The AM Forum
March 29, 2024, 05:26:51 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 [2]   Go Down
  Print  
Author Topic: Simple diode negative peak limiter for line level audio?  (Read 26469 times)
0 Members and 1 Guest are viewing this topic.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #25 on: February 22, 2010, 03:19:14 PM »

Clark,

Yes, the screen dropping resistor works FB for a small rig, but with big HV, the wasted power adds up fast.  At 3KV on the 4X1 final, there wud be 2500V dropped across the resistor.  At 140ma screen current this is 350 watts of heat used up. It would take some big power resistors to handle that continuously. That's why the little 500V screen supply with the choke is used.

In addition, I would lose the ability to fine tune adjust  the screen current on the fly. I use the Variac in the 500V supply now. It helps to adjust screen current as antennas are changed, loading is changed and when HV is stepped up or down.

One last thing to consider:  If by error the HV is turned on before the filament and/or the rig is not keyed to draw the normal 140ma of screen current, there will be 3KV on the screen and associated screen circuitry thru that resistor. Yikes!

Though, weren't screen chokes with LV supplies used all the time in the BC industry? I agree the resistor is better than a choke with high shunt capacitance problems, but think the difference is very small if working correctly.

I'm curious - anyone else using a screen choke for a big rig - opinion?

BTW - I'm gonna try that screen neg peak limiter later today...

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W2XR
Contributing
Member
*
Offline Offline

Posts: 890



« Reply #26 on: February 22, 2010, 03:50:21 PM »

T,

Agreed.

I too am using a screen reactor in the screen circuit of my 2x 4-400A rig. I think a pair of 4-400As qualifies as a high-power rig. Wink The screen supply in my transmitter is Variac-controlled, so tuning the HPA to the resonant point is a breeze; just reduce the Esg to about 200 volts, and there is no fear of pulling excessive off-resonance plate or screen current. I've been doing it this way for over 30 years. It may not be the most optimum solution in a high-power plate modulated rig, but it is possibly the easiest and it certainly works well. If someone can offer a more elegant solution, aside from using triodes in the HPA stage, I am all ears.

I do not see how there could be a large difference either, in terms of audio quality (read final amplifier stage modulating more linearly) when comparing a screen reactor vs. a screen dropping resistor. Perhaps Clark/KE7TRP, can chime in here and describe the difference or improvements he noted with the screen dropping resistor approach.

As you indicated, the use of the screen dropping resistor would necessitate a resistor of significant dissipation capability, along with an attendent large waste of power. Dropping the Esg for final tuning I think would be more problematic as well; a high-voltage switch and additional series resistors to further drop the Ebb would be required, etc.

The screen reactor in my rig is actually 62 hys; I am using a 30 hy choke in series with a 32 hy modulation reactor out of a Gates BC-1G 1KW AM transmitter. The mod reactor is deliberately positioned closest to the 4-400A screen circuit, as this inductor is probably better in terms of lowest HF loss due to capacitance effects to ground, when compared to your average garden-variety filter choke in this application. For what it is worth, I cannot hear any real improvement with the 32 hy reactor in the circuit, so it may be going on the shelf soon, but I first do want to perform a proof-of-performance on the rig to quantitatively see if indeed there is any real improvement. The original 30 hy Kenyon screen choke I use is mounted on ceramic stand-offs to minimize the HF loss to ground, again due to potential capacitive loss issues, and worked great, with excellent low and high-frequency characteristics. I never did a swept frequency response of the rig with the big 32 hy mod reactor in the screen circuit; too lazy I guess. Without it, and with just the Kenyon 30 hy reactor in the circuit, it was about 35 hz to 12 khz at the -3dB end-points.

With the two reactors in series, the predicted -3dB roll-off in my rig at the normal value of screen voltage and current (i.e. the screen impedance) that the two 4-400As pull, is about 18 hz. This does not include the low-frequency roll-off imposed by the mod xfmr, driver xfmr, or mod choke, etc., but rather is the -3dB low-frequency point for the screen circuit only.

In retrospect, the two series reactors in my rig are gross overkill and most certainly not representative of good or practical design. The 32 hy mod reactor in the HPA screen circuit will probably come out sooner than later.

I agree with you; in a small plate modulated rig (<150 watts DC input), the screen dropping resistor is the way to go.

I do recall seeing, however, more than one AM BC rig that used a variant on the screen dropping resistor approach; perhaps it was RCA or Collins in their 4-400A series of 1000 watt-class transmitters. However, I believe this may have more of a design-to-cost issue, vs any potential improvement in the modulation characteristics of the transmitter.

73,

Bruce
Logged

Real transmitters are homebrewed with a ratchet wrench, and you have to stand up to tune them!

