Negative peak limiter, yes or no and where?

[AMLOVER]

I am thinking to construct the 3 diodes negative peak limiter, mostly for modulation transformer protection and why not to get some possible negative peak limitation.
I allready use some compressors and limiters before the modulator to prevent overmodulation and thought why not to have an extra protection after the modulator if this will not disturb the audio quality and in the worst case will only protect the irons from trancients or sparking left right?
If wrong please let me know as soon this is a completely unknown area to me.
My main question is whether should it be involved in a mod. heising circuit?
My setting is like in the attachement because of voltage source modern af amplifier in use, I mean one leg of the modulator grounded.
A. between mod.transformer and capacitor?
or
B. after all the heising circuit  and just before the plate rfc?
All suggestions will be appreciated.

Thank you, Stefano

[AB2EZ]

Stefano

If you want to add a negative peak limiter...definitely use circuit B

Since your audio power amplifier is a voltage source (as compared to a classical tetrode-based push pull audio amplifier... which is a current source), I can't think of any scenario in which a negative peak limiter would offer any protection for the modulation transformer or the Heising choke or the audio power amplifier.

A negative peak limiter might improve the sound of your signal at the listener's end... because most receivers employ a diode peak detector.

I added a negative peak limiter (using 2 diodes... as per the attached diagram) to my Johnson Ranger. The keep alive voltage is 50 Volts (about 10% of the B+)

Sometimes I operate the Ranger with its internal tetrode-based push pull audio amplifier and its internal modulation transformer.

Sometimes I operate the Ranger with its internal tetrode-based push pull audio amplifier and its internal modulation transformer... but using a Heising approach to remove the DC from the secondary of the modulation transformer.

Sometimes I operate my Ranger with a modern audio power amplifier (voltage source), and an external modulation (step-up) transformer, in Heising configuration.

In all cases, I use the negative peak limiter. I believe that it allows me to run the Ranger with a somewhat higher average modulation density, because I'm limiting the negative peaks at the plate of the rf output tube.

I do use a moderate amount of audio compression... and my audio chain does have a negative peak limiter... but with the additional negative peak limiter in the Ranger... and particularly when operating with the Ranger's internal audio amplifier and modulation transformer... I'm not worried about accidently applying 100% negative modulation.


Best regards
Stu

[AMLOVER]

This 2 diode limiter seems quiet cool.
I was reading some topics about the limiter and some users said about protection and I was confused.
I attach what will be my project for tomorrow since I have all the diodes (50x1n5408 and a huge lot of resistors to create 3600ohm/500w).
I just attach what I'll do, I hope that keeps the safety factors, too.
I need to have the best rappans against the "linears" of my neighbourhood 
Do you think Yamaha P7000s or Phonic 5100 are enough to modulate 125% my 8kva - 6.5kw carrier ring?

Thanks
Stefano

[AB2EZ]

Stephano

At these high power and high voltage levels... I suggest that you start out without the negative peak limiter installed... until you are comfortable that everything else is working without problems.

Since the negative peak limiter will not offer any protection in this case... I suggest that you proceed with caution.

In theory, for identical diodes in series, the PIV's will add. In theory, even if the junction capacitance of each diode is +/- a few percent of nominal, the PIV's will approximately add (with some diodes having a few percent more voltage across them, and some diodes having a few percent less voltage across them).

Nevertheless, theory and practice often diverge at these high voltage levels due to subtle effects that were not taken into account in the theoretical analysis.

When I was a sophomore at CCNY, my fraternity held a "Monte Carlo Night" party one Saturday night. The fraternity rented various gaming equipment for this purpose. When I arrived at the party, I made a few calculations in my head... and I concluded that the odds being offered by "the house" (the fraternity) on one of the games were actually in my favor.

At the end of the evening, having lost all of my money, I borrowed 15 cents to take the subway home. The subway ride was about 1 hour... and that gave me plenty of time to figure out what was wrong with my earlier calculation.

