The AM Forum
December 14, 2024, 01:38:10 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: Icoms on AM, external ALC mod  (Read 46569 times)
0 Members and 1 Guest are viewing this topic.
Jim KF2SY
Member

Offline Offline

Posts: 290



« Reply #25 on: February 03, 2011, 06:10:21 AM »

Spent more time on the air last night with my friends 735 and this box.  He sounded really good with it.

C

Clark,

I modified my IC 735 for better sounding AM the old fashioned way ( changing coupling & by-pass caps) after a lengthy on-air tutorial  by Timtron WA1HLR back around 1992 or '93.  There is internal pots for the ALC which can be adjusted to eliminate the backward modulation.  Much of these internal circuitry mods can  probably be skipped with very good results by feeding the audio directly to the back of the radio via the DIN connectors AFSK audio input pin.  This is the way I use it now.  I think Joe, N1VIV does this also with his 735.  The Icom 735 AM transmit does NOT go thru any SSB filtering.  Very good HI-FI results can be had.

Jim
 
Logged
KA2QFX
Member

Offline Offline

Posts: 236

Mark


« Reply #26 on: February 04, 2011, 02:27:49 PM »

   I am intending to write a white paper on ALC, as this is a subject which interests me and I have investigated in some depth over the years on several amateur product lines.  My interest stemming mainly from that fact that ALC is often more of a detriment than asset in transmitting a “good” signal.   “Good” of course being a very subjective term meaning something quite different to an “AMer” than a “Contester”.   However, my outline for the white paper indicated a lengthier document than expected and one which requires a little more bench measurement before I publish.   
   In lieu of that, here’s a brief explanation of how and why the external bias method works on some later transmitters, and what problems it does and does not solve.  The attached diagram (below) describes in simple block form the AM ALC control loop of an ICOM 746.  Other late model ICOM's and some Yaesu's follow similar designs.   
   You’ll notice that for AM there are three variables that control the level of  RF applied to the final amplifier. 
1. The AM carrier level control; this is the amount of unbalance applied to the balanced modulator.  This is the primary factor in determining the optimum ratio of audio to carrier in the radio.  Too low a setting will produce a DSB signal below baseline as – peaks are increased. Too high a setting and the positive peaks will be clipped early as the modulator runs out of headroom.  The optimum setting theoretically is where the RF out, at carrier is ½ of the maximum modulated output.  That however, in this design, is subject to the amount of available RF applied to the modulator.

Which brings us to
2. The “power”control: The output of the BM is proportional to the product of the RF and audio levels applied to the mixer. However, in practice using IC balanced modulators there are limits to the output available depending on input levels, mixer efficiency and power supply voltage. I would much prefer that power be controlled after the BM to avoid the issue of matching audio, carrier offset and RF in the BM when used for AM.   In this case the “Power” applied to the BM is determined by the radio’s CPU and a voltage controlled amplifier (VCA) regulating the 3rd L.O. input to the BM. 

And finally
3. The following IF amplifier chain which contains another VCA controlled by the ALC control voltage.  This control voltage is also developed by several other entities such as reflected power, which I’ve shown.   

Normally the IF chain following point “A” operates at full gain. Let’s arbitrarily say from “A” to output is 30dB of gain.   Let’s also say that the forward power detector develops
-10 volts at 100 watts of RF output. 

