The AM Forum

In the Flex Community forum (link below), someone started a discussion about Reduced Carrier AM which is mentioned in the 6500’s manual. It states, “By reducing the AM Carrier level, the percentage of carrier relative to total PEP can be reduced to increase talk power in the AM sidebands”. I’ve read through the discussion and I just can’t grasp what’s going on. Is Flex saying that by slightly reducing the resting carrier level (which is normally a 4:1 ratio) you can increase the modulated signal in cases where it’s not achieving 100%? I’m not embarrassed to admit that I don’t understand this concept.
 
https://community.flexradio.com/flexradio/topics/reduced-carrier-am?utm_source=notification&utm_medium=email&utm_campaign=new_topic&utm_content=topic_link

As far as I am aware the carrier can be run at ANY level and modulated up to the maximum voltage of the system.

However with simple 'Diode' demodulation going beyond a certain % (150%?) will cause distortion on reception whereas using a synchronous demodulator would not.

This is my humble understanding, I'm sure someone can provide more detailed knowledge.

JB.

I used it on my Flex 6300 as the gain was too low on the Accessory port to fully modulate at 100% on peaks. Reducing the carrier maintained the sideband energy available and afforded full modulation. My new Flex 6500 is a bit more refined in the audio chain and I have a balanced input with plenty of gain to easily over-modulate if I so desire. It's algorithmic in that the processing is done before the PA stage, so reducing the output level does not have the same effect.

W7NGA  dan
San Juan Island, Wa.

You can not always get to 4x peak power and that depends on your voice and phasing as well.
The carrier control allows you to adjust the carrier level separate from the peak power.
Not sure how it works with various software that is out now, it used to be in the flex world that 100 on the carrier was full carrier power, 1/4 pep.
Then they changed it to 50 for full power carrier or something like that.
The ANAN software has it back to 100, but my radio will never hit 4x carrier power at 100, it does so around 70 (on most bands).

If you put the radios into an amplifier, you can make it a bit easier on the amp to have say a 200 watt carrier and 1200 watts pep. 1200 watts would normally be about a 300 watt carrier.

I can do the same thing somewhat with my screen modulators, turn the carrier down but keep the screen voltage swing and get 1400 watts pep with a 200 watt carrier.
In the case of the screen mod, that would over modulate in the negative direction, an sdr prevents that.
Not to say 200% modulation is going to sound good in most receivers...

WBCQ uses reduced carrier AM on their 9330 and 5110 and another 17 mhz TX and it is a challenge for some receivers to demodulate the carrier. I do not know the technical advantage of this process.
WBCQ, 7490 is still a full carrier shortwave service of 50KW.

Here's a short article from the AMfone archives about the subject.

http://amfone.net/ECSound/k1jj22.html

The percent of modulation you run your AM transmitter at will depend much on your own preference and also on what kind of receiver detector the receiving station is using.  

I would guess that 1/4 to 1/3  of listeners are using an SDR sync detector so that we can run any modulation level we wish, even 300% positive and still sound clean to them.

But the majority use older diode detectors which limits distortion-free listening to maybe 120% modulation at most.

I personally adjust my carrier level either through plate loading and/or drive to a linear amp so that I can achieve a natural 140% positive peak level that lets my own asymmetrical voice fill it out where it wants to go.  It's all a compromise. If too many people complain about us sounding like an ssb station, especially during fades, then our modulation level is set too high.

BTW, our voices can change asymmetry levels (and phase at certain frequencies) from day to day or year to year, especially from early morning to night.  It all depends on what our vocal chords are doing. They can be sore or relaxed from use, a cold, or whatever.  So don't panic if you set all the audio gear perfectly and the next day things look funky.  I've seen my own voice change from as low as 110% positive to as high as 140% positive over a week's time. Waking up with some mucus congestion in the throat can get us reaching for the EQ low cut in a hurry.  Lows can sometimes dominate the asymmetrical effect and are much affected by the present condition of vocal chords.

T

There are people that sound good (mellow and relaxing) with a strong carrier and 60% modulation, and people who sound loud with a lot of audio on the carrier.

Its nice to be able to set it however you want.
With the new ANAN radios, the compressors work well and even include look ahead so you do not spike amps and hit the baseline in the negative direction.
Unlike the old software, it seems to work very well.

As an operator, its nice to not have to pay any attention to the audio level, set and forget and you will not be too low or over modulate over a wide input range.
What will they think of next!

