++++++++++++++++++ ++++++++++++++++++ From: "Steve Jackson" To: Date: Fri, 17 May 2002 20:53:16 -0400 Subject: [Elecraft] K1 #1181 on the air K1 #1181 has joined K2 #771 in the shack. The K1, a birthday gift from the XYL, will be the 'frequently portable' radio. I made two construction errors on the 4-band board. I wanted to mention what I did wrong and what happened, so that someone else can benefit from observing my carelessness. Turns out that I installed the premix coupling transformer incorrectly. It was wound fine, and had no PTTL syndrome (I use a solder pot). I just didn't see the green vs. red wires right ... now for the interesting part: the SA602 that drives it ran warm, and sucked down an extra 30 mA doing so. This is a good habit reinforcement: when assembling any kit, power it through an analog ammeter if you have one. (I've been doing this since I was eleven, 30 years' now, and it's always proven to be a good idea.) The rig was working, but not properly, and drawing nearly 100% more juice (85 mA) than it was supposed to. And since when does a 602 dissipate heat? I thought the chip was bad, even though bad '602 chips are darned rare ... and I pulled it and replaced it. (Thanks are in order to Scott and Gary.) New chip : Same symptom! Took me a while to find this first error, with a bright light and magnifying glass. Being on this side of 40 is a new experience ... I easily fixed the 'chip problem' by removing the transformer and reinstalling it correctly, and then the radio sprang to life. On two bands. Whoops. Error #2: I apparently got dyslexic on T3 and T4. They were swapped. I un-swapped them, and now have a smashingly effective 4-band K1. Ordered the noise blanker today. Don't know if I'll need it, but better to have it than not. KAT1 is next. 72 Steve KZ1X/4 Chapel Hill, NC +++++++++++++++++++ Reply-To: "Brian" From: "Brian" To: "K1" Date: Wed, 10 Jul 2002 22:59:15 -0500 Subject: [Elecraft] K1 #1205 - 20m problem solved Well I finally got some extra time tonight to work on my K1 receive issue. The thing receives on 3 out of 4 bands....20m was dead. Tom Hammond sent me some excellent down and dirty signal tracing instructions. They were a big help. Anyway...it boiled down to me pulling the crystal, putting down a piece of tape, and reinstalling the crystal. Now it works very FB. So now on to the next steps...the transmitter! Thanks to everyone that sent me encouragement and instruction. I'm also doing the Elmer 101 class with the SWL+ 20 so I am learning more as I go along. I even built my first RF probe tonight. Hard to imagine I have muddled through for the last three years with out one. Jeeesh. ======================================== KB9BVN/QRP - New Whiteland IN - EM69WN QRP-ARCI #10223 QRP-L #1540 FIST #5695 FISTS CC #764 - Proud Member ARRL HEATH HW-9. 2W or NORCAL 40A. 1.3W INTO INFAMOUS AF4PS ATTIC DIPOLE SOC #400 AND FLYING PIGS QRP #-57 ======================================== +++++++++++++++++++ From: "Wallace, Andy" To: Elecraft Date: Thu, 17 Oct 2002 14:32:11 -0400 Subject: [Elecraft] K1 toroid turns spacing fiddling I was concerned that I couldn't tweak any more than 2W out on 15m with my dad's 2-band (KFL1-2) K1 filter board. After emailing Gary at Elecraft, he reminded me that the output does depend on power supply voltage. In my case, I am using a 12.0V supply, which yields only 11.6V after the protection diode, as shown on the K1 readout... Well anyway, he also said that in some cases, people have coaxed more power out by compressing or expanding the turns on their filter board input and output toroids. Hope he doesn't mind me quoting him: ==== >From Gary Surrency: I've seen the best results by expanding the turns on either L9 or L11 (the input to the LPF), and then compressing the turns on either L12 or L10, depending on which band is set up for 15m. The optimizes the match between the PA and the load, for maximum power transfer. Be sure to do it at maximum power, as the power level is a factor in the PA to load impedance match. Compressing the turns increases the toroid inductance, and vice-versa. ==== Back to Andy... In the case of my dad's rig, it made a difference. Instead of 5W on 20 and 2W on 15, I now get 4W on each band, with the 12V supply. However, I can't say that expanding L9 and compressing L10 did it -- I believe I expanded both of them slightly. Obviously depending on how symmetrically you wind these things, the direction may change. My own 4-band K1 got 5W on 40, and then dropped off from there for 30/20/15, so I thought tweaking my toroids would help also. Gary says to do this with OUT=5W and with a 50 Ohm dummy load on the antenna jack. With the 4-band board, you have two groups of three toroids in the in/out sections. What I did was transmit into the dummy load at 5W and watch the wattmeter needle as I gently squeezed or expanded