+++++++++++++++++++ See also K2 Signal Tracing ++++++++++++++++++ Date: Thu, 04 Oct 2001 09:41:25 -0500 From: Tom Hammond =?iso-8859-1?Q?N=D8SS?= Subject: [Elecraft] K2 deaf on 80M and CAL FIL problems Robert: >Overall things all seem to be working fine. The exception is that on 80M it >seems pretty deaf! I haven't tried to align the BP filters yet in the event >someone has had a similar problem and would like to share their findings. If it's not the BPF, the it's probably a poorly tinned toroidal inductor lead in either the BPF or the LPF. I'm ATTACHing hereto an article I wrote which may assist you in troubleshooting this problem. It's in PDF format... if you can't load the file into your version of Acrobat Reader, you probably need to update to v4. Version 5 of Reader seems to have problems NOT associated with my file, but problems in general, so unless they have a v6 out by now, I'd go with v4. (NOTE that reflector members will NOT receive the ATTACHment as all ATTACHments are stripped from messages passing thru the reflector). If anyone wishes to obtain a copy of my Cheap 'n Dirty Signal Tracing document(s), they are available from: For the K1: http://home.earthlink.net/~n0ss/k1_signal_tracing.pdf and for the K2: http://home.earthlink.net/~n0ss/k2_signal_tracing.pdf I'm told that these will eventually be posted to the Elecraft web site, but I don't believe they have made it there yet. >Also, does performing spectrogram basically set-up the filters so they are >centered between each other? Is this the main function of using >spectrogram? At present when I go from one filter to the next I find I have >to retune the signal in in order to bring back the signal strength and >intellegibility. This is not the way it should be. Have you downloaded the article from the Elecraft web site? If not, I suspect you should. The article is available at: http://www.elecraft.com/Apps/new_fil_docs/k2_cw_gram.pdf You say, "Also, does performing spectrogram basically set-up the filters so they are centered between each other?". I'm not certain of exactly what you mean by "between each other", but when using Spectrogram, you should set up a MARKER line (in Spectrogram) which sits on the same frequency as your SIDETONE is set to. Then, using noise, the noise peak should be CENTERED on TOP OF the MARKER line. This should, if your 'noise' is random enough to give you something close to 'white' RF noise, place all of the filters right on top of each other, thus requiring NO retuning when you switch from one filter to the next. If you STILL must retune each time, and particularly if you must retune more than 20-30 HZ, there's almost a 100% chance that your BFO is set to the WRONG side of the xtal filter and you have (at least) one bandwidth set to receive on LSB while the rest are on USB, or vice versa. Below is an excerpt from an early revision of the K2 manual. While I _think_ all of the steps still fit with current version K2s, if you find anything which doesn't quite fit, see if you can work around it. About the only thing tht might not still fit would be a different letter showing up on the LCD display, probably a "t". I'm including this excerpt because it does something the newer manuals don't do. It instructs you on how to fine the CENTER of your filter passband with the BFO. Finding the center of the filter is ESSENTIAL to ensuring that the BFO sits on the proper side of the filter when you are setting up the filters. CAL FIL Example (setting up all filters) 1. Connect the frequency counter test cable to TP2 (BFO). 2. Connect an antenna so you can hear some background noise. 3. Switch to a band lower than 15m. (The sideband is inverted on 15m and above, which is confusing during filter setup.) 4. Select CW Normal mode using MODE . 5. Tap XFIL until FL3 is selected. 6. Tap MENU and scroll to CAL. Hold EDIT to move the underline to OFF, then scroll until you see CAL FIL. Finally, hold EDIT again to activate the filter display. 