At some point we need to differentiate between "normally generated" AM and that where the carrier is, due to the process, eliminated.  An example of the latter is during the use of a double balanced modulator the carrier is eliminated and only the sidebands appear - in a perfect world or a SDR radio.  The carrier can be reinserted by one way or another so that the individuals at the other end can properly demodulate the DSB modulation / reinserted carrier.  Assuming the DSB generation  occurs without distortion products resulting from several possibilities during the modulation: mixing & amplification process, there is no longer an assumed carrier to sideband relationship (ie: the fabled 4:1 SB / carrier relationship).

If one is to be received with little or no "distortion," it is important to establish a proper relationship between the DSB component and the reinserted carrier.  Otherwise, be prepared to receive "overmodulation" or "distortion" reports.  This problem can be dealt with using various enhanced carrier detection (ECD) methods at the receive end.  The most obvious ECD method is the use of the product detector that is normally used during SSB detection - a less than desirable method.  Another method - the one most satisfactory to all of us is to use Sync  AM detection.  This approach supplies a phase coherent enhanced carrier that is used to detect an AM signal. Yet another method is to route the received carrier, filter it, amplify it and then reinsert it into the detector.  Most of us are not familiar with this method - it has its limitations (notably not being phase coherent with the modulation & preexistent carrier components) and will probably never be a good solution for this form of "super modulation" where the carrier is below what would be the expected carrier level that can be used for older AM detectors.

I think JJ has the best philosophical approach: be considerate of the folks at the other end and limit your positive peaks to around (Tom's value) of 140%.  My class E system produces more than that positive modulation peak value but that's my voice.  I think I have received around 2 "distortion" reports while I have put the class E transmitter on the air.

Conclusion: be considerate while adjusting your DSB / carrier reinserted rig and try to keep the positive peaks reasonable so that those of us who use the old gear so that we won't want to twist the frequency dial.  Please do this with some sort of test equipment so that a QSO won't turn into an "audio clinic" where the poor guy at the receive end has to listen to a bunch of "fives, yayyys or yayllowws" while we dial it in at the transmit end.

Cheers, Al

The understanding I have from Timtron on WBCQ is that they reduce the carrier by 6dB.
Using an SDR in the SAM mode loses sync trying to receive their reduced carrier stations. An R390 didn't seem bothered by the reduced carrier.
Supposedly reduces operating costs of making a full carrier station and putting more modulation into the sidebands. I understand that part of the equation.
Supressed Carrier is for SSB.
What has been mentioned here so far from ops reducing their carrier from their Anan or other SDR Transceiver is not the same as a reduced carrier AM transmitter used in commercial services. Maybe Timmy might chime in and give a more sensible response than mine.
Fred

The basic difference between the two methods described above is the refinement of the process in the case of commercial broadcasting.  We can implement the same refinement - we have the technology

Al

I think it is the same, just done in a different way, but a signal that has the carrier reduced by 6db is a signal that has its carrier reduced by 6 db no matter how its done I would think.
In the sdr's, I think, if you turn the carrier all they down it just runs DSB.

Apples and oranges,  to a point.

Big broadcash and shortwave use controlled carrier,  up to 6 db. I read somewhere that is the limit that doesn't mess with diode detectors.

My assymetry generator just let's me dial in positive peak stretching.   Ie,  all the way down,  I'm at 100 pct pos and neg.   Turn it all the way up,  still 100 pct neg,  but can hit 420 pct pos peaks.

My assymetry generator only increases the pos peaks.   

That doesn't mean the series or pwm (use both,  depending on which TX I'm using)  don't sit generate a DC offset.   But,  looking at the output of the peak stretcher,  there is no DC offset generated there.   

--Shane
KD6VXI

It will not increase talk power(1) or any other kind of power. The Flex has a maximum PEP capability. You can utilize this PEP capability by running the carrier at about 1/4 the max PEP and modulating 100% or you can run less carrier and modulate more the 100% on positive peaks. In any case, you will achieve the same exact output on the receiver at the far end. There is no free lunch.

(1)  Always be wary when you hear someone use a term like talk power. It's nonsense.

What does full power with only downward modulation sound like?
Rough on the final tube I suppose...

I believe that its 6db below the peak power, which is 50 kw, and since -6db would be 1/4, that's the same 4 to 1 ratio as 100% modulation of a full AM signal, but those transmitters at WBCQ are also only running the upper sideband. Lower sideband is suppressed, you can receive it decently on a receiver on USB, when I do this I switch to LSB to zero beat the carrier (no sound so it is easier to hear when you zero beat), then switch to USB to receive the audio.


