Sandy: >Tom - I'm a little confused by Wayne's last posting on the OPT1 filter not >being affected by the filter mod. Your very nice pictures were all labeled >OPT1, and the SSB manual, Table 3, states the 2nd filters should be OPT1 for >USB and LSB, albeit tuned very slightly differently on USB. So what does this >mod affect? FL3 only? What is the difference between the 2nd crystal filter >and FL2/OPT1 in a standardly tuned SSB? Somewhat confused in Maryland - The 2nd Xfil (Crystal Filter) is located ON the RF Board, not on the SSB Option board. It is in the RX line ONLY (NOT in the TX line). When in CW mode, the 2nd Xfil adds a bit of selectivity to the K2, however, assuming you have tuned L34 as you were instructed to do in the basic K2 manual, when you are in SSB mode, the 2nd Xfil not only does not 'contribute' positively to SSB reception, it actually tends to degrade the passband response of the SSB filter if it's been installed. The 2nd Xfil mod basically pads the 2nd Xfil so it no longer rolls off one side of the SSB option filter when it's in line. And since it's switched on ONLY when the SSB filter is selected, it ONLY effects the K2 when OPT1 is selected for receive. Wayne WANTS all K2 users to align their SSB FIL1 OPT1 filter 'by the book'...! No questions asked... period! He wants this because he is concerned that someone may not follow the manual docs and will wind up with their BFO so far out of whack (a midwest technical term meaning FUBARed) that they'll have carrier inside the filter passband, or so far outside the passband that the TX SSB signal quality will be severely compromised. THEN... Wayne wants users to set up SSB FIL2 and OPT1 however they want it set up (e.g. using Spectrogram or any other method you choose). What this does is that it allows you to specially set up FIL2 as YOU want it configured... then, since FIL1 OPT1 is _ALWAYS_ used on SSB transmit (no matter what FILn you use in SSB mode), you are assured that your SSB TX signal is properly aligned (remember, 2nd Xfil is OUT of the line on TX), and you can still have your 'personalized' SSB filter setting in use when you have FL2 OPT1 selected. In actual practice, I've found no significant difference in the setting of FIL1 OPT1 regardless of whether I used the MANUAL defaults for the BFO or whether I used Spectrogram to set the BFO, but Wayne's the boss, and since it IS his mod, I figger it's best to do it his way. I guess there is a chance that someone could really screw things up using Spectrogram and wind up with their BFO so far out of whack that their SSB TX signal could really be messed up... so I think his method is probably best. I hope this helps to better 'splain why both FIL1 and FIL2 are specified as OPT1. The 2nd Xfil mod has NO effect on SSB RX when the CW filter is used for receive. 73 - Tom Hammond N0SS +++++++++++++++++++++++++++++++++++++++++++++++++ Date: Thu, 19 Apr 2001 10:16:57 -0500 From: Tom Hammond =?iso-8859-1?Q?N=D8SS?= Subject: Re: [Elecraft] Spectrogram Question Joel: >My question regards WHEN TO USE spectrogram. From the Builer's Resources >section on the Elecraft website, I gather that the appropriate place to >use Spectrogram is in the section of teh K2 manual entitled "BFO >Alignment" on page 60. You probably won't need to use Spectrogram until you complete assembly. The defaults given in the manual should be adequate to enable you to get the K2 on the air and to begin playing with it... and this is good, as it allows you to get a 'feel' for the rig and the controls before you get down to the real 'business' of doing the final filter alignment. >Are there any other sections, other than "BFO Alignment" on page 60, where >the use of Spectrogram is recommended? None officially published, but there are a couple different articles available from the Builder's Resource page on the Elecraft site. Go to: http://www.elecraft.com/Apps/new_fil_docs/k2_cw_gram.pdf or http://www.elecraft.com/Apps/new_fil_docs/k2_ssb_gram.pdf depending upon whether you wish to align the CW or the SSB filters. >Your helpful comments are greatly appreciated. I want to do this right the >first time! NOTE that if you will be installing the SSB option board, you are strongly urged to also perform the 2nd Xfil SSB mod which is described at: http://www.elecraft.com/Apps/new_fil_docs/k2_2nd_xfil_ssb_mod.pdf This is a VERY SIMPLE mod and uses one resistor (3.3k to 3.9k), two capacitors (.01uF and 120pF), and a generic small signal PNP transistor (PN2222, 2N3904, 2N4124, etc.). The instructions are step-by-step and include photos. 