is no need to match impedance accurately, a solid state output has no specific impedance and tolerates anything from about 3 ohms to no load (open circuit).

Forget impedance matching for a moment.

To modulate a DC voltage to 100% you need a voltage from the modulator capable of doing that. For example, if you want to plate modulate a tube stage that has 700 volts on the plate, you need to have a voltage from the modulator capable of driving that voltage down to zero and up to 1400 volts (+/- 700V).

The average SS amp will not be able to come close to that (regardless of how much power it can produce) without using a step-up transformer with a high ratio.

After you get the voltage you need, then you can worry about impedance matching for maximum power transfer.


Don

[kb2vxa]

With all due respect sir, we were talking about an output transformer from a high wattage tube hi-fi amp hooked up backward as a modulation transformer in a modified Heising circuit. You may have become confused by my mention of Heising modulation aka constant current Class A modulation not using the transformer, capacitor and swinging choke that Class AB1 modified Heising does. The best example I could come up with for original 1930s vintage Heising modulation is pictured below, note the crystal frequency. (;->) A transformer from a 100W "monoblock" hi-fi amp using KT-88s will have an impedance ratio of typically 7500 ohms plate to plate to 4-8-16 ohms that when the 4 ohm tap is connected to the SS amp the 7500 ohm primary will match what I take is an 807 with 700VDC on the plate perfectly. Then like I said you don't have to worry overmuch about RF plate modulating impedance, it will be reflected back through the transformer to a modulator with no particular output impedance. That would have saved me the math involved in choosing the taps on a 100W UTC "Polypedance" modulation transformer when I built a transmitter using a 6146A modulated by a pair of 807s, but I digress.

[KK4YY]

With all due respect sir, we were talking about an output transformer from a high wattage tube hi-fi amp hooked up backward as a modulation transformer in a modified Heising circuit. You may have become confused by my mention of Heising modulation aka constant current Class A modulation not using the transformer, capacitor and swinging choke that Class AB1 modified Heising does. The best example I could come up with for original 1930s vintage Heising modulation is pictured below, note the crystal frequency. (;->) A transformer from a 100W "monoblock" hi-fi amp using KT-88s will have an impedance ratio of typically 7500 ohms plate to plate to 4-8-16 ohms that when the 4 ohm tap is connected to the SS amp the 7500 ohm primary will match what I take is an 807 with 700VDC on the plate perfectly. Then like I said you don't have to worry overmuch about RF plate modulating impedance, it will be reflected back through the transformer to a modulator with no particular output impedance. That would have saved me the math involved in choosing the taps on a 100W UTC "Polypedance" modulation transformer when I built a transmitter using a 6146A modulated by a pair of 807s, but I digress.

Despited my suggested confusion, I stand by my remarks.

How a SS amp has "no particular output impedance" is a statement that I do find confusing. I would suggest that the amplifiers output impedance is that in which it delivers the maximum power transfer to a load. That is pretty specific. Am I wrong or still just confused?


Don

[AB2EZ]

Don

The intuition, theory, and practices that we all associate with classical tube audio amplifiers are mostly not applicable to modern solid state audio amplifiers.

A modern solid state audio amplifier (which employs large amounts of open loop gain and negative feedback in its transistor output stage) behaves as a voltage source. I.e., with an output impedance which is essentially zero. For example, for any given setting of the volume control, the voltage it produces across an 8 ohm load is the same as the voltage it produces across a 4 ohm load. Likewise the maximum voltage it can produce is independent of the load impedance... except that the maximum current that the amplifier can deliver is limited by heating of the output transistors and the power supply components. The maximum output voltage (for any load that doesn’t cause overheating due to excessive current) is limited by the +/- power supply voltages.

Stu

[w8khk]

Stu, I appreciate your description of the capabilities (and limitations) of "modern" solid-state amplifiers.  I wonder if you are referring to the classic analog complementary-symmetry output amplifiers, class H amplifiers, or possibly PDM implementations, when you mention "modern".  Perhaps the limitations apply to all equally.  Additional insight would be most welcome.