Best regards
Stu

[AMLOVER]

Steve,

I have fixed the 2 diode limiter and without modulation there is not any problem.
I didn't see any arc or strange thing in my instruments with full hv and full power.
Of course things will change when the first ac volt will raise voltages up but there I will go turtle way...one step per hour.
I could also increase/decrease the keep alive from 0-600v without any bad sign around.
I made a separate keep alive voltage supply and I will add a ma meter + volt meter and I'll enclose it in a plexiglass box (Tom Vu inspiration) with 2 porcelain feedthrough connectors including a knife switch to overpass it if something is strange or it doesn't help.
I used 2 parallel lines of 10 pieces  1n5408 for series diode and 1 line of 30 pieces 1n5408 for connecting the mod B+ with keep alive voltage, 500v in my case. Too many diodes but safety is my priority.
I brought slowly up the keep alive and nothing bad happened, I took away hv and checked the diodes' temperature also nothing was too hot. The 3600 ohm/600w resistor was not hot, too.
May be I'll have the surprises when I connect the af amp and push some ac volts in the modulation transformer primary...let's see!
I managed to repair the old reactor and now I have series connected from B+ to the rf deck 3 reactor chokes
30H/2A (was 40H but I repaired and took some windings off to add some teflon foil) + 8.5H/2A + 2.3H/5A (as splatter choke just before the negative peak limiter, all with handmade spark gaps (3mm).
I will add to the mod transformer primaries spark gaps,too.
In this week I'll start modulate with the small Yamaha and if all ok I'll replace the big guy Phonic.
One last question about how to use in the rightest way the af amp is included in the attachment.
Thanks again mostly for your example about that 'difficult' night of you.
I will think ten times more but in my life theory, mistakes are the same welcome as success...
They pain but they make success more appreciable...
Mistakes in life is like suffer in love...we will allways remember and talk about them.
Anyway I decided to go the uphill way to am, high level modulation, I like it and enjoy it even sparking!
I hope you keep 15 cents for me in case I 'll loose all 

Thanks for being on the other side and help me with your experience.
Stefano

[AB2EZ]

Stefano

Agreed regarding the value of all types of life lessons (-:

Assuming your amplifier can deliver enough voltage, you want the audio source impedance seen by the modulated rf stage (looking back toward the modulation transformer) to be as low as possible. This will lead to the best low frequency response (since the amplitude of the audio applied to the rf stage will only start rolling off when the modulation impedance looking into the rf stage drops below the audio source impedance.

Also note that when a modern audio amplifier has a 4 Ohm output impedance rating... that doesn't mean that its Thevenin equivalent circuit is an ideal voltage source in series with a 4 Ohm resistor. What it means is that amplifier's output current overload circuit will "kick in" at a much lower output power level if you use a load that is significantly less than 4 Ohms.

Given the above, and given that your audio amplifier appears to have ample excess power capability (vs what you need to modulate your rf stage), I would be inclined to use the smallest step up ratio that gives you enough modulation voltage. I would start with the amplifier in bridged mode across the 240 Volt primary winding...and  using the 5000 volt secondary winding. That would give you a step up ratio of 21:1. The audio amplifier would see a load of 3571 Ohms/ (21 x 21) = 8 Ohms.   If you don't have enough audio output voltage, then try using the 120V primary winding and the 3500V secondary winding (with the amplifier still in bridged mode). That will give you a step up ratio of 29; and the audio amplifier will see a load of 3571 Ohms/ (29 x 29) = 4.25 Ohms. It would be nice to have something in between... but those are your choices with this transformer.

Stu

[AB2EZ]

Stefano

I deleted this post because I was making an arithmetic error. I also made some corrections to the post before this one.