the turns of the toroids. Touching them makes a difference, so you have to let go to see whether it helped or not. (Or use the plastic trimmer tool to push on the turns.) It didn't take a huge amount of compression or expansion, but I was able to see an improvement. Be sure to swap back and forth between bands in the same filter path to make sure you're not improving one at the cost of the other. I then went back and repeaked the trimmer caps at 2W, and now my output with a 12V supply is 5+W on 40/30 and 4W on 20/15, which is an improvement. Receiver seems a little more sensitive now, too. Power output is very even across each band. I did not check to see if power resolution (steps between powers below 5W) was affected. Since my antenna is just a 35' randomwire I generally run at 5W! This was achieved without substituting values for R11, for more drive, etc. Do this at your own risk -- at worst you'll have to readjust the toroids for even spacing and repeak the trimmers or coils as appropriate. My guess is that other K1s (and maybe K2s) could benefit from a little toroid nudging. The design inductance may or may not be critical, but the tightness of the windings and the spacing between turns could make enough of a difference that the rig is not quite "all it can be." Good luck! -Andy +++++++++++++++ From: "Don Brown" To: , Cc: "Elecraft" Subject: Re: [Elecraft] SN Date: Fri, 27 Dec 2002 07:44:59 -0600 Hi The serial number is in the small brown envelope with the heatsink insulators. It is a small silver mylar sticker with black numbers. It is easy to miss and toss out with the envelope. The number is also on the front of the box that the radio is packed into. Don Brown KD5NDB ----- Original Message ----- From: To: Cc: "Elecraft" Sent: Friday, December 27, 2002 4:18 AM Subject: Re: [Elecraft] SN > I'm on the last legs of my K1 build and have also not come across the serial > no. I have kept all the packaging etc and can't find the sticky label > anywhere. > > I've copied this email to Lisa, so perhaps she can help me locate mine. > > The K1 is going well, I'm on the last leg of construction - only two pages > in the construction section go. > > The only mistake found so far was a non-soldered joint on the headphone > socket. > > Regards Tony > G7IGG +++++++++++++++++++ Subject: RE: [Elecraft] FW: K1 : Problems : Any Ideas ??? Date: Tue, 25 Feb 2003 17:54:44 -0500 From: "Morrow, Michael A." To: , Peter Rosney EI2IF wrote: > I am building a K1 with the 4 band option ( no. 1227 )=20 > and I have tonight set up the radio to align the receiver. > However when I connect an antenna to the adjacent Ham > Transceiver and key down all I am getting is a key click on=20 > 10mhz.=20 Were you able to align the two highest bands OK? If you haven't tried = that yet, try to align the 15m (or 17m) band first (caps C1, C2, CA, = CB), followed by the 20m band (caps C3, C4, CC, CD). If you are able to = get a good alignment of those two bands, that would narrow down the = problem area significantly. The manual has you start with the 30m band = first, but actually the only order that you MUST observe is: a. Always align the 15m/17m band before you align the 20m band, and b. Always align the 30m band before you align the 40m band. Any time you realign the 15m/17m band you must follow that with a = realignment of the 20m band. Any time you realign the 30m band you must = follow that with a realignment of the 40m band. > I adjusted c5 & c6 but still not a sausage. I am not=20 > getting any morse audio on receive. I also adjusted ce and cf > and all to no avail. I am getting hissing noise on rx which > is ok but no morse on the headphones. The first time you try to align the filter board, you may need a pretty = strong signal on the frequency to which the K1 is tuned. Since you have = not yet performed the K1 frequency display calibration procedure, the = frequency now being displayed on your K1 may be as much as 10 kHz off. = (Don't forget to assign the proper lower band edge frequency of 10.1, = rather than 10.0, MHz to band B2.) You'll have to have a strong enough = signal to get through the unaligned filter board stages to be heard = weakly in the K1 receiver. Have you tried *slowly* sweeping the VFO of = the transmitter that you are using as a signal generator through a 15 = kHz range on either side of the frequency displayed on the K1 in order = to find that signal? Once you find the signal, then it will be easy to = peak the appropriate caps. If you are able to align the two highest bands, but not the 30m band, = then set the 30m caps C5, C6, CE, and CF in mid-position (screw slot = perpendicular to the flat side of the cap, I think). Then see if you = can get the 40m band to align. If successful, you'll at least know the = problem has narrowed down only to 30m band components. =20 > I went for the 10.1 mhz