150% Positive Peaks

[w1vtp]

I did a spreadsheet on the carrier level vs sideband products.  Here's the JPG plus the worksheet.  I used 5 and 12 KHz delta frequency as sideband products.

[ka1tdq]

Jim, WD5JKO... My on-air schedule is really sporadic. Infant,new baby on the way, job, etc.  It would be better if we could email or text each other maybe a half hour before hand. My phone has email capability. My email address is navy.gov@gmail.com. Let me know the best way to contact you for a sched.

By the way, I went barefoot this morning talking to Tucson. My carrier was 28 watts with peaks hitting 140 watts. That's 156%. Everything sounded fine to me as well as the guy I was talking to who was using an SX-111 receiver. 

So, full speed ahead!

And Jim, I'll leave the linear on for ya.

Jon
KA1TDQ

[VE3AJM]

Those running high power should be more aware of what they are doing whether its class E or plate modulation. These stations tend to be heard out farther away.

I hear this all the time from guys in the Midwest and I hear this all the time. Example: existing AM QSO in Michigan on 3880 running 100 watt class txs, the band is open at 7pm, and the EC group on 3873 fires up 10 mins later running 500-1000w, and has at it. The usual..oh I can hear a weak QSO in there on 3880 is put out there, and it continues almost daily. Who cares about the guys on 3880 or other adjacent QSOs. Their QSO is now being trashed. Sure, get a better receiver...really...??  A red herring? No...its reality...but lets not talk about it..

Someone's operating practices need to be overhauled and re-examined.

Al VE3AJM

[steve_qix]

Yeah, the 3880 thing has been problematic since I've been around.  I'd be up on 3885 talking with the east coast guys, and a QSO would fire up on 3880 from out in Ohio.  The 5kHz hetrodyne is deafening, and of course the sidebands were right there overlapping!  Nothing can be done about this except to not do that (use frequencies that are too close together).

As a rule, I avoid using 3880 if I can because it will cause interference (or be subject to interference) from/to 3875 3885 and 3875, which are well used frequencies.  Yeah, yeah, we can go anywhere, yeah, yeah yeah.... I know that.

This is not a bandwidth problem, this is a "too close for comfort" problem.  The group currently on 3873 moved down from 3875 to get further from 3880.  They would have gone to 3870, but there is a regular group of sidebanders who use 3868, and well you know the scene..

But, I would make a prediction that IF the 3873 group moved to 3870, SOMEONE would try to have a QSO on 3875, and then problem proceed to complain about the "wide" stations on 3870.

If you have a good signal, you can't win sometimes.

IMHO, 3880 is not a good choice because in theory, a 3880 QSO pretty much precludes anyone else using the so-called "AM window" (3885 or 387x) without causing interference or being subject to interference.

It's an OLD OLD problem - goes back as far as I do in this hobby which is over 40 years.

"Doc, it hurts when I do this"  Doc:  "Don't do that!"   

[VE3AJM]

I would say that with the example that I gave, it was a transmit bandwidth problem, whether the cause was overmodulation and/or the 25+ kc audio frequency response from the 73 people. At that time, even 3885 was being trashed. So the entire "AM Window" was unusable. The 3880 QSO was pre-existing. But then, we really shouldn't be using 3880?? and the 3873 group had already done the right thing and moved from 3875 just to help out? Aw..ain't that nice 

What is the purpose of having a transmitter with a 25 or 30 kc audio bandwidth and having the Ss and Ps, and frictive sibilance being heard out so far? We aren't transmitting music. Why do you want music response? Most guys will use the typical vintage receiver with a 6-8kc receive bandwidth, some a little more some a little less. Seems like a lot of wasted energy and power used unnecessarily, that's only going to tick off a whole lot of folks.

I moved down the band years ago to get away from all that. I see that the attitudes haven't changed much.