So, for 100 watts of output, 2.2 volts of signal is required at “A”, or 22 volts at “B”.   Any more than that and the ALC will reduce the gain of the IF amp to maintain 100 watts output.  Naturally, for SSB this gain reduction is easily compensated for by adding a little more audio drive. But since this gain reduction also reduces the AM carrier it is undesirable.  In addition, on AM, a low pass filter is added to the ALC loop to filter out the modulating peaks normally detected on sideband.  Since this filter must have some reasonable response time it is limited in how low a modulating frequency it can filter out. Consequently, signals which have somewhat enhanced low end suffer greatly from premature ALC action.   (Yaesu uses this ALC loop actively to stabilize power output making things much worse in this regard.) 
   This control loop is rather critical in that you’ll see that as the gain is reduced to hold the output at 100 watts, the ALC detected voltage will also remain about -10 volts. So the threshold remains apparently fixed while the gain of the IF amp is reduced.  It is therefore likely that the position of the potentiometer used to develop the external ALC is likely rather “touchy”? 
    OK, so now you set the External ALC potentiometer  to reduce the gain of the IF by 12 dB.  To develop the same 100 watts output still requires 22 volts at “B” but “A” has been raised to 8.75 volts.   That’s 12dB more signal. In terms of the output from that balanced modulator you now need to raise the carrier  level (Power Control) up 12dB to compensate and your audio level also can be raised proportionally. 

   Your normal ALC detection system would now have to see significantly more RF than 100 watts to exceed to the setpoint of the external ALC input. Because of transfer functions of the ALC loop, which I won’t get into, it’s not 12dB more, but it is enough to effectively disable the ALC almost completely; certainly greater power than the final amplifier can deliver.  Unfortunately, the same holds true for the SWR protection system. Since the ALC voltages are all summed to provide attenuation by the same mechanism++, the reflected power would also need to be significantly higher to reduce the gain and protect the final amplifier.   Hence you have no protection either. 

   I chose 12 dB just as a matter of convenience, but it’s likely not far from reality.  To determine just how much attenuation you are inserting do the following.  Set your external ALC as you desire for AM operation. Unplug the external ALC. Set the power level as high as possible WITHOUT developing any ALC. Measure the output. Plug the external ALC in and measure again. The difference in dB is easily calculated as 10 *log(low/high).

++ This likely explains why some of you have reported strange indications of SWR metering.

   A final word on ALC in general.  Overall, I find the affect of ALC on signal quality to be undesirable. But to simply defeat it altogether I consider unwise UNLESS some other method of strictly controlling excessive output is employed.  For many of us on AM that method is a properly set peak limiter in the audio chain.
   In my opinion, ALC is a desirable protection method to destructive overdrive, a last resort so to speak. I rely on ALC when tuning my grid driven amplifier. The ALC in this case is derived from the grid current of a pair of 4CX250Bs. Any grid current (>120uA) develops plenty of ALC and pulls the exciter down preventing blown grids. For tuning it’s great. I would never use it to limit my audio peaks actively during transmit. Contesters (and others) who use it as a form of cheap compression should have their mic cords snipped!  Some have argued that ALC action itself causes distortion and splatter. I’ve read the article posted recently http://www.sm5bsz.com/dynrange/alc.htm  on the subject and cannot concur with its findings, mainly because I find the test methodology flawed. But that’s another matter. My feeling is that splatter which is coincident with high levels of ALC is more the result of overshoot and flat-topping from ALC delay and distortion in the lower level stages and from simply overdriven mixers and amplifiers.  But this is a matter I will address in my paper after much more data collection and number crunching.

Comments and experience of others are welcomed. Perhaps the moderators would choose to make this a different thread altogether.

Regards,
Mark



* RF_ALC.JPG (35.17 KB, 788x581 - viewed 844 times.)
Logged
W3DBB
Member

Offline Offline

Posts: 57


« Reply #27 on: February 05, 2011, 09:04:14 AM »

I've been pissing around with an Icom IC-737 on 40m AM. It sounds like "2 Dixie Cups and a string". In defense of the IC-737, it modulates correctly as it maintains a steady 25W carrier with close to 100W PEP output. This without resorting to the golden screwdriver or an ALC defeating device. I'm pretty sure AM still goes through the IC-737's balanced modulator with its inherent audio bandpass frequency response limitations so the capability of the Heil ICM microphone I'm using is largely unrealized. By comparison the Yaesu FT-920 I've owned since new is awful on AM. The only good one I've heard is NA3CW Chuck's modified FT-920. A Mark V Yaesu recently acquired is much, much better than the unmodified IC-737 and unmodified FT-920 on AM. Plus the Mark V will make a 50W carrier with close to 200W PEP out on AM. Also have another Icom, a barebones IC-707, forerunner of the IC-718. This one is pretty much an unknown at this point when it comes to AM performance but it may be pressed into service before long at the portable/3 location. The moral of the story: solid-state transceivers are are a mixed bag when it comes to their performance on AM.
Logged

Doug

beautiful downtown Strodes Mills, Pennsylvania
KA8WTK
Contributing
Member
*
Offline Offline

Posts: 873



« Reply #28 on: February 05, 2011, 05:47:20 PM »

Built one of these boxes and put it on my IC-730. Here is what I am seeing.