I can also do something like that with my icom 718 (dsb AM, not single sideband), using the external alc controller. With the radio's menu power set to the highest, use the controller to set the carrier, I can turn the thing all the way down to like 10 watts of carrier and be pushing almost 150 watts pep. The techincal term for it would be double sideband reduced carrier, I guess.

Some of the old tube datasheets always had notes about using "modulation essentially negative" or something like that. I tried it once with my icom briefly into a dummy load, it didn't sound bad. Actually, I think it would actually be easier on the finals (in something like a linear) than normal AM would. You would be running the full output at a resting carrier, but at something more like 60-80% efficiency, then with modulation, the tube would probably start running slightly cooler due to the actual power it is putting out being somewhat lower.

Downward modulation in a linear amp is from improper tuning of the amp to achieve full power PEP in the CW mode from the exciter. When properly loaded and tuned the exciter is set for an AM carrier set for 25% of the PEP carrier carrier and the magic happens. Downward modulation in a linear is severe flat topping of the audio peaks and obvious distortion on the air.
Downward modulation in a table top AM  TX is usually too much or too little grid drive from the RF driver to the control grid of the finals. There seemed to be a fix discovered for the venerable DX100 which had a little too much drive causing non-linear response using the  X-Y inputs of a scope viewing linearity of the final by observing the trapezoid pattern on the scope. Perfect straight lines of the modulation. The bright tip of the trapezoid is the negative peak the opening to the ends of the pattern are the pos peaks

Fred

6db of carrier compression means between words,  the transmitter drops to a quarter of the peak power (100 pct Symmetrical mod) AND the carrier drops another 6 db.

So a 50 kw transmitter would have a 50k carrier with no carrier reduction scheme.   With carrier reduction dialed in,  the carrier would drop another 6db, to 12.5 kw.

This is done in the digital transmitters....   So now your talking maybe 18kw input  at carrier,  vs nearly 80 kw on the old style xmitter and 60 kw for a typical digital...

Some others have digital algorithms to change the carrier compression relative to the input.   IE,  if you have a 50 pct mod signal,  the  the carrier compression dials itself down,  actually increasing the carrier power to ensure max coverage.   I believe that's a brown boveri (bbc)  design,  but I could be wrong.

So,  to sum it up,  the difference between carrier and full power modulation can be 12 db.   At 100 pct modulation.

--Shane
KD6VXI

I would think, if you wanted (mostly) downward modulation, you would need to run the rf deck as a linear to reproduce the modulation cleanly. If only in the downward direction.

I have seen people on the air who had very little or no positive modulation, and sometimes they do not sound too bad.
That was surely because something was wrong, but if you WANTED to modulate that way, you might be able to clean it up some.
My 32V transmitters did not like to modulate a lot in the positive direction till I did the screen voltage mod.  
Phase it wrong and you could have little positive modulation with plenty on negative modulation...

Modern versions of the old controlled carrier?

I know for a fact that a lot of stock riceboxes do that, they have no trouble making 100% negative but barely do 80% positive, even with the carrier set to 1/4 the pep. Now if you turn it up the whole way, you'll be putting out full pep power and not have any headroom for positive peaks, only negatives. I tried this as an experiment with my 718 once, one time at full power and a second time with the carrier at 1/2 the pep of the transmitter (so that it only did about 50% positive peaks). The audio wasn't too bad, there did seem to be a slight amount of distortion when running 100% negative and no positive modulation, but it wasn't nearly as bad as what a highly modulated signal sounds like on a diode detector at like 150-200% positive. if you had a linear tuned up to put out full power and fed a signal like that into it, it would put out its full pep at resting carrier and then should modulate 100% negative very nicely. The output components would be under a lot of stress with the full power carrier on them, but the tube would probably be running a lot more efficiently. if I had a rig that could handle running like that I'd give it a try, but I don't think there is sone unless you built it yourself. What I would really be curious about is the different in received signal between say a 375 watt carrier modulated 100% to make 1500 watts pep, and a 1500 watt carrier modulated 100% negative only (no postive modulation).

Brett,

Pretty much,  yes.

The BBC method is slick.   Under a certain pct of mod,  they actually increase the carrier.   This gives more range.   As the density goes up,   amount of carrier being rolled back between peaks changes.

There method actually increases range while decreasing cost.

The normal method is DX 60 style controlled carrier.

