SSB output overshoot
Locked
 
|
G'day,
When I run the K2/100 into my Acom 1000 I usually set the requested power to about 30 - 35W. I seldom drive the amp more than 400 - 500W output. On CW all is fine but on SSB I notice there is a severe overshoot on the output for the first syllable after changing to TX and after a few seconds pause in talking. With the K2/100 barefoot at 30W requested, on CW the RF bar is steady on 3 but on SSB it overshoots to 4-5 before settling at 3. At present the K2 is set up as twins with the KPA100 co-located with the KAT100. Anyone else noticed this? It would be interesting to put a scope on the output to monitor what happens to the RF level, unfortunately I don't have one. On 40M SSB problem is particularly severe, CW no problem. Regards, Mike VP8NO _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
Mike, I notice the opposite (but when running full power: 100-110W).
The first couple of syllables are lower, and then it kicks up. I'll try reducing power and see if it goes in the other direction. I was told that the previous effect was normal (action of the ALC). Bill W1PA K2/100 #61 _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Mike Harris-9
Same thing here, with or without an external amp. CW is as expected. But
when switching to SSB I will get full output (sometimes well over 100 watts) at any power setting over 11 watts and sometimes cause a shutdown of my power supply and a reboot of the K2/100. This behavior goes away when I press RF/ALC to switch the bar display to ALC. I have been told there is a mod that may cure the problem but I have not done anything. I just make a point of switching to ALC display whenever I operate SSB. Don K7FJ > G'day, > > When I run the K2/100 into my Acom 1000 I usually set the requested power > to about 30 - 35W. I seldom drive the amp more than 400 - 500W output. > > On CW all is fine but on SSB I notice there is a severe overshoot on the > output for the first syllable after changing to TX and after a few seconds > pause in talking. With the K2/100 barefoot at 30W requested, on CW the RF > bar is steady on 3 but on SSB it overshoots to 4-5 before settling at 3. > At present the K2 is set up as twins with the KPA100 co-located with the > KAT100. > > Anyone else noticed this? It would be interesting to put a scope on the > output to monitor what happens to the RF level, unfortunately I don't have > one. > > On 40M SSB problem is particularly severe, CW no problem. > > Regards, > > Mike VP8NO > Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Mike Harris-9
I think Don's observations sound like the problem I
and a few others have had. See, for example the symptoms described here: http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-03/msg00676.html http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-12/msg00408.html These two, at least, were due to the KSB2 ALC circuit having insufficient attenuation range to control the output of K2/100's which have above-average overall transmit gain on the lower bands. The fix involves adding a resistor and changing another in the KSB2. Gary Surrency at Elecraft was able to provide me the necessary details. But Mike's problem sounds much less severe. It might be a marginal case of this effect, I suppose. 73, Steve VE3SMA __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
Re: SSB output overshoot
|
|
Have you any thoughts about what might be causing Don's K2/100 to behave
when he switches the display from RF to ALC? I can see how the ALC might 'hunt' on SSB, but switching the display should not make any difference to ALC action I would have thought. 73, Geoff GM4ESD ----- Original Message ----- From: "Steve Kavanagh" <[hidden email]> To: <[hidden email]> Sent: Wednesday, November 01, 2006 1:55 PM Subject: Re: [Elecraft] SSB output overshoot >I think Don's observations sound like the problem I > and a few others have had. See, for example the > symptoms described here: > > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-03/msg00676.html > > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-12/msg00408.html > > These two, at least, were due to the KSB2 ALC circuit > having insufficient attenuation range to control the > output of K2/100's which have above-average overall > transmit gain on the lower bands. The fix involves > adding a resistor and changing another in the KSB2. > Gary Surrency at Elecraft was able to provide me the > necessary details. > > But Mike's problem sounds much less severe. It might > be a marginal case of this effect, I suppose. > > 73, > Steve VE3SMA _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Steve Kavanagh
Steve,