I adjust my monitor scope for 4 divisions in CW transmit at full power (100 watts). I then use the box to reduce the trace on the scope to 1 division high (25 watts). Switching to AM, modulation will hit 100% (4 divisions) on the scope. Power meter deflects up instead of backwards like "normal".

But, I can do the same thing by reducing the RF power control on the front panel to 25 watts out (1 division). So, I don't see an advantage to the box with this particular radio. I wonder if it has had some sort of modification or has had some ALC tweaks made by a previous owner? Or, the ALC setup in the IC-730 is different than the newer Icoms.

Or, and this is a BIG possibility, I am doing something wrong.

Bill KA8WTK
Logged

Bill KA8WTK
KA2QFX
Member

Offline Offline

Posts: 236

Mark


« Reply #29 on: February 05, 2011, 07:55:11 PM »

Bill,
I can't imagine what you could be doing wrong. So I don't think any operator error is possible.  I suspect your ALC has been tweaked as you say.  Without the external ALC do you see ALC action on the meter in the radio normally when you hit 100 watts?  Or, perhaps a better question, at what power level do you see ALC kicking in, or "backwards" swing start to occur? 

Not sure on the exact ALC implementation on the 730, but I doubt it's too different. I'll have to review a schematic. Suffice to say that your 730 is similar in that the "power" control and ALC use separate circuitry to control output level; which is why this method works at all.

Thanks for your info.
Mark
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8301



WWW
« Reply #30 on: February 05, 2011, 08:30:05 PM »

Please write the paper!
Logged

Radio Candelstein
K5WLF
Guest
« Reply #31 on: February 06, 2011, 12:56:10 AM »

I'm in the process of trying to put together a good AM station. I'm currently working with just a Yaesu FT-897D. From all the reading I've done, I have somewhat ambivalent feelings about ALC. I'm definitely looking forward to your paper, which I hope will clear up some of my probably misconceptions.

ldb
K5WLF
Logged
KA2QFX
Member

Offline Offline

Posts: 236

Mark


« Reply #32 on: February 06, 2011, 05:17:21 PM »

For what it's worth my paper will take a while.  I need to get several late model radios across the bench and apply a standard set of tests to determine exactly how various ALC circuits respond and affect the output of various radios. Things like attack delay, attack rate, compression level, averaging filter rolloff (AM), etcetera. 

Aside from the number crunching I think I pretty much summed up my opinion of ALC above. I can understand the ambivalence toward ALC. I try to avoid it completely, but I'm glad it's there when I goof-up. 

Mark

Logged
Steve - K4HX
Guest
« Reply #33 on: February 06, 2011, 08:18:07 PM »

Sounds like a peak limiter would be more appropriate.

The problems with single-channel integrating feedback control systems are well known. Most ALC systems are of this type and will have the same problems. This is why broadcasters went to multi-section, multi-band approaches decades ago.
Logged
WD8BIL
Contributing
Member
*
Offline Offline

Posts: 4413


« Reply #34 on: February 07, 2011, 09:39:11 AM »

KA8WTK wrote:
Quote
But, I can do the same thing by reducing the RF power control on the front panel to 25 watts out (1 division). So, I don't see an advantage to the box with this particular radio. I wonder if it has had some sort of modification or has had some ALC tweaks made by a previous owner? Or, the ALC setup in the IC-730 is different than the newer Icoms.