Al VE3AJM

[w1vtp]

Shelby, Budley

I particularly like your comments about being careful about our measurements.  We need to realize that it's the rx bw and the tx bw that needs to be factored in while determining one's transmitted bandwidth.  The other thing to note is that if an AM station is very strong and we try to QSO 5 KHz away, we are going to hear the sideband products that may be >-30 dBc.  If we are trying to listen to a station with only a few dB signal level above those sideband products from the strong station, we are going to have an interference problem (interference as defined where two signals are detected at the same time).

If we look at that spreadsheet I created and assuming a correlation between the 32 dB / S9 and the screen, and then factor in the receiver bandwidth there will be interference with the weaker station between at 5, 10 KHz carrier spacing.

Steve

AM carrier spacing:  5 KHz is just too close.  Looking at the numbers in the spreadsheet one would have to have a greater than the -70 dBm (3 dB / S9) and then tune off to one side away from the strong station in order to hear reasonably well the station that is being listened to.

Jon

150% positive modulation level?  Yes!  It is very possible with the human voice to achieve these levels with the negative modulation levels of <95%.  That is especially so today with common knowledge of modulation asymmetry coming from the human voice.  

Correct modulation phasing: Decades ago using the correct phase in one's modulation chain was not a common experience - So I think we need to address this reality by using correct measurement methods.  We can probably look for the "pearls" on the negative side of our modulation levels but accurately determining the positive modulation levels is another matter.  One needs to have a good means of measuring both the positive and the negative modulation levels in order to guard against flat topping on the positive peaks. It is very easy to be aware that something is wrong when either asymmetry is the wrong way or there is none.

Finally, we need to measure the distortion products of our AM transmitters as part of our AM experience - that should be our goal of applying good engineering practice of our amateur radio station.  
Yes, with today's advance in technology, we need to know that our stations are putting out reasonably low distortion products and have control over our transmitted bandwidth.

One more thing: Audio bandwidth: We need to have control over our transmitted audio bandwidth - we need to know what it is and have means to control it. I am not affixing a value to this - I think we need to know what it is and be able to control it.  Another goal in good engineering practice.

Another good thread might be the proper demodulation of the modern amateur radio AM station.  We have the technology, we should work towards using it.
Al

[steve_qix]

I would say that with the example that I gave, it was a transmit bandwidth problem, whether the cause was overmodulation and/or the 25+ kc audio frequency response from the 73 people. At that time, even 3885 was being trashed. So the entire "AM Window" was unusable. The 3880 QSO was pre-existing. But then, we really shouldn't be using 3880?? and the 3873 group had already done the right thing and moved from 3875 just to help out? Aw..ain't that nice 

Anyone remember the mid 70s when there were equally high power tube rigs running around everywhere (pre dates the class E rigs).  Pretty wide stuff back then - a WHOLE lot wider than the modern rigs.  And, a LOT of people had these big tube rigs at that time.

What did we do?  We dealt with it.  Maybe something's been lost over the decades.  Maybe we were tougher (or perhaps more realistic and/or more live and let live)?  Don't know, but I do know we didn't yammer about it all the time.  In fact, I really don't remember hearing about "wide" sigs at all (except from the slop buckets) - and those rigs were PLENTY wide - wider than what's out there today because there was no filtering whatsoever.  I used to run such a rig - pair of 304TLs modulated by 833As.  Not a filter in sight.

But, back to "wide" sigs from 3873 for a moment.  There may be a few old-style transmitters on 3873, but all of the modern ones have 5.-something kHz 6 pole audio filters, and an approximate 10kHz 6 pole output filter.  So, it must be an old transmitter you're hearing with 25+kHz audio frequency response.  None of the ones I designed can do that, and I don't think the Flex radios can either.  Hmm... maybe a modified FT101 or something like that?

What is the purpose of having a transmitter with a 25 or 30 kc audio bandwidth and having the Ss and Ps, and frictive sibilance being heard out so far? We aren't transmitting music. Why do you want music response? Most guys will use the typical vintage receiver with a 6-8kc receive bandwidth, some a little more some a little less. Seems like a lot of wasted energy and power used unnecessarily, that's only going to tick off a whole lot of folks.

