Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Icom 756 pro III alc settings for PSK-31  (Read 6480 times)
K4HXC2
Member

Posts: 51




Ignore
« on: June 05, 2012, 07:53:50 AM »

I'm having a ball with this mode but one nagging question remains. Everyone says you should set your output to no alc at all.
Icom says that if the signal being output stays in the alc range shown on the alc meter then no limiting is taking place.
If I set the drive so that no meter movement shows then I get zero output. But I seem to get decent reports when I drive the transmitter just enough to keep the needle within the alc arc, reports drop if I let it get above the alc arc.
I guess it's a question of what's your definition of no alc. No alc needle deflection or no alc limiting (where the needle is above the arc)?

How are fellow 756 operators setting up for PSK-31 on drive?
Logged

Dennis, WK1A
Formally K4DAZ
In honor of Dad, K4HXC(sk)
AA4PB
Member

Posts: 12836




Ignore
« Reply #1 on: June 05, 2012, 08:41:53 AM »

There should be NO alc indication at all. ALC works fine with voice signals as long as it stays below limiting but it usually causes poor IMD with PSK31 signals. The purpose of ALC is to compress the signal peaks a little - which IS distortion of the original signal. On a 2.8KHz wide signal like voice it isn't a problem but with narrow PSK31 it will make the signal wider than it should be.

Again, no ALC means no reading at all, zero, zippo, nada  Wink
Logged
K4HXC2
Member

Posts: 51




Ignore
« Reply #2 on: June 05, 2012, 09:32:39 AM »

Well here's the excerpt from the manual regarding psk-31 operation:
 Operate the PC (software) to transmit.
• When operating in SSB data mode, adjust the AF output
level from PC so that the ALC meter reading doesn’t go
outside the ALC zone.

See the confusion?

Logged

Dennis, WK1A
Formally K4DAZ
In honor of Dad, K4HXC(sk)
AA4PB
Member

Posts: 12836




Ignore
« Reply #3 on: June 05, 2012, 09:47:41 AM »

I'd say the manual is wrong unless they have designed the ALC specifically to function with PSK31. I doubt that because it then wouldn't be as effective with voice. I've measured the IMD on my IC756PRO and any ALC action will cause the IMD to deteriorate. I turn the power control to maximum and then turn down the audio drive (mike gain and/or interface gain) to limit the power output to 30W. This eliminates ALC action and keeps everything nice and linear.

Now that's with PSK31. Some data modes, like RTTY, don't have any amplitude variations and the ALC doesn't matter in that case as long as its in the normal range.
« Last Edit: June 05, 2012, 09:49:43 AM by AA4PB » Logged
K4HXC2
Member

Posts: 51




Ignore
« Reply #4 on: June 05, 2012, 09:55:05 AM »

Ok, They're specifically referring to USB-D mode, I wonder if that's different?
Anyway, I'll give that a try tonight, Thanks
Logged

Dennis, WK1A
Formally K4DAZ
In honor of Dad, K4HXC(sk)
AK7V
Member

Posts: 250




Ignore
« Reply #5 on: June 13, 2012, 10:25:05 AM »

How much power are you running?

I have the pro 3 as well.  Like AA4PB, I keep the RF output at maximum and adjust my output power by adjusting the audio level from the sound card (or Signalink USB in my case).  I keep output at or below 25 watts and never have any ALC indication.
Logged
K4HXC2
Member

Posts: 51




Ignore
« Reply #6 on: June 13, 2012, 10:51:42 AM »

Actually, I have a RV setup with a 31' vertical and the steel framework as a counterpoise with a tuner at the radio end, so I'm calculating there's enough feed losses that I probably need to run about 45 watts to get 30 or so at the antenna.
After a careful read of the posts I've found that using about 75% power on the radio ( not the suggested 100% so if I forget and switch on the amp I won't overdrive the amp for other modes) and adjusting the tx level on the Signalink until about 40 watts is indicated at the output of the tuner, I seem to get great signal reports and IMD around -17db.. This shows nil to a slight, occasional, needle barely moving from zero on alc.  When I'm at home with the big beam 25 watts seems to do well, probably because there is almost a perfect match for the entire system there. The RV is altogether different.
Logged