Hi Bill.
If you're gonna see a difference it will be on voice peaks. "P" pops and atomic yeeaallooowwws may drive the normal ALC to cut the carrier. If this isn't happening then there's a good chance someone has tweaked it already.
Logged
W2VW
Contributing
Member
*
Offline Offline

Posts: 3483


WWW
« Reply #35 on: February 07, 2011, 09:49:48 AM »

I used a 730 back in the late 80s. Ran one of those Radio Shack 5 band equalizers right into the mic input. It didn't seem to have any ALC problems at all with 20 watts carrier.

A much easier radio touse on AM than many later Icomz.

Built one of these boxes and put it on my IC-730. Here is what I am seeing.

I adjust my monitor scope for 4 divisions in CW transmit at full power (100 watts). I then use the box to reduce the trace on the scope to 1 division high (25 watts). Switching to AM, modulation will hit 100% (4 divisions) on the scope. Power meter deflects up instead of backwards like "normal".

But, I can do the same thing by reducing the RF power control on the front panel to 25 watts out (1 division). So, I don't see an advantage to the box with this particular radio. I wonder if it has had some sort of modification or has had some ALC tweaks made by a previous owner? Or, the ALC setup in the IC-730 is different than the newer Icoms.

Or, and this is a BIG possibility, I am doing something wrong.

Bill KA8WTK
Logged
WA3VJB
Guest
« Reply #36 on: February 07, 2011, 10:00:34 AM »

Sounds like a peak limiter would be more appropriate.

The problems with single-channel integrating feedback control systems are well known. Most ALC systems are of this type and will have the same problems. This is why broadcasters went to multi-section, multi-band approaches decades ago.

Yep.  The KO6NM approach.


Logged
KA2QFX
Member

Offline Offline

Posts: 236

Mark


« Reply #37 on: February 07, 2011, 10:10:50 AM »

Steve is of course quite correct. While I neglected to include it in the above writing I advocate the use of limiters and compressors as a much more practical way of controlling the output of one's transmitter.  In this instance I tried to focus simply on the mechanism of using external ALC to work around the problem in these rice boxes.

Regarding the problems of integrated feedback control systems being "well known" I suppose that's true amongst engineering oriented operators. However, listening to the bands during a typical SSB corn-test it's obviously less well known.  Smiley Hopefully, I'll be able to put some more tangible numbers to the detrimental affects of ALC; not that it's likely to solve any problems, but it can't hurt.  

One of the hypotheses I hope to prove is that later implementations of ALC in ham gear, which moved ALC from the audio stages to the RF stages, made things much worse. It is likely that the RF designers were seeking to reduce some of the problems inherent with integrating the feedback for audio and applying it so early in the RF chain. By moving it to the later RF stages (post filtering) they may have sped up their integration and reduced some of the delay/overshoot issues but inadvertently made matters worse as far as widened bandwidth.

Some of SM5BSZ's later analysis indicates the ALC loop behaves as an under-damped control loop. The compromises available between speed, damping and increased bandwidth are likely all poor. But I'd like to know what compromise the ricebox designers chose just because I'm curious.  Granted, integrated DSP and SDRs resolves much of this problem having the ability to "look ahead". But I'm still inclined to satisfy my curiosity.

Thank you Steve, you have a gift for being concise.

Mark
Logged
Steve - K4HX
Guest
« Reply #38 on: February 07, 2011, 10:52:40 AM »

I generalized more than I was concise.  Grin  That's why I'm all for getting some measurements on actual radios and doing some circuit analysis (specifics). It should be quite interesting, and as you say, may open some eyes. Go for it!
Logged
K5UJ
Contributing
Member
*
Offline Offline

Posts: 2814



WWW
« Reply #39 on: February 07, 2011, 01:30:45 PM »