Again, no one at least on 3873 can do that - well, ok, there is one person, but he's using vintage AM equipment that's been modified..  The modern rigs are not able to do that sort of high frequency response.  Too many filters.  It's just a fact.  If a modern rig is "wide", it's operator error which can happen with ANY transmitter - AM, SSB, etc. etc. and we've all heard it.

I moved down the band years ago to get away from all that. I see that the attitudes haven't changed much.

Al VE3AJM

Well, I was down there in the low end of the badn within the past 6 months or so and my QSO got pretty heavily interfered with because someone fired up about 6kHz from where I was.  We moved down the band a bit and carried on.  That's life on ham radio.  It was a WHOLE lot worse 40 years ago than it is now, that's for sure.  The bands are so much less crowded now.  And, we don't have the slop-bucket wars we had back then.  I'm sure glad THAT isn't happening too much - at least not around 3885.

[steve_qix]

Something just occurred to me.  I operate on 3885 all the time when the group 3873 is on and I never hear any interference from them at all.  And sometimes I'm even n 3880 and I don't get it.

[Steve - K4HX]

Not my experience and not what the spectrum analyzer shows.

[W2VW]

Reality check needed.

[N1BCG]

Having built a Class E rig and several tube rigs, I don't see how the Class E design results in any more bandwidth issues than the more traditional designs. Any one of these transmitters can be made to sound great or terrible depending on how the modulation is controlled.

Since this thread began as a discussion of asymmetry, I also feel that there's far too much of a fascination with positive peaks given their negligible contribution to perceived volume, which comes from average modulation, not peaks which contain very little energy. Additionally, there's a point where voice asymmetry maxes out and anything beyond that is artificially generated by clipping the negative peaks. That's where its very easy to get into trouble.

Yet another argument against excessive positive peaks is the much greater risk of punching a hole in the insulation in expensive mod iron or damaging other components. For those who care, a 375 watt carrier exceeds 1500 watts PEP above +100% mod. It's more effective to maximize the carrier, reign in the peaks, and properly process the audio.

Broadcast engineering has been driven for almost a century to develop transmissions that were both clean and loud because big dollars depended on it. There's an easy lesson in that.

[steve_qix]

I've been tweaking my home-brew linear/transmitter combo with audio and power drive to get the perfect combo.  I think I'm at a point where I've found the perfect match.  I'm using 200 watts carrier and my PEP-hold feature on my Daiwa wattmeter is showing that I'm at 1200 watts peak.  My math tells me that I'm at 150% positive peak.

I was talking with WV5G last night in NM who told me that my audio sounded great.  We ran a test and I increased audio to 1400 watts, but he started to notice distortion at that level.  In our discussion, he mentioned that he'd never heard a class E rig on the air, and that the audio level between my rig and an E rig would be equivalent.

I thanked him for the compliment, but that I had heard many E rigs on the air back east and the audio quality is unmatched.  

I would like to say thanks to all who have helped me get my station to where it is today, from parts to technical support.  As I gather parts for my 40 meter E rig, hopefully I can put super-efficient and quality sounding RF on the air from Phoenix as well.

Jon
KA1TDQ

Anyway, back to the original topic.  It would be interesting to know what sort of receiver the other fellow was using.

I went over to another ham's place a few years ago and Brent W1IA (who sounds really good) was on the air.  Well, he sounded quite poor on this guy's R390 (might have been an A variant -  can't remember).  Point being, the detector in that particular R390 was clipping QUITE BADLY all modulation above about 80%.  That was one of the worst detectors I had ever encountered until in one of my own receivers, an RME45 (an RME forty-something - I don't remember right now) was a REALLY bad detector.  I ripped that one out and put in a precision rectifier type of detector and cleaned it up pronto.