Thanks Steve.. I was hoping there was something I could do. Do you (or anyone) have the details of the necessary resistor changes? Don K7FJ ----- Original Message ----- From: "Steve Kavanagh" <[hidden email]> To: <[hidden email]> Sent: Wednesday, November 01, 2006 5:55 AM Subject: Re: [Elecraft] SSB output overshoot >I think Don's observations sound like the problem I > and a few others have had. See, for example the > symptoms described here: > > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-03/msg00676.html > > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-12/msg00408.html > > These two, at least, were due to the KSB2 ALC circuit > having insufficient attenuation range to control the > output of K2/100's which have above-average overall > transmit gain on the lower bands. The fix involves > adding a resistor and changing another in the KSB2. > Gary Surrency at Elecraft was able to provide me the > necessary details. > > But Mike's problem sounds much less severe. It might > be a marginal case of this effect, I suppose. > > 73, > Steve VE3SMA > Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Geoffrey Mackenzie-Kennedy-2
--- Geoffrey Mackenzie-Kennedy
<[hidden email]> wrote: > Have you any thoughts about what might be causing > Don's K2/100 to behave > when he switches the display from RF to ALC? Yes, that is quite unexpected ! David, M0DHO, brought this peculiarity to my attention and I noticed the same behaviour on my rig. I understand it may have to do with firmware differences between normal and ALC mode operation but I haven't seen a definitive answer. Nor do I know if Elecraft has dealt with this general issue in more recent production units. 73, Steve VE3SMA __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Don Ehrlich-2
G'day,
I have now done the resistor mod to no effect and will probably revert 'cos I don't like how I had to add the 1K0 in series with the base of Q1. Driving the K2/100 into a dummy load and just using the bar led display I see the following: Set requested power to 30W. Mode SSB. SSbA = 2, SSbC = 2-1, Mic Kenwood MC-43C. Mic jack grounding mod done. On 80M I see up to three bars ALC. Set to RF, regular peaks of 50W. On 40M I see up to seven bars of ALC. Set to RF, regular peaks of 50W and occasional peaks of 70W or more. On 30M I see up to six bars of ALC. Set to RF, regular peaks of 50W, occasional peaks to 60W. On 20M I see one bar of ALC. Set to RF, regular peaks to 40W. 18M - 10M, one bar of ALC. Set to RF, peaks to 40W. Set mode to CW. Request 30W. All bands power peaks to 30W, no overshoot. I dropped the requested power to 5W thus switching out the KPA100 and similar SSB overshoot problems presented. Conclusion, in it's present state my K2/100 is unusable on 40M SSB with an amp. If this is a common problem then it looks like the KSB2 is in need of some serious work unless the excess TX path gain in the 7 - 10MHz spectrum can be tamed. Maybe it also gives a clue to those reasonably frequent Hi Current warnings when running barefoot at 100W Regards, Mike VP8NO | Thanks Steve.. I was hoping there was something I could do. | Do you (or anyone) have the details of the necessary resistor changes? | | Don K7FJ | | | | ----- Original Message ----- | From: "Steve Kavanagh" <[hidden email]> | To: <[hidden email]> | Sent: Wednesday, November 01, 2006 5:55 AM | Subject: Re: [Elecraft] SSB output overshoot | | | >I think Don's observations sound like the problem I | > and a few others have had. See, for example the | > symptoms described here: | > | > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-03/msg00676.html | > | > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-12/msg00408.html | > | > These two, at least, were due to the KSB2 ALC circuit | > having insufficient attenuation range to control the | > output of K2/100's which have above-average overall | > transmit gain on the lower bands. The fix involves | > adding a resistor and changing another in the KSB2. | > Gary Surrency at Elecraft was able to provide me the | > necessary details. | > | > But Mike's problem sounds much less severe. It might | > be a marginal case of this effect, I suppose. | > | > 73, | > Steve VE3SMA _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Steve Kavanagh
G'day,
I haven't found any difference switching about from RF to ALC and back to my overshoot problem. Neither any effect by sneaking up on the requested power from a previously lower setting or down from a higher setting. Regards, Mike VP8NO | > Have you any thoughts about what might be causing | > Don's K2/100 to behave | > when he switches the display from RF to ALC? | | Yes, that is quite unexpected ! David, M0DHO, brought | this peculiarity to my attention and I noticed the | same behaviour on my rig. I understand it may have to | do with firmware differences between normal and ALC | mode operation but I haven't seen a definitive answer. _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Mike Harris-9