I read the SM5BSZ article a few years ago (this was back before I got into AM) and hauled out the service manual and schematics for my FT1000MP Mk V to see how its ALC works.  It is the typical PA coupler that sends a control voltage back to a VCA at I think, the 455 KHz IF.  I did not and do not know enough to understand all the precise design details but I was able to grasp that the fundamental flaw with ALC for peak power limiting is that once a too-high peak has gotten to the coupler, it is to late to do anything about it.   It seemed like the circuit, which seemed to make sense for VSWR protection, was a cheap way of providing some sort of forward peak limiting, but I  never heard of such a scheme being employed in any professional gear like broadcast rigs.  I figured since broadcasters seem to employ agressive audio peak limiting, I'd go that route and I did, originally for SSB, but I soon realized I could operate AM with it and I did and proceeded to make that my primary voice mode.   But before I did that, I would set the peak limit on the audio, to not let the rig go over say, 150 watts, crank the "power" pot on the rig almost all the way up and limit the rig's peak power with no ALC intervention.   I never did anything to the rig"s ALC so on AM I can run it as a 20 w. exciter and not see any ALC intervention.  I think I see it at around 30 watts and 120 to 130% positive modulation.

Rob
Logged

"Not taking crap or giving it is a pretty good lifestyle."--Frank
W1FVB
Member

Offline Offline

Posts: 77


VB Radio


WWW
« Reply #40 on: February 22, 2011, 02:29:57 PM »

I've tried this on my ic-735 and it doesn't make sense at all. If you limit the carrier to about 20 W with external ALC control, it limits the radio's TOTAL output to 20 W. No room for decent modulation what so ever.
Am I missing something here? or is it my 735

Frits
Logged

vbradio.wordpress.com
Steve - K4HX
Guest
« Reply #41 on: February 22, 2011, 03:19:47 PM »

Set the external ALC control so there is NO ALC on modulation peaks when you run 20 watts of carrier.
Logged
W1FVB
Member

Offline Offline

Posts: 77


VB Radio


WWW
« Reply #42 on: February 22, 2011, 03:54:15 PM »

Set the external ALC control so there is NO ALC on modulation peaks when you run 20 watts of carrier.

Thanks for clarifying Steve, I thought this topic explained on how to "easily" set the AM carrier level on a old rig like the 735.
Another way I've noticed to not have ALC action on modulation peak is to set the internal RF Max output
potmeter a little bit higher.

Frits

Logged

vbradio.wordpress.com
W2VW
Contributing
Member
*
Offline Offline

Posts: 3483


WWW
« Reply #43 on: February 23, 2011, 03:36:37 PM »

Set the external ALC control so there is NO ALC on modulation peaks when you run 20 watts of carrier.

Thanks for clarifying Steve, I thought this topic explained on how to "easily" set the AM carrier level on a old rig like the 735.
Another way I've noticed to not have ALC action on modulation peak is to set the internal RF Max output
potmeter a little bit higher.


Frits



That does work but now audio compression is really needed on ssb.
Logged
W1FVB
Member

Offline Offline

Posts: 77


VB Radio


WWW
« Reply #44 on: February 23, 2011, 04:00:04 PM »

Quote

That does work but now audio compression is really needed on ssb.


I have not noticed that, but then again I'm barely using it on sideband anyway.

Frits
Logged

vbradio.wordpress.com
Pete, WA2CWA
Moderator
Member

Offline Offline

Posts: 8146


CQ CQ CONTEST


WWW
« Reply #45 on: February 23, 2011, 05:43:37 PM »

Be aware that some of the earlier Icoms, the pre-DSP's, handled the internal ALC differently then some of the more recent rigs like the 756 series, 7600, 7700, and 7800. These newer rigs also don't use a balanced modulator. So, depending on the rig, this external voltage tweaking, may or may not result in useful results. Also, if using this voltage tweaking, and the rig is driving a linear amplifier, diligence should be done in monitoring the amplifier output to prevent over-driving the amplifier.
Logged

Pete, WA2CWA - "A Cluttered Desk is a Sign of Genius"
KW4DE
Member

Offline Offline

Posts: 20


« Reply #46 on: January 30, 2019, 08:06:15 PM »

Has anyone experimented with the external ALC mod on a 7300 or 7610?  Works well on older Icoms.
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 0.064 seconds with 19 queries.