Point being, some receivers are just plain BAD.  Too fast AGC is another big cause of distortion in receivers.  I gave up about 20 years ago and finally built a receiver (build a solid state receiver using a BC-1004 as a platform) and that one has very good fidelity.  The absolute best I've used is a sync detector - a Sherwood - hooked to a receiver with a well filtered AGC line.  Great sound!

[Steve - K4HX]

No kidding. None are so blind that will not see (hear).

Reality check needed.

[ka1tdq]

That's a good point.  Unless we're using a receiver that has been engineered for high fidelity audio, there's no good judge out there.  99% of hams (me included) are using a rice box receiver or something out of the 50's that's only getting older (and possibly not working properly on top of that).

Having said that, with all my operating down here, I've only had good to great comments about my audio.  Twice people have said, "Wow, there's no distortion in your audio."  And I've received no complaints about anything (too loud, too wide, etc).  

And, I'm just a ham.  I've made a real effort to put together a good sounding home-brew station.  Not corporate standard... just within the FCC parameters and a desire to sound "good".  And, people say I sound "good".  

Mission Accomplished

Jon

[W3GMS]

Not my experience and not what the spectrum analyzer shows.

Same here.  Its impossible for me to operate on 85 .  Most of the guys on 73 are not the problem but one in particular completely wipes things out.  Before you assume I am using a poor receiver that is not the case. 

So for me, the solution is just go to some other part of the band.   Its something that is likely not to change. 

On another note, some folks like to be extreamly wide to keep the SSB ops from getting to close to them.  The SSB - AM wars are not dead if you listen to 73 at all. 

Joe - GMS   

   

[N1BCG]

We all want to be clean and loud, and the b'cast industry figured it out decades ago. Unfortunately, two circuits in every commercial processor sold that contribute a lot to clean and loud are also the most misunderstood: Phase rotators and clippers. Let me explain...

Phase rotators increase audio density (loudness) by reducing asymmetry. Since compressors reduce gain based on the peak to peak voltage of an audio signal, asymmetric waveforms are *reduced* in level more than symmetric waveforms. Average volume ends up being higher for waveforms that require less gain reduction. Phase rotators are available online for under $20, completely built, and can easily be added to your audio chain before your processing.

Peak clippers have gotten a very bad rap because of circuits that utilize just two diodes, and when driven hard, generate loads of harmonics and distortion. What makes clippers invaluable for clean loudness is the careful regulation of the driving audio and proper low-pass filtering that follows. Some form of peak audio limiting is needed prior to the clipper circuit to maintain a reasonable clipping level. Since any clipping creates harmonic energy, a Low Pass Filter should be used after the clipper. Also suffering from a bad rap, these circuits are usually the first to get torn out of boatanchors, and understandably so since they aggressively strangle high frequency response (clippers in Johnson Viking Valiants are driven with unprocessed audio...yikes...thus the need for radical filtering). The LPF should only filter out everything above the highest audio frequency you want to transmit. Multi-pole filters with steep slopes will give you the best highs and neighbor-friendliness.

It's easy to borrow these proven concepts from broadcast processor manufacturers, and by employing both circuits with a consumer audio processor in between, you will achieve a very clean and loud signal.

[steve_qix]

That's a good point.  Unless we're using a receiver that has been engineered for high fidelity audio, there's no good judge out there.  99% of hams (me included) are using a rice box receiver or something out of the 50's that's only getting older (and possibly not working properly on top of that).

Having said that, with all my operating down here, I've only had good to great comments about my audio.  Twice people have said, "Wow, there's no distortion in your audio."  And I've received no complaints about anything (too loud, too wide, etc).  

And, I'm just a ham.  I've made a real effort to put together a good sounding home-brew station.  Not corporate standard... just within the FCC parameters and a desire to sound "good".  And, people say I sound "good".  

Mission Accomplished

Jon

Maybe someone running a Flex or other good receiver can make a recording and we can hear it!!  Unfortunately, propagation to your part of the world from my part of the world is a bit unreliable, at least on 75 meters.