crystal option and all components seem to be=20 > correctly installed. Another useful thing to check is that your filter board crystals are = oscillating when the appropriate band is selected. The 30m crystal X3 = should easily be heard in an external receiver, even before you align = the board, within a few kHz of 18.100 MHz. Just run a short wire near = the filter board to the antenna of a receiver that can tune that = frequency. Best of luck. 73, Mike / KK5F +++++++++++++++++++ Subject: RE: [Elecraft] K1 RF board resistance, voltage, L1 Date: Wed, 23 Apr 2003 15:41:27 -0500 From: "Morrow, Michael A." To: Cc: Mark Fickett wrote: > I've just finished with building the transmitter portion=20 > of the board, and have run across a resistance check and a=20 > couple voltage checks that are off... > > Resistance (p. 45): ground (-) to J8 pin 5 should be=20 > 90-110k, but I'm getting 1.5k. I'd suggest re-checking the orientation of D15. > Voltage (p.46): all from ground (-), with about 12.8v in=20 > (the supply is rated 13.2v, but the K1 says it's 12.8). No problem there. There'll be about 0.3 vdc drop (exact drop depends on = D16 junction temperature and current flow) across the reverse polarity = protection diode D16. 13.2 - 0.8 =3D 12.9. > U8 pin 1 should be 1.3-1.5, and is 1.05 I'd say no problem, that's close enough. > U8 pin 8 should be 5.1-5.5, is 5.8 I'd say no problem, that's close enough. > D11 cathode should be 4.0-4.6, is 0.8 I don't know about that one. Are you sure you had the positive lead of = your voltmeter connected to the junction of D11 and D12? > And finally, a question: if I readjust L1 (for the VFO),=20 > will I need to recalibrate the reciever, or will it just=20 > affect the VFO (and .: the band edges)? No sweat! When you did the receiver display calibration, you were = calibrating the LCD display to take into account variations on the = various heterodyne crystals RF-X5 and filter X1 through X4. This cal is = necessary because the LCD display frequency counter only counts the = output of the VFO (L1), and not the other oscillators. =20 So, you can adjust L1 turn spacing after the earlier band calibration of = the LCD display, and still have an accurate LCD display of the receiver = frequency. Adjusting L1 will, as you suspect, only alter where the band = coverage starts and ends. Squeezing together turns on L1 to make the = VFO operate at a lower frequency will RAISE the K1 lower and upper band = ends. Any L1 adjustment will affect all four bands by the same amount. = You need not re-do the band cal procedure. 73, Mike / KK5F +++++++++++++++ To: "Morrow, Michael A." , elecraft at mailman.qth.net Date: Wed, 23 Apr 2003 14:38:21 -0700 From: "Mark Fickett" Reply-To: throughtheye at lycos.com Subject: RE: [Elecraft] K1 RF board resistance, voltage, L1 Organization: Lycos Mail (http://www.mail.lycos.com:80) Hello, In response to Michael Morrow " in response to myself: >>Resistance (p. 45): ground (-) to J8 pin 5 >> should be 90-110k, but I'm getting 1.5k. >I'd suggest re-checking the orientation of D15. I re-checked it, and the banded end is indeed facing left. (No luck with spontaneous change of resistance, either.) >> D11 cathode should be 4.0-4.6, is 0.8 >I don't know about that one. Are you >sure you had the positive lead of your >voltmeter connected to the junction of >D11 and D12? I'm not sure what I had wrong before, but it now measures as 5.1v. I think I /may/ have been measuring the wrong diode or wrong end. >> if I readjust L1 (for the VFO), ... >So, you can adjust L1 turn spacing >after the earlier band calibration of >the LCD display, and still have an >accurate LCD display of the receiver >frequency. O-/kay/. I readjusted L1 so I'm closer to the bottom edges, and get a smidge more of 15m. Thanks! -Mark Fickett, KB3JCY ++++++++++++++++ Date: Wed, 23 Apr 2003 15:59:52 -0700 From: Wayne Burdick Organization: Elecraft To: giuliano Cc: elecraft at mailman.qth.net Subject: [Elecraft] Re: K1 SSB? Hi Guiliano, Thanks for the further clarification. giuliano wrote: > I have made a polarization network; i have about 150 ma IC current without > modulation, i think that now the PA is running Class-AB OK. You'll want to check the PA device temperature during long key-down periods, but the side panel should provide adequate heat sinking at this bias level. > > - How did you reverse the I.F. strip to provide SSB filtering on transmit? > Or > > did you add a separate TX SSB crystal filter? > > Yes i use two separate crystal filters . Reception filter with 6 crystal and > TX with 4 crystal Did you measure the opposite sideband rejection and carrier suppression? We needed 7 crystals in the K2 to obtain really good results. > > - How did you handle the VFO range issue? > > Now i have add a 68 NPO cap. on parallel to the 120 Cap. supplied. > So the tuning range is about 250 KHz.... > > I