AGC Design of the K3
Locked
 
|
I noticed a remedy for the following problem on the K3 enhancement list -
Hardware AGC threshold change Some users prefer tuning the band with a wide filter bandwidth selected (for example, using a 2.7 or 2.8 kHz SSB filter in CW mode rather than a 500 Hz or narrower filter). Strong off-frequency signals (above about S9+5) that are inside the crystal filter passband can then activate the hardware AGC circuit, reducing the level of desired signals. Shouldn't a radio's AGC only use signals within the selected final IF passband width to control gain? Wasn't this kind of AGC "pumping" described above a problem in older radios when DSP initially featured and the AGC wasn't included within the DSP functionality? That is, AGC would operate using circuitry prior to the DSP, users would then narrow the passband width using the DSP but then found that signals outside this passband would reduce the desired signal. |
|
|
No because the DSP stage doesn't have enough dynamic range to handle the full 140 dB range of possible input signals. There are two AGCs in the K3, Orion, etc...analog ahead of the DSP stage and digital within the DSP. The analog AGC ahead of the DSP is needed to protect the ADC from overload and the digital can then handle about 100 dB internally. 73, Bill W4ZV |
|
Administrator
|
In reply to this post by VK5ABQ
Shane White wrote:
> I noticed a remedy for the following problem on the K3 enhancement > list - > > Some users prefer tuning the band with a wide filter bandwidth > selected (for > example, using a 2.7 or 2.8 kHz SSB filter in CW mode rather than a > 500 Hz > or narrower filter). Strong off-frequency signals (above about S9+5) > that > are inside the crystal filter passband can then activate the hardware > AGC > circuit, reducing the level of desired signals. > > Shouldn't a radio's AGC only use signals within the selected final IF > passband width to control gain? Shane, The K3 has two kinds of AGC: DSP AGC, which is effective up to about S9+20 or +30, and hardware AGC, which only kicks in above this level to protect the A-to-D converter. This is very typical for DSP-based radios. Ideally, the operator selects the narrowest crystal filter they can, even though the DSP can be further narrowed within that passband. The K3 does this for you as you adjust WIDTH and other DSP controls. The majority of radios on the market today provide only a wide "roofing" filter at the first I.F. These are rarely any narrower than 6 kHz -- sometimes 3 kHz. This doesn't help protect the front end if you wanted a narrower passband for CW, narrow SSB, or data modes. The K3 helps solve this problem in three ways: 1. It uses a very low first I.F. (8.215 MHz), where crystal filters as narrow as 200 Hz are easily fabricated. (Most other rigs have a first I.F. in the low VHF range.) The K3 has a very high-dynamic-range front end to take advantage of this design strategy. As a result, it's #1 on the Sherwood Engineering receiver test site, and is the only rig for which the ARRL has measured over 100 dB IMDDR3 at close spacings. 2. It provides slots for up to five first-I.F. crystal filters for the main receiver and another five for the subreceiver. (You could call these roofing filters, if you like this term -- I don't, since they're far higher performance than the usual filters so designated.) Thus you can put in a variety of close-spaced filters to optimize for a given mode. As I mentioned, the WIDTH and other controls then select the optimal filter as you adjust them. 3. Later this year we'll be offering variable-passband crystal filters that each cover a significant bandwidth range, e.g. 200-400, 400-800, etc. (These are not finalized.) Each of these filters will have 8 discrete bandwidths. As you adjust the DSP controls, the K3 will again optimize the passband of the selected filter. 73, Wayne N6KR --- 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 |
|
|
In reply to this post by wayne burdick
Wayne, I don't understand the concept of variable passband filters. Surely the K3 will just choose a filter based on the current operating frequency, and that filter will be used over a range of frequencies. How are these new filters you describe different?
|
|
Administrator
|
The K3 selects a filter based on the user's WIDTH setting, which
usually varies with the mode of operation. Typically you might select 50 Hz to 1000 Hz WIDTH while using CW mode. We offer several crystal filters in this range (200, 250, 400, 500, and 1000), and if you have more than one installed, the K3 will select the optimal one automatically. The idea of a variable-passband crystal filter is that it could replace two or more fixed-bandwidth filters. For example, if we made one with a BW of about 200-500 Hz, it could replace up to four crystal filters (in practice, it would be replacing two or three at most). This would free up one or two slots for other fixed- or variable-bandwidth SSB filters. Since this variable-passband filter has 8 steps of bandwidth, we can optimize automatically it to be equal to or just a bit higher than the selected DSP bandwidth (i.e., the WIDTH control setting). So you could think of them as "DSP-tracking crystal filters." Wayne N6KR On May 12, 2008, at 7:23 PM, Shane White wrote: > > Wayne, I don't understand the concept of variable passband filters. > Surely > the K3 will just choose a filter based on the current operating > frequency, > and that filter will be used over a range of frequencies. How are > these new > filters you describe different? > > > wayne burdick wrote: >> >> >> >> 3. Later this year we'll be offering variable-passband crystal filters >> that each cover a significant bandwidth range, e.g. 200-400, 400-800, >> etc. (These are not finalized.) Each of these filters will have 8 >> discrete bandwidths. As you adjust the DSP controls, the K3 will again >> optimize the passband of the selected filter. >> >> 73, >> Wayne >> N6KR >> >> > > -- > View this message in context: > http://www.nabble.com/AGC-Design-of-the-K3-tp17199520p17199832.html > Sent from the Elecraft mailing list archive at Nabble.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 > > --- 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 |
|
|
Very nice.