73 - Tom Hammond N0SS +++++++++++++++++++++++++++++++++++++++++++++++++++ Date: Fri, 08 Jun 2001 05:53:56 -0000 From: "Marinos Markomanolakis, M.D." Subject: Re: [Elecraft] SSB narrow filter performance improvement. John, Thanks for the info about the CW filters being used with SSB. As a matter of fact, I do have the KSB2 option installed, but I was using the narrow filter settings (based on the recommended spectrogram adjustment) centered at the same frequency as the OP1 filter, a fact that due to their wide skirts was not succeeding in any significant QRN rejection. Eric Swartz, in his last email suggested to leave the bandwidth at OP1 and move the BFO frequency up or down in the different filter options so that you obtain a pass band tuning effect. I tried it tonight and indeed it seems to make a good difference in the QRN rejection with some sacrifice of course in the sound fidelity of the respective filters. So I ended up setting the different filter options one for high end cut, the other for low end cut leaving the OP1 filter centered properly as per the spectrogram. I am still not completely finished with all the filter adjustments (USB and LSB) but I think I am in the right tract. Many thanks to Eric, you John, Tom Hammond and all the Elecrafters with their good ideas and the willingness to help. Marinos, sv9dru/w6 >From: "John Grebenkemper, KI6WX" >To: "Marinos Markomanolakis, M.D." >Subject: Re: [Elecraft] SSB narrow filter performance improvement. >Date: Thu, 07 Jun 2001 18:37:32 -0700 > > >Marinos; >When I was measuring the CW filters, I also made some measurements on the >wide bandwidth mode of the CW filter and the SSB filter. Basically, the CW >filter run in wide bandwidth does not have a very good skirt response. >Putting the variable bandwidth option on the filter doesn't allow you to >tune it properly to get both a narrow bandwidth and a wide bandwidth. The >KSB2 SSB filter is optimized for SSB, and has a 2 kHz bandwidth with much >steeper skirt selectivity. If you really want better SSB reception, you'll >need to order a KSB2. >-John > KI6WX > > > Since there is considerable interest about simple mods intended to >improve > > the performance of the K2 CW filters,which is excellent anyway, I was > > wondering if anyone has ideas about getting better rejection from the >narrow > > filters at the SSB part of the rig. > > > > At least for mine (SN 1659), although the wide filter setting gives the > > expected width ~2.4Khz, change to narrower settings e.g. FL2=2.01, >FL3=1.80 > > or even FL4=1.70Khz settings, does not significantly alter the rejection >of > > unwanted high or low end signals. > > It does change the audio pitch, but the QRN remains. > > > > I have come to the conclusion that these narrow filters have rather wide > > skirts, a fact that makes them essentially unusable... > > > > Now do not get me wrong, I love my K2 and I really believe that it is a > > superb performer, but it can certainly lents itself to improvements >based >on > > experimentation. That is its major advantage over the commercial rigs as >far > > as I am concerned anyway. > > > > 73 to all, > > Marinos sv9dru/w6 Moving back to SV9 land in less than 2 months. ++++++++++++++++++++++++++++++++++++++++++++++ Date: Thu, 7 Jun 2001 23:59:30 -0700 From: "Ron D' Eau Claire" Subject: RE: [Elecraft] SSB narrow filter performance improvement. Marinos, sv9dru/w6 wrote: > Eric Swartz, in his last email suggested to leave the bandwidth > at OP1 and > move the BFO frequency up or down in the different filter options so that > you obtain a pass band tuning effect. > > I tried it tonight and indeed it seems to make a good difference > in the QRN > rejection with some sacrifice of course in the sound fidelity of the > respective filters. Did you also note that he pointed out that without the SSB option you will NOT get especially good skirts at SSB bandwidths? The OPT1 filter on the SSB module is designed to provide good skirts at the wide bandwidths needed for SSB, while the CW filter on the r-f board are designed for narrower bandwidths. The CW filters can be set for a wide bandwidth in order to listen to SSB, but they will not have the skirt selectivity of the OPT1 filter on the SSB