Stu

[AMLOVER]

Stu,

I somehow feel that bridging is not a very safe mode even the productor of the amp says the oposite.
I'll go for the 8 Ohm (240v primary) to the 3571 Ohm (5000v secondary), this transformation will keep the modulation level low giving me the time to study the behaviour of the transmitter, later I''ll upgrade to the powerfull 4 Ohm.
I am also thinking to unkey the Hv from the anode after I have unkeyed the screen and before the driver.
I have a vacuum relay that I could use it after the Hv supply to interupt it before the driver is off so reactor's energy will have the time to go through the tube. Some guys here are connecting the vacuum relay in parallel with the reactor and they just short the reactor in receiving periods, they say that like this they have not the big energy back on the reactor when they key off. What is more safe cutting the HV or leaving it on the plate but shorting the reactor before the driver is off?
Another guy mentioned that turns his driver down slowly (1sec) by using the power adjustment potentiometer and he keeps all voltages on the tube all time, even slow operator, he never had a spark in his mod transformer or reactor.
I can make any of the three solutions but I don't know what seems the most efficient to start with.

Thanks
Stefano
 

[AB2EZ]

Stefano

If you are usinging tetrodes in the rf output stage, then reducing the screen voltage to zero, as a first step toward going from transmit to standby, is still (in my opinion) the safest way to shut down.

After that (2nd step), closing a relay to place a resistor (not a short circuit) across the Heising reactor would be a good thing to do. The resistor should have a value approximately equal to the modulation resistance of the rf output stage. That would keep a non-zero load on the audio amplifier and the transformer... and the reactor.

If you place a short circuit across the reactor, then your powerful audio amplifier might damage your modulation transformer (too much current).

You should disconnect the audio input to your audio amplifier when in standby... but do not disconnect the output of the audio amplifier from the primary of the modulation transformer... and do not turn off the audio amplifier.



Stu

[AMLOVER]

Stu

I use a tetrode so screen voltage is firstly reduced to 0v.
Your suggestion reminded me the protection resistor I used in parallel with the mod. transformer secondary when I didn't use the modified Heising except that it was 500 watts.
The audio chain will be first off, then the screen voltage from primary(I counted almost 0.5 secs to 0v), then the resistor parallel to the reactor and finally the driver together with the bias resistor out/full fixed bias into.
When going to the transmit I'll go exactly the oposite way.
I think that a 50w resistor would be enough as far there is no load when it is involved.
I also counted the total reactors' resistance to 86 ohms for a total no load 42H which will keep the dc a bit back for a higher mod percentage.
I'll always keep the af amp on and will not disconnect it from mod transformer primary as you suggest.
I'll set up all and come back with news.

Many many thanks,
Stefano

 

[AMLOVER]

Stu

I took away the limiter because I'll use the 3600 Ohm resistor for the reactor protection, limiter must wait.
I was thinking to keep the resistor always in the sircuit as far it is 600w and will not suffer during transmitting.
Is it a certain reason to switch the resistor off when transmit or it is not a problem to let it involved?
I am thinking to do as in the attachment hoping to get the advantages + simplicity.
If wrong I'll enter the vacuum relay to on/off it.

Thanks
Stefano

[AMLOVER]

Stu

I am very happy to tell you that with the resistor as you suggested, connected permanently without relay,
the on/off is perfectly safe without any sparking even without shuting down the screen voltage before.
I can turn on and off the full load with only the driver on/off switch. I regulated the fixed bias to -180v for both on and off and all seem quiet and relaxed. The resistor is absolutely cold so I'll try a much smaller one around 50w and return the giant one back to the limiter which is in the standby mode.
I can't believe yet that I'll not need any sequencer. Audio chain off, driver off...monster falls asleep and receiver starts work!!!
Now it's time to connect the af amp back to back with mod transf primary.
I'll do it tomorrow morning, now I go to have a beer with some friends to your health 

Thanks
Stefano

[AB2EZ]

Stefano

There is a problem with keeping the resistor across the reactor during transmit.

The audio power amplifier will be looking into the parallel combination of that resistor and the modulation resistance of the rf output stage.

That means that (roughly) half the audio (modulation) power delivered by the audio amplifier is going to be dissipated in that resistor

The resistor may not be hot now... but it will be really hot when you apply modulation (-:

Stu

[AMLOVER]

Stu

I am so relief with the on/off easyness that it will be a pleasure to switch out the resistor with the vacuum relay when transmit. I'll add the vacuum relay and test its reliability before add any audio on the sircuit.
After lanch, the "Philadelfia experiment" will be repeated in Athens-Greece this time... 

Thanks
Stefano