Arthur C. Clarke's Third Law: "Any sufficiently advanced technology is indistinguishable from magic".
Steve - WB3HUZ
Guest
« Reply #27 on: February 22, 2010, 03:54:04 PM »

Assuming you have a good choke (low shunt capacitance and sufficient inductance) there will be zero difference when compared to a resistor. Impedance is impedance. The modulating audio doesn't know the difference.  Cheesy
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #28 on: February 22, 2010, 05:12:51 PM »

I just tried the screen diode negative peak limiter. It had no effect as I increased the voltage from 0-150V or changed the series diode resistor from 400 ohms to 5k.  It was like it wasn't there.

The circuit:  cathode of diode to screen pin, anode of diode in series with 5K resistor, resistor to 0-150V + supply.  Supply neg to ground.

I felt the resistor getting warm, but no neg peak shunting effect on the scope looking at the modulated RF envelope.


** So... I have enuff HV diodes to build a 3-diode circuit for the plate circuit. Gonna give that a try.  I know it will work cuz I've done it many times b4.  Better to have some protection than none.


T

Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #29 on: February 22, 2010, 09:06:44 PM »

I put together the 3-diode negative peak limiter in the 4-1000A plate modulator circuit.  I used 6A diodes and made three 15kv legs.  The power resistor is 4K and I found about 50V was needed for the "keepalive" supply.  The negative band will will get as thick or thin as I need depending on the keepalive supply.

The scope pictures below show the intentionally overmodulated signal to show the negative band is slightly thicker, maybe negative -95%.  

It works FB as desired for a safety net.

T


* 4X1 Rig 560.jpg (329.17 KB, 1280x960 - viewed 487 times.)

* 4X1 Rig 549.jpg (316.47 KB, 1280x960 - viewed 508 times.)

* 4X1 Rig 575.jpg (326.46 KB, 1280x960 - viewed 478 times.)
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W2XR
Contributing
Member
*
Offline Offline

Posts: 890



« Reply #30 on: February 22, 2010, 09:16:28 PM »

I put together the 3-diode negative peak limiter in the plate modulator circuit.  I used 6A diodes and made three 15kv legs.  The power resistor is 4K and I found about 50V was needed for the "keepalive" supply.  The negative band will will get as thick or thin as I need depending on the keepalive supply.

The scope pictures below show the intentionally overmodulated signal to show the negative band is slightly thicker, maybe negative -95%.  

It works FB as desired for a safety net.

T

Hmmmmmmm..............Ok, I'm sold. T, you're the man!

I guess I'll do the same thing now, based upon your success. It appears the low-level route does not cut it, much to my dismay. What did you do differently this time around, that you did not do previously with regard to the 3-diode negative peak limiter?

And no audible artifacts in the modulation whatsoever?HuhHuh

Why the 6A diodes? Availability of the parts on-hand?

I have a bunch of the Amperex 15 KV @ 0.75A stacks I could use, and I'd use a 5K ohm loading resistor to simulate the class C HPA load impedance in my rig.

How about posting a quickie schematic for the great unwashed masses such as myself? I'm looking to implement this as the protection of last resort for my mod xfmr.

Thanks for the good work!

73,

Bruce
Logged

Real transmitters are homebrewed with a ratchet wrench, and you have to stand up to tune them!

Arthur C. Clarke's Third Law: "Any sufficiently advanced technology is indistinguishable from magic".
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #31 on: February 22, 2010, 09:16:58 PM »

Here's the 3-diode circuit as drawn by Steve, QIX.


Bruce:  I didn't spend much time testing it yet, just socked a few Yallos into the dummy load. I'll get some on-air reports for side crud tmw in the quiet day.

It does appear to work and lets the positive peaks continue higher as the neg band stays fat.

I used the 6A diodes cuz I have a bag of them. Your blocks will work FB, of course.

All in all, it's a great way to creep up on the audio while hitting the baseline regularly without generating the predictable trash of -100% negative. Plus the mod xfmr is protected with another layer of safety.

I don't expect to be slamming it regularly like the scope pics show, but will try it anyway and see what the reports show on the air out of curiousity.

T


* 3-diodeCircuit.jpg (77.87 KB, 592x417 - viewed 494 times.)
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W2XR
Contributing
Member
*
Offline Offline

Posts: 890



« Reply #32 on: February 22, 2010, 09:25:07 PM »

Tom,

I'm going to guess your success this time around was related to having the correct keep-alive voltage available. It appears that perhaps a greater percentage of voltage is required, at least in your transmitter, compared to what 'QIX had originally recommended.

If 50 VDC did the trick, what is your HPA plate voltage? I'm looking to scale this value for use in my rig.

Thanks,

Bruce
Logged

Real transmitters are homebrewed with a ratchet wrench, and you have to stand up to tune them!