am working, to solve better this problem, with a DDS AD9834 and a tuning > encoder with 10/100/1000 Hz selectable tuning step. > It is the next modification step. This will work, but are you also planning to include a phase-locked loop to clean up the DDS's output? Using DDS as the injection oscillator by itself will significantly degrade the performance of the K1's receiver. The spurious-free dynamic range of any DDS with a 10-bit DAC, including the AD9834, will only be in the 60 to 70 dB range at best. Large in-band signals will mix with the spurs to produce numerous "ghost" images within the passband. This would be especially noticeable during contests, but could be a problem anytime the band is open or if you have strong nearby stations. For this reason, raw DDS VFOs are only acceptible the simplest of equipment, or if the operator is willing to accept the compromise. I wouldn't bring this up except that the K1's present performance is outstanding, and anyone contemplating making a DDS mod should be aware of the limitations. > > - How did you handle the band-pass filter bandwidth issue? ...it's of critical > > importance that both the premix and RF band-pass filters be as narrow-banded as possible > > I have measured the Filter selectivity, (see also my filter plot on my Web > page) and seems that selectivity is adeguate to cover 350 KHz on 20 M and > 150 KHz on 40 m. This would provide useful range on 20 m and up, but on 40 m the range would be adequate only for non-U.S. K1 owners. In this case, as well as on 80 m, you'd need to tune the filters for either CW or SSB. By the way, did you make these measurements using only the 2-band filter board? The 4-band board's bandwidths are quite a bit narrower because of the high-Q inductors and the multi-band switching arrangment. It just covers 80 kHz on 40 meters, for example, and I believe the usable bandwidth on 20 meters is 200 kHz or less. Thanks, Wayne N6KR ++++++++++++++++ From: "Trevor Jacobs" To: "Wayne Burdick" , "giuliano" Cc: elecraft at mailman.qth.net Subject: Re: [Elecraft] Re: K1 SSB? Date: Wed, 23 Apr 2003 16:07:24 -0700 Hi Wayne and all, What about the new AD995X series with the 14 bit DAC? The inproved resolution of the DAC should lower the spurs well below the 60 dB range. I haven't played with them yet but they are worth looking into for this type of application. What is the SFDR of the current VFO in the K1? I haven't seen that information published. How far down do the spurs need to be before they approach the kind of performance that we are seeing in the K1 or K2 at the moment. The big advantage is the phase noise. Looking at the output of a AD9850 on a spectrum analyzer, the phase noise is almost non existant. Take care... 73's Trev KG6CYN http://www.qsl.net/kg6cyn > This will work, but are you also planning to include a phase-locked loop to > clean up the DDS's output? Using DDS as the injection oscillator by itself will > significantly degrade the performance of the K1's receiver. The spurious-free > dynamic range of any DDS with a 10-bit DAC, including the AD9834, will only be > in the 60 to 70 dB range at best. Large in-band signals will mix with the spurs > to produce numerous "ghost" images within the passband. This would be especially > noticeable during contests, but could be a problem anytime the band is open or > if you have strong nearby stations. For this reason, raw DDS VFOs are only > acceptible the simplest of equipment, or if the operator is willing to accept > the compromise. > > I wouldn't bring this up except that the K1's present performance is > outstanding, and anyone contemplating making a DDS mod should be aware of the limitations. ++++++++++++++++ Date: Wed, 23 Apr 2003 16:35:57 -0700 From: Wayne Burdick Organization: Elecraft To: Trevor Jacobs Cc: giuliano , elecraft at mailman.qth.net Subject: [Elecraft] Re: K1 SSB? Hi Trevor, Trevor Jacobs wrote: > What about the new AD995X series with the 14 bit DAC? The inproved > resolution of the DAC should lower the spurs well below the 60 dB range. The SFDR of the AD995X series is about 95 dB, I believe, and it's certainly a cleaner part than the 983X series. But it will still not be nearly as clean as an analog (L-C) VFO. Its spurs would still be the limiting factor in receiver in-band image performance. It's also quite expensive. You'd also need: - a low-jitter, VHF crystal reference oscillator for good phase noise performance - at least 3 spare I/O lines on the MCU to drive the DDS, and more if you include a PLL (the K1's microcontroller at present has no spare I/O lines) - a shaft encoder (preferably optical: $15-$20 minimum; contacting-type encoders are a lot less expensive but have about 1/100th the expected rotational life - some means of reading the position of the RIT offset pot and controlling the LEDs and LCD (this implies integration of