Will this upgrade be able to be installed on older model K3s? |
|
Administrator
|
Yes. There is no option or upgrade that only works on newer K3s. They
can all be upgraded, and always will (Elecraft company policy). 73, Wayne N6KR On May 12, 2008, at 8:26 PM, Shane White wrote: > > Very nice. > Will this upgrade be able to be installed on older model K3s? > > --- 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 |
|
|
Cool. Nice to see logical thinking reign for a change.
|
|
|
In reply to this post by wayne burdick
I like the description "DSP-tracking crystal filters", it really
says what they will do. 73 Stewart G3RXQ On Mon, 12 May 2008 19:36:16 -0700, wayne burdick wrote: > The K3 selects a filter based on the user's WIDTH setting, which > usually varies with the mode of operation. Typically you might select > 50 Hz to 1000 Hz WIDTH while using CW mode. We offer several crystal > filters in this range (200, 250, 400, 500, and 1000), and if you have > more than one installed, the K3 will select the optimal one > automatically. > > The idea of a variable-passband crystal filter is that it could replace > two or more fixed-bandwidth filters. For example, if we made one with a > BW of about 200-500 Hz, it could replace up to four crystal filters (in > practice, it would be replacing two or three at most). This would free > up one or two slots for other fixed- or variable-bandwidth SSB filters. > > Since this variable-passband filter has 8 steps of bandwidth, we can > optimize automatically it to be equal to or just a bit higher than the > selected DSP bandwidth (i.e., the WIDTH control setting). So you could > think of them as "DSP-tracking crystal filters." > > Wayne > N6KR > > On May 12, 2008, at 7:23 PM, Shane White wrote: > >> >> Wayne, I don't understand the concept of variable passband filters. >> Surely >> the K3 will just choose a filter based on the current operating >> frequency, >> and that filter will be used over a range of frequencies. How are >> these new >> filters you describe different? >> >> >> wayne burdick wrote: >>> >>> >>> 3. Later this year we'll be offering variable-passband crystal filters >>> that each cover a significant bandwidth range, e.g. 200-400, 400-800, >>> etc. (These are not finalized.) Each of these filters will have 8 >>> discrete bandwidths. As you adjust the DSP controls, the K3 will again >>> optimize the passband of the selected filter. >>> >>> 73, >>> Wayne >>> N6KR >>> >>> >> -- >> View this message in context: >> l >> Sent from the Elecraft mailing list archive at Nabble.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 >> >> > --- > > 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 _______________________________________________ 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 VK5ABQ
Wayne, as you know from my private correspondence, I think the variable filters are the best way to go on this, as there are just not enough slots to get to an adequate level of protection for anyone who uses a lot of modes. Another example of thinking outside the box. I LOVE the use of the term "tracking filter", BTW. You should use that term in all future references to the new filters. Any idea when they might be available? 73, Larry N8LP > Date: Mon, 12 May 2008 19:36:16 -0700 > From: wayne burdick <[hidden email]> > Subject: [Elecraft] Re: AGC Design of the K3 > To: Shane White <[hidden email]> > Cc: [hidden email] > Message-ID: <[hidden email]> > Content-Type: text/plain; charset=US-ASCII; format=flowed > > The K3 selects a filter based on the user's WIDTH setting, which > usually varies with the mode of operation. Typically you might select > 50 Hz to 1000 Hz WIDTH while using CW mode. We offer several crystal > filters in this range (200, 250, 400, 500, and 1000), and if you have > more than one installed, the K3 will select the optimal one > automatically. > > The idea of a variable-passband crystal filter is that it could replace > two or more fixed-bandwidth filters. For example, if we made one with a > BW of about 200-500 Hz, it could replace up to four crystal filters (in > practice, it would be replacing two or three at most). This would free > up one or two slots for other fixed- or variable-bandwidth SSB filters. > > Since this variable-passband filter has 8 steps of bandwidth, we can > optimize automatically it to be equal to or just a bit higher than the > selected DSP bandwidth (i.e., the WIDTH control setting). So you could > think of them as "DSP-tracking crystal filters." > > Wayne > N6KR _______________________________________________ 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 |
|
|
Hi,
I am very curious about the variable filters, a concept which has been proved in the K2. However I also wonder if the non-linearity of the diodes (and on top of that the crystals) will not cause more problems in the K3 environment than they solve. In many modern receiver concepts with very strong front ends, the crystals in the filters have become the limiting factor. I wonder if the 8 steps, which Wayne mentioned, mean that diodes will not be used but switched Cs... It will be very, very interesting to see what Elecraft comes up with. If the IMD behavior is equal to that of the current filters, this will be really cool. Fingers crossed, vy 73 de toby _______________________________________________ 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 N8LP
I also like the term DSP Tracking Filter very much and wish it had been
used initially instead of "Roofing Filter" so the K3 IF filtering technique would not have been confused with those receivers using a VHF first IF. 73, Don W3FPR Larry Phipps wrote: > > Wayne, as you know from my private correspondence, I think the > variable filters are the best way to go on this, as there are just not > enough slots to get to an adequate level of protection for anyone who > uses a lot of modes. Another example of thinking outside the box. I > LOVE the use of the term "tracking filter", BTW. You should use that > term in all future references to the new filters. Any idea when they > might be available? > > 73, > Larry N8LP > > 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 |