Arthur C. Clarke's Third Law: "Any sufficiently advanced technology is indistinguishable from magic".
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #33 on: February 22, 2010, 09:37:20 PM »

Bruce,

Here's something about the crud level of this circuit as written by Steve/HUZ. His calculations show the crud to be minimal. Tests will tell.

I used 2200V for this particular test. (low tap)  I wud make a supply that will do ~0-150V to have a full range of adjustments. The neg band seems quite firm and you can set the keepalive so that the band is thin and forget about it. I wud set it to a minimum for the job.  The tests will show us.   ie, How little we can use and still not generate band side crud while keeping close to -95% - -98% neg..

T



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

"Although the zener regulation action will clip the negative audio peak near 100% negative peak modulation, the resulting distortion is minimal. Remember, at the instant the clipping occurs, the voltage on the final tube(s) is only about 5% the normal B+ voltage. This also means the instantaneous final plate current is also only about 5% of normal (or what you read on the transmitters meter which reads the average value). Knowing power is the product of voltage and current (P = EI), then the input power to the final at the instant the negative peak clipping occurs is only 0.25% normal. Thus, the distortion products created by the clipping will be no worse the 0.25% of the input power or unnoticeable. For example, if you are running 100 watts average input, the power input at the instant the negative peak clipping occurs is just 0.25 watts!
  With the circuit in place, the negative peak modulation is absolutely limited to slightly less than 100%. If your rig has sufficient modulator power, you can crank up the audio gain some, achieve higher levels of positive peak modulation, and never produce splatter. Remember, splatter is produced by modulation in excess of 100% on the negative peaks, not the positive peaks. Modulating over 100% on positive peaks will make you audio sound louder at the receive end or give you more "talk power".
  The circuit also provides protection for the modulation transformer. A big killer of modulation transformers is large transients created by modulation in excess of 100% on negative peaks. Since the zener circuit prevents negative peak modulation from exceeding 100%, no damaging transients are produced. Long live your modulation transformer!"
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #34 on: February 23, 2010, 10:15:23 AM »

Tom, Bruce,
You may have had problems in the  past if the keep alive voltage could not supply enough current to the NPL when the mod transformer went beyond the negative peak setting. The negative cycle loading draws current off the keep alive supply. If that voltage pulls too low you will hit baseline. I suggest a big cap on the keep alive input to handle transients. I had the same problem in my PDM rig during over modulation. fc
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #35 on: February 23, 2010, 10:59:52 AM »

Referring to Steve's article on NPL above...

He mentioned a 100w transmitter's negative peak crud is 0.25 watts when limiting at about -95%.  This would be down maybe -27db or so? (100w > .25w)   If someone is S9 +40 over, then the NEW side crud generated by the NPL would be S9+ 13 over.  Of course this is better than the crud generated by a -100% negative transmitter, but not as clean as one that uses no NPL at all and naturally stays above -95%.

I ran some tests last night into the dummy load using the local receiver as a spec analyzer and think I could pour on more audio without the bad effects of -100%, but will need to use on-air reports to be sure.

At least the NPL appears transparent when below the threshold use level, acording to crud measurments with it on or totally out of the circuit.  So far I'm pleased with it and will make up a permament ~50V NPL supply.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
W2XR
Contributing
Member
*
Offline Offline

Posts: 890



« Reply #36 on: February 23, 2010, 11:28:41 AM »

Tom, Bruce,
You may have had problems in the  past if the keep alive voltage could not supply enough current to the NPL when the mod transformer went beyond the negative peak setting. The negative cycle loading draws current off the keep alive supply. If that voltage pulls too low you will hit baseline. I suggest a big cap on the keep alive input to handle transients. I had the same problem in my PDM rig during over modulation. fc

Hi Frank,

Many thanks for the input.

You make a very valid point here. How big of a cap do you suggest?

As I indicated to Tom earlier, the biggest problem I had was finding the real estate to mount this thing. Playing a little hookie from work this morning, I think I found a good location right over the mod reactor. I really just have to get one HV rectifier stack and then start the mechanical layout, etc. I think I have all of the other parts.

A very simple project with hopefully good protective benefits, etc. I'm really looking to protect the mod xfmr in the rig more than anything else, and certainly not to increase the density of the modulation.

73,

Bruce

Logged

Real transmitters are homebrewed with a ratchet wrench, and you have to stand up to tune them!