the new modification with the K1's present microcontroller) In summary, adding DDS control to the K1 (inculding the recommended PLL) would require numerous hardware changes. This would make much more sense as a total redesign, taking advantage of the fact that no premixer or TX mixer would be required, and adding broader band-pass filters, a display with more digits, etc. By the time you finished, it would be a much more complex and expensive rig--perhaps a K2 would be a better choice ;) 73, Wayne N6KR +++++++++++++++ From: M0TLTHF at aol.com Date: Sun, 4 May 2003 10:38:24 EDT To: elecraft at mailman.qth.net Subject: [Elecraft] K1 #1518 Completed and on air Finally completed K1 #1518, after having a couple of problems on receive and transmit. Problem 1 - Receive was very quite Cure - faulty 2N7000 in the AF Preamp/Mute Circuite, replaced Q10 2N7000 audio bust into life. Problem 2 - No transmit power P=01 when power set to 2 Watts Cure - Faulty resistor R30 which was causing a problem with Q6 2SC2166 replaced resistor R30 K1 tuned average of 7 Watts output on all 4 bands. Hope this may help someone Trevor, M0TLT ++++++++++++++++ From: "John Webster" To: Date: Mon, 26 May 2003 00:20:06 -0700 Subject: [Elecraft] Re: Help, I think I've killed my K1 Greetings: [Rather Lengthy] This is a long-overdue final report on a thread started back in February. While the problem was solved by mid March, I have been away so much that I have not had a chance to wrap this up properly, and thank those who helped. The symptom was sudden relay chatter, LED flashing and then silence from my K1, while tuning a manual antenna matching device (a K.W. Electronics E-Zee Match), and keying my 100 watt base station rig (Kenwood TS-850). The initial problem turned out to be a blown Front Panel Board 6v regulator, due to high RF field from the tuner while the K1 was on standby (but antenna disconnected). The damage seems to have been caused by induced high voltage/current. The K1 was too close to the antenna tuner (only some 3 inches away from it), and was thus affected by the very high energy fields that can be generated by large air core inductors with high SWR on the feedline. It took a little work to find and solve the damage to the K1. The members of the list proved invaluable. Particular thanks are due to: Mike KK5F; Bob AG5Q; Stuart K5KVH; Trevor KG6CYN; and Jessie Oberreuter. It was quickly determined that the 6v regulator on the Front Panel Board was blown. None of the other voltage regulators were faulty, and the problem was thus isolated to the FP board. Replacement (with a locally obtained 6.2 volt reg., NTE-988) brought the K1 to life, but not to full health. I suspected the NTE-988 and ordered the correct ZR78L06C from Elecraft. But much to my chagrin, this did not solve the anomalies. Upon checking, the voltage output was being pulled down to 4.3 volts (only 232 ohms to ground). Now the search got serious. With help from Mike and Bob, I proceeded to isolate the output of the regulator from everything else connected to it (even C4 & C5). As Murphy's Law would have it, the last part I checked was U2 the D/A Converter (MAX518). On lifting U2 pin 7 the low resistance to ground disappeared. But the output of the 6volt regulator was still too low (5.25 volts). Some- thing was still wrong! Again, with help from Mike, it turned out that the regulator was not happy with C4 and C5 lifted. Finally, with a replaced D/A Converter from Elecraft ($10), and everything else reconnected, the K1 has once again worked flawlessly every since. Lessons learned: 1. This list is an incredibly valuable resource. The combined talent, skill, experience and know-how on here is priceless. 2. I gained enough confidence to fix a non-trivial problem on my K1, and learnt a whole lot about the circuit in the process. 3. To remove an IC, cut its pins, and then desolder them carefully. To isolate a trace, you can cut a single pin and lift it up, and then resolder it back, without leaving hardly even a trace. 4. I need a Hakko 808 desoldering tool! 5. Don't put your K1 (or K2?) near Antenna Tuners (particularly those with air cores). It turns out that almost the identical problem has happened to two others before (Jessie and Trevor). Beware. [I take it that the toroid cores of the ATU's in the K1, K2 and in the KAT100 are less of a problem]. 6. Be careful with the K1 when you have it switched to "S-meter" (reading relative signal strength). With no signal present, and the volume turned down, you have no way of knowing whether the rig is on or off. I always turn the display back to "normal" if I leave the rig on, while unattended. [Doesn't help in the dark, of course!] 7. I'm so impressed with the K1 that I have bought a K2 to build . . . now to find the time! Thanks for reading this long winded and long overdue post. 73 John, N6JW +++++++++++++