board. In order to realize the selectivity the K2 is capable of using SSB bandwidths, you need the filters on the SSB mod itself. Ron AC7AC K2 # 1289 +++++++++++++++++++++++++++++++++++++++++++ Date: Fri, 8 Jun 2001 07:57:06 -0600 From: "Santa Fe" Subject: [Elecraft] ssb narrow filtration My method, for what it is worth, was to roughly adjust the filters for SSB while looking at Spectrogram, THEN... I tuned in an SSB station and LISTENED to it. While listening, I adjusted both the width and the BFO. I found that moving the low frequency skirt up was not a good idea, but moving the upper skirt down was. I also found that, because of the bumpy filter peaks, some widths sound better than others. Instead of deciding before hand, OK, let's get 2.0 and 1.8 for example, instead I ignored that and just tuned around. I found very unround number widths provided "pretty good" fidelity at narrow widths, and kept those. The selected widths proved to be 2.12 and 1.69 on my radio, but am certain that someone else's will sound best at other random widths. 73, FD w5ya / kt5x ++++++++++++++++++++++++++++++++++++++ Date: Fri, 8 Jun 2001 09:45:34 -0700 From: "Ron D' Eau Claire" Subject: RE: [Elecraft] Mod to K2 CW Filter To Improve Rejection John, KI6WX wrote: > I used both versions 6.0.8 and 5.1.6 and found no difference in their > performance and results. I'm not convinced that it makes any difference > which version you use. I've had the same experience, at least as far as the noise floor is concerned. However, there a significant difference between computers. My five-year-old NEC laptop shows a noise floor of about 83 dB, while my new IBM Thinkpad is clean all the way down to the Spectrogram limit of 90 dB. I would suggest that anyone looking at what's happening way down there near that noise floor to first run Spectrogram with all the inputs to their computer disconnected and muted using the Windows audio controls and see where the sound floor is for that computer. I say to mute the inputs because I have noticed that even an input with nothing connected to it can generate some noise at those very low levels if that input is enabled by the Windows audio controls. And then, of course, there are those 'built in' microphones in most notebooks that become active automatically unless you specifically turn them off . Ron AC7AC K2 # 1289 +++++++++++++++++++++++++++++++++++++ Date: Sat, 09 Jun 2001 00:52:50 -0000 From: "Marinos Markomanolakis, M.D." Subject: RE: [Elecraft] SSB narrow filter performance improvement. Hi to all, I did experiment with all the possible filter bandwidths, having the BFO frequency moved up or down for the different filters (i.e. FL2, 3 and 4) and keeping the BFO frequency centered (per the spectrogram)for best audio. Here is how I did it last night: 1)I do not have a signal generator, so I chose strong SSB signals and then moved the frequency of the receiver 2Khz above and below them, essentially mimicking the situation when you have interference from strong nearby signals when you are trying to listen to a weak one. 2)Then, I adjusted the BFO in a way that the interference essentially disappeared. 3)I returned to the exact frequency of the SSB signal with the new BFO settings, and I played with the bandwidth of the filter, in order to obtain the best possible intelligibility. During the last step, I found that the best possible bandwidth for these displaced BFO filters was not the OP1, but rather narrower, which depended on how much the BFO displacement was for that specific filter. Below, I provide my exact BFO settings combined with the respective bandwidths, realizing that they may vary between different K2s depending on the exact frequency of their crystals. My KSB2 crystal calibration # is 3.7 and their respective frequencies are 4913.6 for LSB and 4916.3 for USB. LSB: USB: Bandwidth BF# BFO Freq Bf# BFO freq FL1 OP1 135 4913.67 215 4916.5 FL2 2.00 105 4912.75 220 4916.65 FL3 1.80 130 4913.50 185 4915.45 FL4 1.50 116 4913.06 175 4915.07 With the above adjustments, I have the FL2 committed for low end cut and the FL3, FL4 committed for increasingly tighter high cut. I was amazed last night to be able to copy 100% stations with signal (0)transmitting just 2Khz below an S9 station using the FL4 or even the FL3 with the S9 guy barely perceptible, while with the OP1 setting the weak station was totally covered by the "highs" of the strong