When looking at the limitations of an analog output stage, does not the manufacturer "imply" a specific output impedance for a given power level, based upon the limiting factors of transistor dissipation and heating, as well as power supply load regulation?  In order to achieve a reasonable level of distortion at the rated power, I would think the power supply voltage regulation would play a major role.  Thus would it not be practical to specify an available maximum power level working into a specific load impedance, not only to arrive at the optimal speaker configuration, but also the impedance transformation in modulator applications, to maximize power transfer with minimal distortion and risk of component failure?

I have constructed, and serviced, many analog solid-state amplifiers.  Most of these include output current sensing circuitry, and thus fold back the output power when working into unreasonably low impedance loads, thus protecting the output devices from thermal runaway.

I would suppose another issue is that in amateur modulator applications, we are not so much concerned with continuous steady-state power, rather, we are looking at handling the low duty cycle of the human voice, configuring our installation to handle the occasional peaks with minimal distortion.  The output stage feedback is only as good as the power supply and it's capability to support the feedback's demands.

[KK4YY]

Stu,

Your final point about the maximum attainable output voltage being limited by the power supply rails of the amplifier is essential to the point I was trying to make - a transformer with a sufficient turns ratio is required to bring that voltage high enough to properly modulate the high voltage of a tube RF amp.

I tried to make this point because, as a young ham in my teens, I built an AM transmitter without knowing that important fact and wanted to pass along my lesson-learned.

As to the output impedance of SS amplifiers I will defer to your expertise. I have read (and re-read) your posts in this forum which date back many years. I consider them to be a great technical knowledge base from which I have benefited and enjoyed.


Thanks,
Don

[AB2EZ]

Don

Thank you for the kind words

Best regards

Stu

I have used this approach (modified Heising / solid state audio amplifier) quite successfully in the past, with a variety of transmitters.

The audio amplifier must be rated at 0.5 x the input power level (at carrier) of the modulated RF stage... i.e. 0.5 x the plate voltage x the average plate current... or higher.

In the past, I used an audio amplifier rated at 600W (into a 4 ohm load) to modulate a home brew 375W (RF output at carrier) class C RF amplifier. The plate voltage was 1800V and the average plate current was 0.3A. Therefore, the input power (at carrier) was 540 watts. The audio power required for 100% modulation with a sine wave was 0.5 x 540W = 270W.

At present, I use an audio amplifier rated at 200W to modulate (externally) my Ranger. A 50W amplifier would be more than enough, but I already have a 200W amplifier that I can use for this purpose. Any load greater than 4 ohms will not cause overheating (due to excessive output current) of the amplifier's output transistors or the power supply when the input and the volume control are adjusted to deliver maximum power to the load. I.e. 200W into 4 ohms, 100W into 8 ohms. Since we won't be requiring the full 200W output, a load smaller than 4 ohms would be okay... but l designed this modulator to produce a 4 ohm load on the audio amplifier.

I used a modern, ferrite core, toroidal transformer as a step up transformer.

The Ranger, RF output stage has a modulation resistance of roughly: 525V plate voltage/0.125A plate current = 4200 ohms. The audio amplifier is specified to drive up to 200W RMS into 4 ohm load. If I want to present the audio amplifier with a 4 ohm load, I need a transformer with a turns ratio of around: the square root of 4200/4 = ~ 32.4. This is also the associated voltage step up ratio. I use a backward connected toroidal transformer... with a pair of 120V primary windings (connected in series) and a pair of 7.5V secondary windings (connected in parallel)... rated at 100W total. 240V/7.5V ~ 32 ... so this results in a load, seen by the audio amplifier of 4200 / (32 x 32) = 4.1 ohms.

Note that 200W into a 4 ohm load corresponds to an RMS voltage of 28.3V and a peak voltage of 40V. The peak output voltage of the step up transformer (with 200W of audio input) will be 32 x 40V = 1280 V ... which (as expected) is more than 2.4 x the voltage needed for 100% positive peak modulation. If I used an amplifier rated at 50W into a 4 ohm load, the peak output voltage would be 640V... which is still more than enough to modulate the Ranger to 100% positive peaks.

 

[w4bfs]

hi Stu .... it has been a while 

[Steve - K4HX]

Used this set up on a Viking II. Could easily modulate 150-200% positive with a 200 watt amp. I tube output tranny works FB.

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