Nice work !!!!!

Regards,  Steve

[steve_qix]

Same here.  Its impossible for me to operate on 85 .  Most of the guys on 73 are not the problem but one in particular completely wipes things out.  Before you assume I am using a poor receiver that is not the case. 

So for me, the solution is just go to some other part of the band.   Its something that is likely not to change. 

On another note, some folks like to be extreamly wide to keep the SSB ops from getting to close to them.  The SSB - AM wars are not dead if you listen to 73 at all. 

Joe - GMS   
   

Apparently I'm missing all the "fun" in the AM Window.  I only hear about the Texas traffic net and they're more of a minor annoyance.  Haven't had a run in with a disgruntled SSBer is ages.... and don't get any interference when I'm up on 3885 from any one on 3873.

But, I'm not on every night anymore since I've been back in the workforce, and when I am on, it's usually late at night.  Only on sporadically in the early evening when all the action is apparently taking place.  Hmmmmmmmm  maybe that's a good thing?  Of course there's the 160 meter portion of the 75 meter band if one wants some solace.

Regards,  Steve

[Pete, WA2CWA]

  Of course there's the 160 meter portion of the 75 meter band if one wants some solace.

Regards,  Steve

I didn't know about this. Is this something new??? Where do I set my dial?

[steve_qix]

  Of course there's the 160 meter portion of the 75 meter band if one wants some solace.

Regards,  Steve

I didn't know about this. Is this something new??? Where do I set my dial?

Ah !  Joe GMS can tell you better about this than can I

[w1vtp]

All:

I thought I should redo the spreadsheet that I did earlier.  I'm calling the first line in Brett's first line "Example 1" and the second line "Example 2"   Remember this is an extrapolation between the S meter readings as the carrier level and the display in dBm.  I picked peaks at (very) approximately 5 KHz and 12 KHz on the left hand side of the display.

Recap: I do not know what levels of modulation the example stations used so cannot characterize which signal is "friendlier" for co-channel operation.  So I redid the spreadsheet to include what I could read off the screen.  It is very hard to pick a screen shot that represents accurately what was going on but I went with what I had.  While I was at it, I did some more work on my S meter vs dBm table.  The PDF will not show the math behind the table but I'll include the XLS file so I can be checked on my math.  Some of you SDR guys might find the S meter vs dBm table helpful.

If I had to pick a "neighbor,"  I'd pick the station as represented in Example 2.  I offer no further characterization other than that.  I stand by my previous comments.  Let's all be good stewards of the frequency spectrum that has been given to us

Later on, I'll probably strip the signal display pages and upload the S meter info as a stand alone file for future use in the "Hints and Kinks" section

Al

[Steve - K4HX]

No guarantee in that portion of the band. Several years ago I was trying to work some European AM stations on 3705 kHz. This was just past midnight local time and when the band would usually be pretty quiet. I kept getting strong splatter. I tuned up the band and eventually came across the source - a Class E station on 3725 kHz.

  Of course there's the 160 meter portion of the 75 meter band if one wants some solace.

Regards,  Steve

I didn't know about this. Is this something new??? Where do I set my dial?

[flintstone mop]

Pete, I think Joe makes a reference to 3.705 that it can resemble the gentleman's band, on 160M at times.
I'm starting to study for the Extra class to get below 3.700 and enjoy some AM there.

And Jon,,,is there another one in the oven?? b-a-b-y??
Fred

[w1vtp]

Pete, I think Joe makes a reference to 3.705 that it can resemble the gentleman's band, on 160M at times.
I'm starting to study for the Extra class to get below 3.700 and enjoy some AM there.

And Jon,,,is there another one in the oven?? b-a-b-y??
Fred

Fred - Sounds like it    Congrats Jon and to your wife.  Nice family

Al

[ka1tdq]

Yup. It's one is 13 months old and #2 is 3 months in the oven.

Jon