It's not as elegant as having an automatic circuit handle the issue, but one
way to reduce the amount of ALC and associated "overshoot" on 40 is to reduce the audio to the rig. Switch in the audio attenuator at the MENU or otherwise reduce the audio to the mic jack until you are getting just one bar of ALC flickering, or no ALC bar at all, on voice peaks. Ron AC7AC -----Original Message----- G'day, I have now done the resistor mod to no effect and will probably revert 'cos I don't like how I had to add the 1K0 in series with the base of Q1. Driving the K2/100 into a dummy load and just using the bar led display I see the following: Set requested power to 30W. Mode SSB. SSbA = 2, SSbC = 2-1, Mic Kenwood MC-43C. Mic jack grounding mod done. On 80M I see up to three bars ALC. Set to RF, regular peaks of 50W. On 40M I see up to seven bars of ALC. Set to RF, regular peaks of 50W and occasional peaks of 70W or more. On 30M I see up to six bars of ALC. Set to RF, regular peaks of 50W, occasional peaks to 60W. On 20M I see one bar of ALC. Set to RF, regular peaks to 40W. 18M - 10M, one bar of ALC. Set to RF, peaks to 40W. Set mode to CW. Request 30W. All bands power peaks to 30W, no overshoot. I dropped the requested power to 5W thus switching out the KPA100 and similar SSB overshoot problems presented. Conclusion, in it's present state my K2/100 is unusable on 40M SSB with an amp. If this is a common problem then it looks like the KSB2 is in need of some serious work unless the excess TX path gain in the 7 - 10MHz spectrum can be tamed. Maybe it also gives a clue to those reasonably frequent Hi Current warnings when running barefoot at 100W Regards, Mike VP8NO | Thanks Steve.. I was hoping there was something I could do. Do you (or | anyone) have the details of the necessary resistor changes? | | Don K7FJ | | | | ----- Original Message ----- | From: "Steve Kavanagh" <[hidden email]> | To: <[hidden email]> | Sent: Wednesday, November 01, 2006 5:55 AM | Subject: Re: [Elecraft] SSB output overshoot | | | >I think Don's observations sound like the problem I | > and a few others have had. See, for example the | > symptoms described here: | > | > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-03/msg00676.html | > | > http://www.ac6rm.net/mailarchive/html/elecraft-list/2005-12/msg00408.html | > | > These two, at least, were due to the KSB2 ALC circuit having | > insufficient attenuation range to control the output of K2/100's | > which have above-average overall transmit gain on the lower bands. | > The fix involves adding a resistor and changing another in the KSB2. | > Gary Surrency at Elecraft was able to provide me the | > necessary details. | > | > But Mike's problem sounds much less severe. It might | > be a marginal case of this effect, I suppose. | > | > 73, | > Steve VE3SMA _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
G'day,
| It's not as elegant as having an automatic circuit handle the issue, but one | way to reduce the amount of ALC and associated "overshoot" on 40 is to | reduce the audio to the rig. Switch in the audio attenuator at the MENU or | otherwise reduce the audio to the mic jack until you are getting just one | bar of ALC flickering, or no ALC bar at all, on voice peaks. This I have tried. Setting the gain to SSbA = 1 reduces the overshoot at the 70W+ level and reduces the ALC to three bars. However, it still kicks up to twice the requested power on occasion. Not good for my GU74 cathode that. So to change from 20M to 40M, maybe in the heat of a contest, I now have to press Band- twice, tap MENU, tap up to SSbA, hold MENU to select the SSbA parameter, tap down from 2 to 1, exit MENU, tune the linear, remember to select a band variable power level much less than I need for CW. Not the best option methinks, not with modern gear at least, in the old days it was normal to be acting like a one armed paper hanger, not now. Band decoders, automatic switching etc. Maybe I should buy one of those expensive mikes that don't seem to give enough output, that would do it. The curse of having one of the few K2's on the planet at has a surplus of AF gain. Regards, Mike VP8NO _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
In reply to this post by Mike Harris-9
At 22:55 02/11/2006, Mike Harris wrote:
>G'day, > >I have now done the resistor mod to no effect and will probably revert >'cos I don't like how I had to add the 1K0 in series with the base of Q1. Hi Mike, ISTR making the ALC *very* much more aggressive to tackle this, especially on 80m. I think I cut the track to include a resistor to the Q1 base and changed another resistor. I can't remember the details (such as the values I ended up using) since it was more than a year ago that I was experimenting with this. When I next get a chance, I'll open up my K2 to refresh my memory. Since I made those changes, I don't see any significant difference in SSB output levels between bands and it's reasonably well tamed. I do find the peak RF on SSB higher than the keyed RF level on CW, but that applies to all bands. I've just got used to turning the requested power control down a bit to keep within the 400W UK legal limit with the K2 driving an Acom 1000, so it's not really been a problem. You might want to experiment with more aggressive ALC around the Q1 base on the KSB2 to see if you can tame it. BTW, I have never seen a good explanation that includes an overview of the firmware's part in the ALC control. There appear to be two separate control loops here. With the later firmware changes to the main PIC, one of those control loops only appears to be active in ALC meter mode. With earlier (pre KPA100) firmware versions, it would apply in RF meter mode, albeit being polled much less frequently, and in ALC meter mode with very frequent AUXBUS operation. I had found that by monitoring the AUXBUS with a scope in both modes and both firmware versions. Elecraft were unaware of this difference but Wayne was too busy to respond in detail at the time. The other control loop is based on RFDET being fed back and compared with a DAC level using the PIC on the KSB2. This seems to be an area of the K2 design that would be good to have a decent explanation of including a summary pseudo-code algorithm of what the firmware is doing to control the ALC. I have yet to see a detailed description how BOTH of these control loops work, so if anyone has any more details on this, I'd appreciate the wisdom being shared :-) Good luck, 73, David M0DHO _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
Re: SSB output overshoot
|
|
On Friday, November 03, 2006 at 3:27 PM, David M0DHO wrote:
<snip> >.I have yet to see a detailed description how BOTH of these control loops >work, so if anyone has any more details on this, I'd appreciate the wisdom >being shared :-) ----------------------------------------------------------- Likewise. I have never been too comfortable with RFDET having two sources tied together, one source at the output of the K2 and the second at the output of the KPA100. 73, Geoff GM4ESD _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
|
|
Geoff,
Your question is a lot easier than the control loop question. The series resistor R66 on the RF board will limit the response of the basic detector in the base K2 when the other devices are installed - consider that the KPA100 (and other options that provide the VRFDET signal) use an op amp output with no series resistor - all options other than the KAT2 use a tri-state device so it can be enabled or disabled under control of the microprocessor(s). This is a bussing technique as opposed to simply connecting several outputs together. 73, Don W3FPR > -----Original Message----- > >.I have yet to see a detailed description how BOTH of these > control loops > >work, so if anyone has any more details on this, I'd appreciate > the wisdom > >being shared :-) > > ----------------------------------------------------------- > > Likewise. I have never been too comfortable with RFDET having two sources > tied together, one source at the output of the K2 and the second at the > output of the KPA100. > > 73, > Geoff > GM4ESD > _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
Re: SSB output overshoot
|
|
Don,
Thank you for putting me straight on this. On the control loop question I agree with David, M0DHO, that an explanation of the workings of the ALC system would be useful not only in its dynamic state but also in its static state. 73, Geoff GM4ESD ----- Original Message ----- > Geoff, > > Your question is a lot easier than the control loop question. > The series resistor R66 on the RF board will limit the response of the > basic > detector in the base K2 when the other devices are installed - consider > that > the KPA100 (and other options that provide the VRFDET signal) use an op > amp > output with no series resistor - all options other than the KAT2 use a > tri-state device so it can be enabled or disabled under control of the > microprocessor(s). > > This is a bussing technique as opposed to simply connecting several > outputs > together. > > 73, > Don W3FPR > _______________________________________________ Elecraft mailing list Post to: [hidden email] You must be a subscriber to post to the list. Subscriber Info (Addr. Change, sub, unsub etc.): http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/subscribers.htm Elecraft web page: http://www.elecraft.com |
«
Return to Elecraft
|
1 view|%1 views
| Free forum by Nabble | Edit this page |