7. The display will now look like FL3 0.40c. This shows that CW filter FL3 is set for a bandwidth of roughly 0.4 kHz. The filter parameter has a range of 0.00-2.49. Above 2.49, the filter parameter changes to OP1-OP5 . These 5 settings are used to select optional fixed filters, such as on the SSB adapter. Above OP5 there is an OFF setting (FL2, 3, and 4 only). If you set a filter to OFF, it will no longer be accessible with XFIL on exit from CAL FIL. 8. Use the VFO knob to vary the bandwidth. Notice that the "shape" of the noise changes. Return to 0.40 for now. 9. Tap BAND- to display the BFO setting, e.g. BF3 110c. The BFO parameter can be varied from 000-255, and allows the BFO to cover about 4 or 5 kHz. (BAND+ returns you to the FL3 display, but for now stay with BF3.) 10. Sweep the BFO value over its full range using the VFO knob. You'll hear the pitch of the noise change. At some point in the adjustment range you'll hear the pitch go very low and then to zero--this is where the BFO has passed through the center of the filter. You may also see a peak in the S-meter reading here. Note this zero-pitch BFO setting (around 130). BE SURE to NOTE the ZERO-PITCH setting of the BFO! Now, given this information, ALWAYS REMEMBER that when performing CAL FIL, _and_ when using a band lower than 15M (80-17M), the BFO settings for filters set using CW (normal) will ALL be BELOW the ZERO-PITCH setting of the BFO, and all BFO settings for CW REV will be ABOVE the zero-pitch setting of the BFO. If you don't ensure that this step is strictly followed, you WILL have to re-tune your VFO in order to find the signal when you switch filter bandwidth settings. IF you must use a band HIGHER than 17M (15M-10M, in order to obtain enough band noise), due to the IF mixing scheme, your CW filters will be reversed, so in this instance, your BFO settings will be ABOVE the zero-pitch setting when aligning CW (normal) and they will be BELOW the BFO zero-pitch setting when aligning for CW REV. BE CERTAIN YOU KNOW WHAT BAND YOU'RE USING and use the appropriate set of BFO settings. >I'm still learning how to use all the functions but a nice rig. I agree completely. Good luck, Tom Hammond N0SS +++++++++++++++++ Date: Sun, 9 Dec 2001 09:40:44 -0600 From: "Jay Henson" Subject: [Elecraft] K2 Signal Tracing Tom and the group, I was asked by a friend to look at his K2. It seems that the rig had low (very little) audio. The audio was there but could barely be heard, even with the audio control at max. I remembered Tom's article on signal tracing the K1 and K2 so I dug it out. The rig had low audio on ALL bands and I knew it had worked at one time. We had used spectrogram to align his filters. A gut feel told me the problem was downstream of the LPF's and BPF's. The attenuator selection did attenuate the input signal. Selecting/deselecting the preamp did produce a change in the signal level. Again, a gut feel told me the problem was downstream of the RF preamp. The sidetone level was loud and changeable from the menu and the volume control caused change in the audio level (what there was of it). Gut feel said the problem was upstream of the audio amp. Tom's article on signal tracing started at the TUF-1 and worked back to the antenna. What the heck. I tried his method at the product detector and IF amp. 4.915 mHz is still in the RF region. There were signals to be heard. Very weak but there. As I worked back through the crystal filter, the injected signals seemed to disappear. Rule #1 - Make the basic radio work first so I removed the SSB adapter and added the W2/W3 jumpers. Now there is plenty of audio. To make this long story short, the input side of the SSB adapter had a bad solder joint. The important thing I found was that Tom's signal tracing technique can be used in the IF sections also. The signals will be weaker as there is very little amplification but the signals CAN be heard. Many thanks for the technique, Tom. My best to all! Jay AJ4AY Mobile, AL FP -115, SOC 220, ARCI 8131, FISTS 7917 K2 # 1073 ++++++++++++++++++++ Date: Sun, 09 Dec 2001 21:58:54 -0600 From: Tom Hammond =?iso-8859-1?Q?N=D8SS?= Subject: [Elecraft] Re: K2 Signal Tracing Jay: >The important thing I found was that Tom's signal tracing technique can be >used in the IF sections also. The signals will be weaker as there is very >little amplification but the signals CAN be heard. > >Many thanks for the technique, Tom. My best to all! Thanks to YOU, Jay, for a bit more insight into troubleshooting the K1/K2! APPRECIATED! Season's greetings, Tom N0SS ++++++++++++++++++++ Date: Sat, 07 Sep 2002 13:22:23 -0500 To: Unifiedtx at aol.com, elecraft at mailman.qth.net From: Tom Hammond =?iso-8859-1?Q?N=D8SS?