--Shane
KD6VXI

I think you could play with circuit values or even make it adjustable.
I have heard people running the controlled carrier and it did not sound too bad, and its easy on the amplifier if you use one.
I think it gets bad when you have almost no carrier when not speaking, but at something like 50% you almost need a scope to tell.

This is kind of what I've been thinking, but due to my lack of a clear understanding I decided to introduce the topic and use it as a learning experience.

So last night I tried an experiment.
I currently use my 6500 to drive an ACOM 1000. With the AM Carrier set to the default 1:4 ratio I adjust the unmodulated AM output of my 6500 to produce 200W out of the ACOM 1000. This produces ~ 800W out on peaks.

Next, while applying an unmodulated signal, I reduced the AM carrier setting to decrease the ACOM’s output to 100W.I made a few contacts and nothing negative was noted. To stations where I was weak, no differences were noted when comparing normal operation to the reduced carrier. HOWEVER, I definitely noted a decrease in the ACOM’s air exhaust temperature. If for nothing else, AM Reduced Carrier can probably be used to reduce power consumption and amplifier wear and tear. I'm going to do more testing to see how much I can reduce it. I don't have a mod scope, so I am very curious how this will all effect the modulation. Audibly, it sounded fine.

As the 'guilty party' who started off the thread in the Flexradio forum, I'd like to thank everybody here for their contributions.

Mni tnx de Guy G4DWV/4X1LT

I learned a lot from this Guy!
Thanks

Going from 200 to 100 watts carrier (even in a normal signal) would likely not show up much at the far end.
When testing, try reducing the overall power from 200/800 to 100/400 and see what the difference is, then do 100/800 and see.

I was always amazed how things seem to jump up from about the 25 watt level, power below that does not work very well, and someplace between 25 and 100 watts is a big improvement, then from 100 watts to 500 watts seems to do little.

In the new flex software, how is the carrier setting done, 0-100, 0-50 or what?

So last night's testing was the 100/800. I'm thinking to try 50/800 tonight.
Without looking, I'm not sure what the numerical scale is. At the max position (of the AM Carrier adjustment) is where you get 25 / 100 or the 1:4 ratio. I start there and adjust the transmit output to get my 200W. From there I just back the AM Carrier slider down, to get 100W (unmodulated) out of the amp. That provides a 100/800 end result.

BTW, I see you gave up on the Anan - that was fast!

I still hang on the thought that the carrier is reduced within the apparatus of a commercially built transmitter.
Here is an example of what transmitters WBCQ is using with this REDUCED CARRIER capabilities

http://www.transmitter.be/tmc-gpt40k.html

What you guys are doing is not connecting to a TRUE reduced carrier operation, for whatever reason you are trying to implement this in your SDR transceiver. Reducing the drive of the SDR in AM only reduces the power out of the linear and ends up as a full carrier with a lot more AM modulation on the distant receiver.
Sorry for some attitude in this reply. A little bump in the road in the hospital today. It's minor; I'll be ok
Now looking at a reduced carrier on an SDR panascope will really show you the really wimpy reduced carrier and how easily it unlocks when you try to run SAM.
What I remember seeing on WBCQ on their other services is the reduced carrier usually running in USB mode. A plain regular AM shortwave radio will receive this signal and detect it as AM because there is some sort of a carrier there.
This method is a power saving shortcut for a shortwave broadcaster and is allowed by the FCC.
WBCQ, 7415 is using a FULL carrier mode of a similar GPT 40K TX. I do not know what happened to the Harris 50KW MW50.

Thanks

Fred

You'll find even more efficiency if you roll the bias up (increase it to more negative)  and increase the drive to compensate.

For instance,  a 500z at 3.2 kv needs about 5 to ten volts of bias.   In am,  I run 20 or more volts of bias.

My tubes run culo,  and nobody complains.

I do NOT run this way on ssb.   Needs to be switchable.

--Shane
KD6VXI

I think the net result is what counts, not how you do it.
In the TMC transmitter, they likely do it all at low level with a balanced modulator they can unbalance and get any amount of carrier out of it.
I do not see how that is different from an SDR where the same thing is done in the digital relm.
There are three separate controls, overall power output, audio level, and carrier level.
The power control changes carrier and audio, the carrier only does carrier, and the audio only does audio.

In a screen modulator, you can have a carrier control and an audio control, as you can in plate modulation.

