The AM Forum

I had a nice chat with Dave W2VW this afternoon on 75 meters and this fine business old fellow joined us. It sounds like he's not feeling very well. If you come across him, encourage him to keep talking to clear out those lungs! A short mp3 file is attached to help you identify him.

I think he lost a lung there at the end  :-[ :-[ I know I almost my lunch  :-\ :-\ :-\ :-\

Oh my....  1:10 - 1:24 -  did anyone call 911 ?!!!   ;D


Just add in the SBE rig and create a duet entitled: "Drowning in their own puke"  ... :o


T

Must be winter time - Tron is coughing his brains out. I pity his mic!

Yeah, Tim does that all the time - I worry about it because it sounds like more than a cold.

What AGC speed were you using? It sounds VERY compressed. Is that a Flex?  The newer versions of the Flex software have BAD AM  AGC - the AGC responds to the audio (not good).  So, I always run the AGC on the slowest possible setting (long, usually) to avoid the compression effect and back down the AGC-T.  AM sounds a whole lot better under these conditions.

The AGC code, at least for AM needs to be rewritten.  Yeah, Yeah, I know - one of these days I'll dig into the code see how they're doing the AGC and make a modification and fix it.... in my copious free time  ;)

I think that maybe a "cough" button next to the mike might be a good idea, you could mute the mike without un-keying the transmitter. They were standard in broadcast, at least when they did live radio.

Hi Steve,

Tim mentioned that he and Marcy both came down with it after exposure several days ago to someone's children who had the same symptoms.

The audio sounds strange for a lot of reasons. I was riding the AGC-T because Tim's signal was subject to quite a bit of QSB. Additionally, a sideband station came in about 4 kc below us, so at some point I pulled the lower sideband in to about 3 kc on the low side of the carrier. Then an AM QSO started up 9kc above us, so I pulled in the upper sideband to about 4 kc above the carrier. The clip is actually a cut and splice of segments from a single 10 minute transmission, so the levels and bandwidth filtering are different throughout. Since I was using synchronous detection, there is also a very low frequency component that gets recorded as a result of the slight error in the lock to the carrier. I ran it through a high pass filter to scrub out everything below 20 cycles to clean that out. Because of the changes in amplitude throughout the spliced segments I ran the whole clip through Audition's hard limiter to keep everything fairly close in volume. That raised the level of everything 2 dB but limited the maximum amplitude to -.5 dB below the range pre-set in the hard limiter. I believe the last step is the biggest contributor to what you are hearing. So, the clip has been highly processed (and mutiliated) by someone who is NOT an audio engineer!

As far as the PSDR's AGC goes, I also use the "Long" setting and keep the AGC-T low and the AF high to avoid what you were referring to. It seems to behave very similar to a manual RF gain control. There is a definite sweet spot where most of the signals, weak or strong, will have very similar amplitude without pumping. It is certainly not an automatic setting though!

I think you should definitely take a run at the code and make it better! Before you do though, if you have not done so, you might want to look at the adjustments available on the AGC setup page. You can adjust the AGC slope, max gain, attack, decay, hang, hang threshold, and fixed gain. Through trial and error I found settings that seem to work very well for most casual AM QSO conditions. When the going gets rough though I find that it's necessary to ride the AGC-T lever. That whole aspect of PSDR could use some help. Go for it! If you fix it, I'll buy you a sausage sandwich at Angie's wagon at Nearfest!

Man, that was retched. Think about that next time your at a fester wondering how that sweet looking used microphone on someone's table might sound on your station.

Any used mics I get that I can't take apart and boil on the stove for an hour get tossed directly into the garbage.

Oh, ok Rob - the compression after-the-fact explains everything!  The signal was obviously compressed, and I just assumed it was the AGC (which gives a very similar effect when in fast mode).

I've played around with the AGC settings quite a bit, and can't seem to do anything to prevent the modulation from generating AGC.  I think that's going to involve a code change which will do the AGC function based on the average (filtered) value of the AM signal and not the peaks.