= Subject: Re: [Elecraft] Cheap'n Dirty Signal Tracer Roy: >Can anyone tell me where I can find the schematic for the Cheap'n Dirty >Signal Tracer? Thanks Roy Morris W4WFB K2/100 #2225 Here 'tis... it's not a circuit, just a method of trying to find out where you're losing the received signal... I've seen some references to the fact that it's been posted to the Elecraft website, but have never been able to find it. Since the Elecraft server strips ATTACHments, the ATTACHed file will not be sent to reflector readers. However the PDF files (one for K1 and one for K2) are available from me website, at: http://home.earthlink.net/~n0ss Go to the homepage (above), drop down to the bottom, click on DOWNLOAD PAGE and you'll find the PDFs either under K2 (or K1) Related files. Be sure to READ the docs at the TOP of the download page with regard to how to easily download a PDF without having it first come to your screen. Faster and LOTS LESS bandwidth required. 73, Tom Hammond N0SS ++++++++++++++++++ Date: Tue, 10 Sep 2002 17:59:51 -0700 From: Eric Swartz WA6HHQ - Elecraft Organization: Elecraft To: elecraft at mailman.qth.net, QRP-L , GQRP Subject: [Elecraft] New Signal Tracing Articles on Elecraft Tech Notes Page We have just uploaded two excellent articles on Signal Tracing techniques for debugging by Tom Hammond, N0SS. They are titled "K1 Cheap and Dirty Signal Tracing" and "K2 Cheap and Dirty Signal Tracing", but the techniques are applicable to any radio design. Thanks Tom! They can be found on the "Tech Notes" link at http://www.elecraft.com 73, Eric WA6HHQ +++++++++++++++++++ Date: Wed, 21 May 2003 18:52:55 -0500 From: "John R. Lonigro" To: Tom Cc: Elecraft at mailman.qth.net Subject: Re: [Elecraft] Need help tracing signal Tom wrote: > >The troubleshooting section of the manual tells us how to build a RF probe, >which I did. It seems to work fairly well, but this raises my first >question. Can I use the probe that came with my oscilloscope? I like the >idea of using the DMM as it is easier for me to read than the scope. > > The scope and its probe will show waveforms, not just voltages, which might be more enlightening. The RF probe that came with the K2 is for us poor slobs who don't own a decent scope. Either should work, but if you have a scope, go for it. > >What I do have is the Ft. Smith Marker/Generator, which puts out a signal >every 5 kHz including (I suppose) 10 MHz. When I measure this signal with >the RF probe which I built, its reading well over 2 Vrms. The K2's >troubleshooting section is talking in the 0.14 Vrms range. So I have much >more output than is called for. > >So, dear gurus, what do I do? I can breadboard the K2's suggested signal >generator and use a bunch of fix resistors in series and parallel (that >sounds kind of crazy). Or can I put a fixed resistor in series with the >output of the Marker/Generator and then use a pot that I have, in the 25K >range, to get the output down to 0.14 Vrms or so. (Oh, and it calls for a >non-inductive pot - how do I tell that?) > > A non-inductive pot is one that is not wirewound. Most volume or tone controls from old radios are typically non-inductive. I believe the pots Radio Shack sells are suitable. >Or, can I just use the marker/generator as is with the 2.x Vrms and adjust >my expected readings through the receiver accordingly (this seems to be the >easiest). > > I would NOT pump 2.x Volts rms into the front end of any radio. The Elecraft manual states that the receiver output resulting from the .14 Volt injected signal "might be quite strong even if your receiver is attenuating the signal somewhere." Just think how strong it would be if you injected 2.x volts into the front end and the receiver weren't attenuating the signal somewhere! >I appreciate your help, I don't think I've ever come away from this list >without a satisfactory answer to whatever I've asked (except maybe when I >asked to be bumped up on the list for shipping my DSP module). > >Thanks in advance > >Tom >WB2QDG >K2 s/n 1103 +++++++++++++++++