one. So it turns out that if you use those filters properly, they will do an amazing job at the SSB too, and if you combine the good rejection with the K2's lack of desensitization from near by strong signals, you get a result hard to beat. I hope others can use the above data or follow these steps to get some performance improvements.... Marinos, sv9dru/w6 K2# 1659 ++++++++++++++++++++++++++++++++++++++++++ Date: Sat, 09 Jun 2001 06:43:29 -0400 From: Charles Greene Subject: RE: [Elecraft] SSB narrow filter performance improvement. At 12:52 AM 6/9/2001 +0000, Marinos Markomanolakis, M.D. wrote: Hi Marinos, I tried to set the filter settings you recommended but found that my K2 doesn't respond as yours. For example for FL1 LSB, I had to set the BF# at 122 giving a BFO Freq of 4913.39 although the USB setting was ok. I just set your USB setting then matched the LSB using Spectrogram. So if you could give the upper and lower frequency cutoff using Spectrogram for each FL, them I could adjust my filters to meet those frequencies goals rather than try to use your settings. It's a very good idea, and we need to pursue it farther and come up with an adjustment procedure that will apply to all K2's. Some may be closer to yours than others, particularity if they have any mods to the original SSB circuit. Mine is SN #462, and I haven't made any filter mods; yet. TNX, >Hi to all, > >I did experiment with all the possible filter bandwidths, having the BFO >frequency moved up or down for the different filters (i.e. FL2, 3 and 4) >and keeping the BFO frequency centered (per the spectrogram)for best audio. +++++++++++++++++++++++++++++++++++++++++++++++ Date: Sun, 10 Jun 2001 08:22:55 EDT From: WA9GQT at aol.com Subject: [Elecraft] Re: SSB Filter Settings Eric Wrote: If you set the SSB filters to OP1 for all four SSB settings you will be using this bandwidth for each setting. You can then move the BFO for each filter setting to change the filter center (pass band tuning in effect.) I agree with Eric: I have done the 2nd Xfil Mod and have checked my filter settings for FL2 thru FL4 with spectrogram. All my filter positions are set on OP1 for all SSB filters. I then adjusted bfo freq for FL3 & FL4 to give good sound characteristics while being narrower. FL2 was left at spectrogram setting. I would not suggest using CW filters for SSB reception since they make the human voice sound very strange. I like the way SSB sounds using OP1 for all filters. 73, Rod WA9GQT K2 #1900 ++++++++++++++++++++++++++++++++++++++++++++ Date: Sun, 10 Jun 2001 21:22:01 -0000 From: "Marinos Markomanolakis, M.D." Subject: [Elecraft] Re: SSB filter performance and more. Jay, Without being an expert on electronics and after doing some limited reading of the ARRL handbook and the K2 manual, I came to the understanding that the radio signal (RF) in the (k2) receiver after being tuned by the antenna/front end circuit is basically being subjected to 2 changes before it can be demodulated into AF voice or audible CW tone : First it goes through the IF mixer where it is combined with the VCO generated signal, which will always be either (RF + IF) for the bands 17m and below, or (RF - IF) for the bands above 17m, where IF is 4.915Mhz. After these 2 signals are combined in the mixer, you have 2 possible signal combinations coming out of it one being the sum and the other being the difference of the 2. If we examine the case for the lower bands (below 17m) it will be either RF+(RF+IF) or RF-(RF+IF). >From those 2 possible outputs, with filtering, only the 2nd one is selected which is exactly equal to the IF(4.915Mhz). Now, we have the SSB signal carried within that IF frequency which then is subjected to a 2nd mixing process, this time with the frequency generated in the BFO (4913.6Mhz for LSB and 4916.3Mhz for USB demodulation) which is slightly different than the IF frequency. The purpose of this (with a process same as described above for the 1st mixer) is essentially to "strip" the AF SSB signal out of the IF carrier. That final signal product comes to the AF amplifier an then to the speaker.By adjusting the BFO frequency, essentially you adjust the frequency of the product coming out of the BFO mixer into the AF section. Now it seems that different K2s have slightly different BFO frequencies depending on the kind of crystals that are supplied with that particular unit, which have been