Steve,
I think you are suffering from softrock hanging off an IF issues. The gain ahead of the softrock may be too high. Also the gain of the softrock op amps.
I found the op amps want less that X10. I think softrock runs around X50.
You don't need that kind of gain hanging off an IF. Also the softrock needs to see 50 ohms for the tayloe detector to work properly. Softrock is designed to hang off an antenna so needs a lot more gain to drive a sound card. Your receiver should be in manual gain mode so there is no interaction between the two AGC systems.

I noticed the same overload with my SoftRock off the SP-600 455 IF.  I ended up adjusting the RF gain just cracked open at about "2" most of the time. In fact, I set the RF gain until I JUST begin to see the noise floor come up.  The dynamic range is better this way for measuring IMD too.

Rob, so that intense hacking segment was a chop-chop of a 10 minute OB transmission?  Man, you sure know how to make a guy look bad.... ;D ;D

Poor Tron and Marcy... about four years ago I had a cough like that that stayed with me for about 2 months. No cure but rest.

T

Yea, making old buzzard xmission probably isn't a recommended cure. I do worry about the guy when I hear that nasty cough.  :(

Hopefully Tim will feel better soon. Yes indeed, being largely unsupervised all day while my wife is at work leaves lots of time to exploit the dark side on the bands. By the way, I posted a sound clip of the infamous high voltage "Fabio Shakedown Transmission" on the new AM Facebook page. You are more famous than you know Mr. Vu.

Unfortunately, I wish it were this simple!  This is definitely a software problem.  With my own setup, I don't use the AGC in the PowerSDR software at all - keep it on manual because I have enough AGC available in the front end receiver itself.  So, personally I can work around the problem.

All of the new versions of PowerSDR suffer from this problem.  Even a "real" Flex radio (I just used one at Al W1VTP's place) will do this if they are using the newer versions of the software.

I know, for instance, that Frank WA3JBT has not upgraded to the newest version of PowerSDR for this very reason.  I've heard other recordings from Flex setups - same thing.

Interestingly enough, the WinRadio doesn't have the problem - this is only a receiver, however.  I get fantastic recordings from Bill KC2IRF without the usual AGC anomaly.  Of course, WinRadio is using their own software.

Anyway - something to fix over a non-busy week (HUH!).   :D

The 811 rig doesn't help.  Needs a filter on the outpoot or maybe less filament voltage.

Or just sprinkle it on a pizza.

Yeah, the Tron is an old vet at battling these colds for years now.  He'll be OK by Nearfest, guaranteed.  ;)

I haven't seen the Facebook AM site yet, Rob. Do you have a link?  


Steve: So Bill is using a Winradio?  He's sent me some recordings over time and I've always thought his were the best I've heard, followed by Al/VTP with his Flex.   I was considering a WinRadio at one time, but the HPSDR  boards looked better suited for experimenting so I went wid those for now.

T

http://www.facebook.com/#!/pages/Amateur-Radio-AM-mode/167171563328186 (http://www.facebook.com/#!/pages/Amateur-Radio-AM-mode/167171563328186)

You could probably enter a "bug report" over on the Flex web site. They're still tweaking the Version 2 software.

Once we did an interview with John Popper, harmonica player for a band called Blues Traveler.

He mentioned that they did a show when he had the flu and he barfed into his harmonica while playing it...Had to replace it on stage..

Gross.

I do the muting thing with the plantronics headset console when I am on a conference call. Does anyone use a headset on the ham radio (aside from doing so to not annoy others in the room)?

I think it's been reported in the past - considered a "non issue", I believe.  But, I could try I suppose...

Thanks !

Steve

I found four AGC user bug reports that were related to AM reception. Two of them seem to fit closely the behavior we are seeing with certain software versions:

http://support.flexradio.com/ReviewBug.aspx?id=2282&it=B

http://support.flexradio.com/ReviewBug.aspx?id=2240&it=B

Another two related to AM reception are located here:

http://support.flexradio.com/ReviewBug.aspx?id=2274&it=B

http://support.flexradio.com/ReviewBug.aspx?id=2498&it=B

I have not yet installed the latest beta version (2.0.16) so I have no idea if changes that were made (there were many) have had any impact on whether the AGC still reacts to the audio on AM or SAM. I do know that Flex is not very interested in pursuing any bug reports with versions other than the latest beta. I suspect that you could articulate the issue better than most of us who use the software Steve. Definitely report it as you see it.