What might be different is what happens in the negative direction, plate and screen modulation close the carrier off in basic form (no negative limiting).
In SDR, before it gets to 100% negative it starts going the other way, it inverts it and starts going positive again, but they can limit or compress the audio in only the negative direction.

I have a problem picturing what the TMC rig does in reduced carrier when you have plenty of audio to give 40KW peak on a 5000 watt carrier.
It brings the carrier up to 20KW only during the peaks?
What is the time aspect of the AGC sort of gain control look like, and how does it impact the sound?
 

The "digital Realm" is only in the software......once it exists the computer into the SDR transceiver its back to analog transmitted the old fashioned way. We are not transmitting digital data over the air with SDR. The latest greatest Anan's and and their kind grab the analog RF into their magic FPGA (on the receive side) processors and do the digital thing in your "digital realm' and spits it out of the computer as an analog audio signal. All magic processing, for the ultimate dream receiver is obtained through the magic of computers. Specs for receivers that the best analog receiver ever conceived tube or solid state could NEVER compete with.

Fred

As flintstone said, it is done fairly easily using software. Here is a summary:

Audio input to most station transmitters today is usually in the form of a digital audio stream,

If the audio coming in is balanced analog, it is first converted to digital using A-to-D converters and stored,

The digital data is then sent to a processing unit which does a number of things:

1. messaging of audio if necessary, such as predistortion for type of modulation (PDM, digital PA's, etx.),

2. look ahead processing to determine what carrier level and modulation is needed when the modulation envelope is created

3. digital data is released to control carrier and modulation, all happening at a high-Megahertz rate.

Phil - AC0OB

The TMC units are very far from any type of computer control. These are 51 yr old retired military units!!!!!!!!!!!
Possible date of manufacture around 1964!!!!!!
I never mentioned use of software used in the TMC transmitters


Fred

The method I described is exactly how Brown Boveri Co.  implements reduced carrier AM.

Now,  it's mostly done in software.   What I described wasn't, the tx's I studied where pwm,  analog input.   Circa mid to late 80s.

Nail on the head time,  they mostly all accomplish the same thing:  Saves broadcash and shortwave tens to hundreds of thousands of dollars in utility bills.....   And allows hams to use amps like the sb220.

--Shane
KD6VXI

It's called Modulation Dependent Carrier Level (MDCL). There are two basic approached, Amplitude Modulation Companding (AMC) and Dynamic Carrier Systems (DCS).

In AMC the carrier and sideband power are reduced with modulation. With no or low levels of modulation the transmitter output is not reduced at all. Above a certain level, the power is reduced in a linear fashion relative or the level of modulation. At 100 %, maximum reduction occurs (usually 3 or 6 dB).

With DCS, only the carrier power is reduced and it is reduced the max amount generally when there is no modulation. As modulation increases, carrier power increases.

Fred, I was referring to your reply #32  in an attempt to answer Brett's question. ;)

Phil - AC0OB

So last night I did some testing with N1BCG.

I started out with 200W of carrier coming out of the amp. At peaks I was getting about 800W.
Using the AM Carrier control on the 6500 I reduced the carrier to 50W, while still achieving the 800W peaks.
Amp runs much cooler with no apparent degradation. Need to repeat this test with someone who has a mod scope.
Seems like an easy way to reduce power consumption and heat dissipation.

Thank you Phil for your kind re-direction.
Sometimes I get a little rattled sitting in this hospital dealing with another cancer. Clarity of mind and reading responses gets scrambled and I get stupid

Fred

Fred. Good luck in your battle. I hope you are out and about soon and fighting fit.
I recently had a longish stay in hospital and did not take any toys in (phone, books or tablet) to bust the bordom as I did not know if there were any lockable facilities available (there were not). You are lucky that you did not spend all your waking hours staring at the walls ;).
With vy vy best 73 de Guy G4DWV/4X1LT

No problem Fred.

Keep up the good fight and here's hoping for your quick return to Modulatin' in the AM mode. ;D

Phil - AC0OB

What was he using for a receiver, and how was it set up?
An SDR in sync mode and I would expect that.
An R390a maybe not.

I would think 100 watts carrier and 800 watts pep would be ok in most receivers and save an amplifier some work.

I don't think Clark is using any SDR equipment. If I hear you on this weekend we can do the test again.

Here is Nautel's presentation on MDCL if anyone is still interested in how one company implements this for broadcashers.

About 40 minutes long and would recommend skipping the first two minutes.


http://www.nautel.com/webinar/am-power-saving-techniques/

Phil - AC0OB