Arthur C. Clarke's Third Law: "Any sufficiently advanced technology is indistinguishable from magic".
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #37 on: February 23, 2010, 12:27:46 PM »

I=c dv/dt

Since you are charging the cap with 120 Hz pulses it has to carry you through a  charge cycle. This means it needs to hold the voltage stable for 8.3 ms. Say you swing the voltage 500 volts negative across the negative cycle load. Now calculate the current through the resistor. The keep alive needs to hold the voltage so the final stays lit.
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #38 on: February 23, 2010, 01:19:33 PM »

In looking at this thread, I seem to remember a 'log speech processor' from back in the 80s marketed by a kit mfg. that did just what people are looking for, create asymmetry in the audio chain and block negative going peaks.

I remember building one and taking it into electronics class on a (also homebrew) AM transmitter.  My electronics teacher, Dan White (KB6TDW, SK), was AMAZED at the pic on the o-scope.  Where was the negative mod?

I remember the name now, it was CBC Intl. that did the kits.  Worked quite well, although I don't think they where anywhere near hi fi.  A friend built one last year, I just got it in the mail to play with.  If anyone's interested, I can do some before and after's on a scope when I get "round toit".

--Shane
 
modified to correct SK callsign
Logged
ke7trp
Contributing
Member
*
Offline Offline

Posts: 3659



« Reply #39 on: February 23, 2010, 01:22:47 PM »

Good job on the limiter.  

I know you will have 350watts of heat.  However,  I ran back to back tests with the choke and the Dropping resistor. The difference in audio quality and sound was huge.  I dont know what else to say. Sometimes people get stuck doing things one way and then defend that solution.  I have an ultra modulation unit that contains three 866s.  That unit is designed for BC use at 1000 watts Am transmitter. In the instructions it tells you to convert the screen to a Dropper and to NOT use the choke or Fixed Screen.  We converted it and where shocked at the results.  I think its because the screen and plate where not in phase. With the choke. There was clear distortion. Not bad but you could hear it. With the Dropper, This cleaned up and flat out talked.

There are obvious dangers here..  You decide if you want those pitfalls.  This might not be the best for Fabio. But now you have a neg peak limiter on the Plate only.  This makes the screen and plate not modulated the same.

I am in the process of changing the T3 over to modulated Screen.  I have studied the schematic and this seems very simple to me since the T3 has a Dopper on the NON modulated B+ now.  I have a feeling when I talk to you next, You are going to notice it Smiley

Clark
Logged
W3RSW
Contributing
Member
*
Offline Offline

Posts: 3308


Rick & "Roosevelt"


« Reply #40 on: February 23, 2010, 01:44:53 PM »

Something else is going on.  Time constant of L & screen C vs. R & screen C?
-shouldn't be that big a difference.  What is the value of your screen bypas caps.?

I don't detect a whole lot of audio loss using 20 H screen reactor and 780 pf/tube screen capacitors on my 813 rig (screen fed w/seperate 350v. supply.)

and,
Quote
Assuming you have a good choke (low shunt capacitance and sufficient inductance) there will be zero difference when compared to a resistor. Impedance is impedance. The modulating audio doesn't know the difference. 

Right.
After all, it is just audio, not 60 MHz RF.
Logged

RICK  *W3RSW*
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #41 on: February 23, 2010, 02:35:42 PM »

I just wanted to add the keep alive needs to be stable or it will appear as negative modulation at a phase lag which will be distortion.
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #42 on: February 23, 2010, 02:49:59 PM »

I just wanted to add the keep alive needs to be stable or it will appear as negative modulation at a phase lag which will be distortion.

Frank,

How would you measure this.... just use a scope and look at the keep-alive DC component for movement under modulation?  Or, are more elaborate tests required?

I just ran a Fabio RF dummy load test with high tap HV on the modulators and final. Gawd, those modulators are capable of some clean and serious audio when the negative peaks are limited. I never ran them that hard cuz of fear of the mod xfmr breaking.  I still plan to keep things naturally near neg -95% despite the NPL, but it's good to know the headroom is there for best linearity.  Still need to do some live on-air side crud tests with the new NPL kicking in.


BTW, just received a pair of new Tron-recommended Sony 7506 headphones today. What a difference over the crap I had before. The listening low end is huge now. No wonder I had trouble EQing my own audio in the mod monitor before.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
WA1GFZ
Member

Offline Offline

Posts: 11152



« Reply #43 on: February 23, 2010, 03:21:05 PM »

Tom,
Yup just monitor the keep alive voltage to see that it is stable. Any change in voltage will appear as modulation superimposed on the B+. Ripple also needs to be as low as possible. At some point the mod transformer will saturate limiting the peaks. Also keep an eye on the load resistor temperature.
Here is a good test for the transformer. Measure the resistance of each winding cold. Then run the crap out of it for a while. Shut it down and measure the resistance again. Knowing those values you can determine the power in the windings and the real temperature of the wire.
Logged
Pages: 1 [2]   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 0.075 seconds with 18 queries.