Dennis, WK1A
Formally K4DAZ
In honor of Dad, K4HXC(sk)
AA4PB
Member

Posts: 12836




Ignore
« Reply #7 on: June 13, 2012, 01:31:21 PM »

The thing about ALC is that the time constants have a large impact on what the ALC does to a PSK31 signal. Those time constants vary from one radio mfg/model to another. With some radios a little ALC won't cause much IMD while another radio will have very poor IMD with the same amount of ALC. As a result, the **safe** recomendation is to keep the power down where there is no ALC indication unless you have an IMD Meter or a second receive setup to actually measure it (which isn't a bad idea anyway).
Logged
K8AC
Member

Posts: 1471




Ignore
« Reply #8 on: June 16, 2012, 03:33:25 PM »

Quote
I seem to get great signal reports and IMD around -17db

First of all, many PSK operators don't know what a good signal looks like - as long as they have solid copy, the signal report is good.  The fact is, a -17 dB IMD measurement is just short of TERRIBLE!  A reasonable and achievable goal is - 30 dB or more.  Anything less than -20 is a problem.  While adjusting the drive for no ALC action is better than nothing, it's no guarantee that you've adjusted things for the lowest possible IMD level.  The only way to know for sure what the level is, is to measure it using one of the RF sampling devices being sold today for the purpose, or to watch your own signal on a separate receiver and let the software measure the IMD while your signal is idling.  Don't forget there are other things in the signal chain (such as the soundcard itself) that will contribute to the IMD problem (cheap integrated sound cards/chips being worse than professional grade external sound cards).

I've been monitoring PSK transmissions for a couple of weeks now to see what typical IMD levels are.  It's not unusual to see a signal with an IMD in the -9 to -11 dB range get a rave review from the other guy in the QSO.  Granted that a weak signal may not be strong enough for the software to be able to measure the IMD, but some of these signals have sidebands covering hundreds of KHz each direction from the signal.  Minimizing IMD on PSK signals is not a simple thing and everyone who can should measure their own signal and play with the variables to achieve the desired result.  

73, K8AC
Logged
K4HXC2
Member

Posts: 51




Ignore
« Reply #9 on: June 17, 2012, 05:31:48 AM »

Short of investing in expensive equipment or another computer I suppose the best way is get reports from others.
Which I'm guessing is not entirely accurate. However too much IMD could account for those times when I can make far contacts but nothing close (discounting propagation) Perhaps there's too much distortion up close. Being fairly new to this digital thing, can you tell on the waterfall when a station has too much IMD? Maybe a pic?

Also begs the question: Wouldn't somebody come over and let you know like:

WK1A, WK1A de K4DAZ, K4DAZ, Hey Nitwit! You're blasting IMD all over the band, lighten up!  sk   Grin
Logged

Dennis, WK1A
Formally K4DAZ
In honor of Dad, K4HXC(sk)
K8AC
Member

Posts: 1471




Ignore
« Reply #10 on: June 17, 2012, 06:40:30 AM »

You don't need another computer to do the testing.  I do it with a single PC, but two different soundcards and two different pieces of software (requires a second receiver).  Too much IMD probably won't have any effect on your ability to make contacts, although the power in the unneeded sidebands means less power in the fundamental signal where it's needed.  But - if your signal has a high level of IMD, you sure can effect the ability of others to make contacts. Provided that the received signal is of reasonable strength, you can easily see the IMD in the waterfall and with experience make a decent guess as to what the numeric level might be.  In this case, "reasonable" strength means that the signal has to be above the noise level enough that the IMD at 20-30 dB down is visible out of the noise.