given different code numbers e.g.3.7 etc, in order for the builder to recognize the exact frequency of his K2's BFO, and be able to adjust the different BFO frequency settings for proper demodulation. I hope that the above simplistic and probably not entirely accurate explanation helps. More comments or corrections are certainly welcome from any more knowledgeable fellow "Elecrafter" 73, Marinos, sv9dru/w6 >From: "Jay Rutherford" >Reply-To: k1uc at amsat.org >To: "Marinos Markomanolakis, M.D." >Subject: SSB narrow filter performance improvement. >Date: Sun, 10 Jun 2001 11:06:49 -0400 > >Hi Marinos, > >Could you explain to me the crystal calibration number you wrote >about in a recent email on SSB filter settings? I've been trying for >over a year to determine the "center frequency" or "IF frequency of >my rig to understand the relationship between the IF and the actual >BFO readings. > >I'm interested in where you got that number and how you applied it >to the nominal 4.915 MHz IF. > >Thanks & 73, Jay > >Marinos Markomanolakis, M.D. wrote: > > > My KSB2 crystal calibration # is 3.7 and > > their respective frequencies are 4913.6 for LSB and 4916.3 for USB. > > > >K1UC in Waterbury Center, Vermont, USA >GRID FN34 Email: K1UC at AMSAT.ORG K2#1034 ++++++++++++++++++++++++++++++++++++ Date: Sun, 11 Nov 2001 07:02:19 -0600 From: Tom Hammond =?iso-8859-1?Q?N=D8SS?= Subject: Re: [Elecraft] Question on 2nd Xtal filter mod for SSB David: >The directions to the mod suggest that one may choose to tweak the BFO >settings for FL2/OP1. I realize that this is entirely subjective, but I >don't really see the need to fool with the BFO settings. So finally, here's >my question... do most K2 owners, upon doing this mod, feel the need to >change the settings? Is it usual for them to be well out of whack after >doing the mod or not? They should not out of whack following the mod, however... Before performing the mod, when many builders initially set their BFO for SSB receiption (when the SSP Option IS installed), they may attempt to compensate for the lack of highs (or lows, depending upon the band/SSB-mode being used when the sdjustments are made) by tweaking the BFO too far in one direction ofr the other. Once the SSB filter has been compensated for its original 'tilt' in passband flatness, they may find that additional tweaking of the BFO is required to better center the received signals within the filter passband. 73 - Tom Hammond N0SS ++++++++++++++++++ Date: Sun, 11 Nov 2001 06:50:07 -0700 From: "=?iso-8859-1?Q?=22D=2ER=2E_Weiss_KI=D8RP=22?=" Subject: Re: [Elecraft] Question on 2nd Xtal filter mod for SSB My experience was like yours; the sound was so similar between LSB and USB that I decided not to change anything until I was able to do it using Spectrogram. Dave Weiss KI0RP K2 #2249 At 02:57 AM 11/11/01 -0500, dmaliniak at penton.com wrote: >Got a question for the assembled multitudes. > >I recently finally performed the mod referenced in the subject on K2 #265. >The mod was easy, went fine and seems to have worked. When I now switch >from USB to LSB, the overall pitch of the band noise is much closer than it >seemed to be before I did the mod. --- snip --- ++++++++++++++ Date: Sun, 11 Nov 2001 11:26:56 -0800 From: "Ron D' Eau Claire" Subject: RE: [Elecraft] Question on 2nd Xtal filter mod for SSB David,AD2A/6, asked: > I recently finally performed the mod referenced in the subject on K2 #265. >> The directions to the mod suggest that one may choose to tweak the BFO > settings for FL2/OP1. I realize that this is entirely subjective, but I > don't really see the need to fool with the BFO settings. So > finally, here's > my question... do most K2 owners, upon doing this mod, feel the need to > change the settings? Is it usual for them to be well out of whack after > doing the mod or not? I'll echo the other reply you have - it's something that I'd wait to use Spectrogram on. I did tweak my FL2/OPT1 filter to cut off the bandpass a bit lower than FL1 - about 1.8 kHz I think - which greatly reduces QRM in crowded conditions and, to my tin ear, produces much more pleasing audio on most SSB signals - especially on the K2 speaker. Before I did the mod the tilt in the bandpass caused so much low-frequency emphasis on one sideband that the audio became very mushy without the filter OPT1 filter set at FL1 all the time. Ron AC7AC K2 # 1289 +++++++++++++++++++