 

If you're interested in following through with your observations, I can recommend sending them to the attention of Greg Jurrens - K5GJ, VP, Sales and Marketing at Flex.  I've been exchanging some emails with him and he is puzzled that he has not heard anything about these issues in the 18 months he's been with the company.

Not only did he express high interest in pursuing the cause of any anomalies like the ones you, Rob/AEX, and perhaps Al/VTP are hearing, but he plans to set up a test to see if he can replicate the problems reported in the bug report links AEX compiled from the Flex website.

To all three of you, I recommend making audio recordings and describing your receive conditions, and send them to him with notes matched with timings so that he can precisely analyze, at least by ear, what you're hearing.

He plans to make such recordings and send them along if he finds anything to confirm or dispute what you've noted.

I think this represents an unusually receptive level of interaction with an equipment manufacturer, and I hope you pursue some answers with him.  I have an email address and a direct phone number to share as well. Email or text me 202 841 3208  anytime.

Hi Paul,

That sounds like an excellent idea. I have a short audio sample that I believe demonstrates it very well. It was recorded a few minutes before Tim joined Dave and I on 75 meters. If you listen to the MP3 file, you'll immediately notice that it sounds like Dave is using a poorly adjusted limiter, but in fact, he was not. The limiting effect was the AGC in Power SDR version 2.0.8 responding to his audio peaks. Note the gain reduction that immediately follows each word. The screenshot of the audio sample taken in Audition shows the gain change between each word as Dave is speaking.

http://members.cox.net/w1aex.fn31/PSDR_AGC_AM.mp3

Perhaps Steve QIX can confirm that this is the effect he is referring to. I would have to say that this is an extreme example, and I have no doubt that my AGC-T level was not optimal, but it is a strange anomaly.

Rob

I just wanted to check in to say, after listening to that audio clip, how much I miss the northeast AMers.  Ay yi yi.  

Still in Va. Beach and and occasionally listen for the gangsters on a random wire and the FRG100, but sadly all I seem able to hear are 4-land slopbuckets which is too depressing for words.

73
Paul, K2ORC

Actually, that wasn't too bad.  You'd really have to know what the station in question sounds like for "real", but you could hear the background come up a bit when there was no modulation.  If you put the AGC on fast, and turn up the AGC-T a bit, it will be REALLY bad.  Lots of pumping and artifacts like that.

 Interesting on the AGC action on newer versions of the PowerSDR.. Here is a recording of my recently finished cheap SDR general coverage kit .. (fun working with surface mount!).. The PowerSDR version is the most recent 2.0.16 Beta set up to work with softrock... AGC T is set nominally (at 80)..

opinions?

http://dl.dropbox.com/u/17340243/come%20and%20go.mp3

take care,
Jeff

I heard parts of that same transmission and didnt even realize who it was until the ID. Didnt stick around as I was working on taming birdy generators throughout the house.

Carl

That sounds (and looks) very smooth Jeff. I lifted a small sample and looked at it with Audition. It shows no variation in gain with modulation. The screenshot shows a flat-line between words instead of the rising gain symptom. Was that recorded using AM or SAM? I think I'll install 2.0.16 this weekend and see how it looks.

73,

Rob

I always called the crud in mikes "moustache dandruff". My friend Jack calls it "radio breath". Either way it's nasty. An ultrasonic bath will do wonders for all but the element itself.

"I love you sweet lief"   B.S.

Hi
I'm not aware of the AF pumping the AGC on my Flex.  I believe I can put an unmodulated carrier on and vary the AM modulation % and the output will accurately track the mod level.  I'll have to take a function osc home from work to check this out.

I do have a custom setting where I really draw out the hang time for several seconds but have found that the "Long" setting that PSDR offers is OK with me.  Sounds like a nice evening project when band conditions go bye bye.

Al

It's how fast you vary the modulation level that's the issue.

I don't know about the Flex rig but HPSDR there seems to be a lot more audio gain than nesessary using PSDR. I tried SSB last night and was way over modulated until I went into the EQ and set the main level pot way back. I was using an EV664.  I was running "6" on the compression setting.
High audio gain and a fast time constant would make things go crazy.