Take a look here to see several pics of good and bad PSK signals.  There's a bit of info on SSB signals there that you can just skip over. 
www.k8ac.net/signal_images.html

I avoided putting IMD numbers on the pics since the receive software can't make that determination unless there's a sufficient period of idle time to do the calculation on.  Note that I didn't have to spend a lot of time hunting for the bad signals - they're quite common on a Saturday morning on 20M !
Logged
K4HXC2
Member

Posts: 51




Ignore
« Reply #11 on: June 17, 2012, 06:57:42 AM »

Oh yes, I can see it quite well.   And the are good explanations as to potential causes, thanks.
Maybe you could pm me. We could do a sked this week?  I'm currently in the NC foothills so we might be able to do something on 20m.
I'd value a good and useful report.
Logged

Dennis, WK1A
Formally K4DAZ
In honor of Dad, K4HXC(sk)
AA4PB
Member

Posts: 12836




Ignore
« Reply #12 on: June 17, 2012, 08:16:08 AM »

http://www.ssiserver.com/info/pskmeter/

For about $50 you can get a PSKMETER kit that will measure your IMD.

On air reports are okay, but the measurements must be taken while the signal is idling and while the noise and QRM are pretty low in order to get a valid measurement. Also, any distortion in the receiver (fast AGC for example) can cause a good signal to look bad. IMD is a total measurment from the transmitter sound card through the receiver sound card. Any non-linearity anywhere in between will show up as poor IMD at the receiver. Thats why something like PSKMETER is ideal to measure your own transmit signal while eliminating all the other variables.
Logged
K8AC
Member

Posts: 1471




Ignore
« Reply #13 on: June 17, 2012, 12:04:39 PM »

The PSKMETER is one of the devices I was referring to earlier, another one is: http://kk7uq.com/html/imdmeter.html

Under what conditions do you think fast AGC would affect the distortion on a received PSK signal?  I haven't noted any difference here between AGC speed and my observations.  What makes most sense to me is to turn the AGC off, since with it on, any strong signal in the SSB passband will desense the receiver a bit.  Having said that, I guess I usually run mine at medium or fast. 

It would be interesting to know just how much each element in the total system effects IMD observed at the receiver.  Clearly, a low-end sound card will have an impact, but just how much I don't know. 

73, Floyd - K8AC
Logged
AA4PB
Member

Posts: 12836




Ignore
« Reply #14 on: June 17, 2012, 04:23:58 PM »

I ran some test several years ago using an IC765. I found a very noticeable degrade of IMD with the fast AGC setting and a test signal at an S9 reading. With a lower signal level (S4 or S5) the AGC activity wasn't high enough to "compress" the received signal. With a slow AGC setting the decay time was slow enough that it wouldn't modify the signal waveform - rather just hold at an average gain level.

Basically, AGC does the same type of thing in a receiver that ALC does in a transmitter. It automatically changes the overall gain to maintain a constant output level. The question is whether the response times are quick enough to change the gain to follow the modulation envelope and flatten the peaks. If the response time is slow enough to maintain and average gain while permitting the modulation envelope to change without being modified by the AGC or ALC then neither one will cause the IMD to degrade. I found that I could modify the ALC decay time on a transmitter and permit it to function without IMD being degraded. The benefit of doing that is that it makes audio level adjustments much less critical as the ALC can take care of the gain automatically. One of the things you notice on some transmitters (depending on the quality of the filters - the amount of ripple) is that moving the audio frequency around the waterfall will cause the transmitter output power to change.

The potential issue with turning the AGC off is that strong signals in the passband could overdrive the receiver and cause distortion. If you can adjust the RF gain control to have sufficient strength from the desired signal without the stongest signals overloading the receiver then AGC off can work. Personally I leave the AGC on at a slow setting with the RF gain wide open and just let the AGC handle it.
« Last Edit: June 17, 2012, 04:29:42 PM by AA4PB » Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!