Re: [Elecraft] 10M birdie and other assundries
Sorry to be a wet blanket, but from my current standpoint, I'd be grateful to find just the one spurious signal from the KPA. I have literally dozens of spurii on all bands that are S2 or greater. Some read S9+40 and would read higher if the S-meter could indicate that. They all emanate from that dratted HV bias supply and moving between high and low power on the K2's power control doesn't affect the birdie count. I haven't found any that move, either. Turning off Q8, the bias supply osc. transistor, turns off all the spurii. All components have been checked most carefully for value and all have been re-soldered. The KPA was constructed to the latest revision and with R4 at 100k, the bias voltage was just over 140v. Increasing R4 to 220k has dropped the bias voltage to 128v, but hasn't reduced the spurii. They are not entering the K2 via the speaker lead, aux. power or RF connections but are present at rock-crushing levels with just the ribbon cable connected. It's been a great disappointment to find this problem, as everything else has worked beautifully (after minor debugging). I note the various comments regarding the difficulties experienced by others in their attempts to tame the situation, so with my limited expertise, I don't hold out a lot of hope. Just in case it's the MAX1406, I'll try lifting one end of D7 to disconnect its input, but after that I'll probably resort to replacing each and every component in that area one by one, in case there's a rogue. 73 and Season's Greetings to everyone. Dave L G3TJP, who's feeling very sorry for himself just now ;-( Quote: Julius, N2WN Just out of curiousity, after spending some time on 10M during ARRL 10, I noticed a birdie around 28.019/20. Is there a way to eliminate this or move it below 28.000? ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
If PC5M's calculations are correct, and these birdies are caused by mixing products of the 4.9136 MHz IF crystals in the K2 and the 18.432 MHz bias supply crystal in the KPA100, and assuming his spreadsheet calculations are correct, it seems that a crystal at 15.8735 MHz (+ or - about 9.5 kHz) would not produce mixing products in any US ham band (including 5.3305 through 5.4065 MHz for 60m). Since this is not a standard crystal frequency, I wonder what the cost would be of a batch of custom crystals at this frequency? It seems that even within general tolerances for calibration and temperature stability for both values of crystals this should be feasible. Like I said, I don't know what the cost of a run of custom crystals would be, but it may be something for the folks at Elecraft to look into for future outgoing KPA100's, and then those of us who wish, could purchase one as a spare part to retrofit our existing amps. Another alternative would be a separate individual purchasing a batch of custom crystals for sale to Elecraft users. Just a thought. What do you all think? -Ben K1NT ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
Hi Ben, My assumption is that the birdie is due to mixing of the VCO with X-tal frequencies. At certain a (VCO) frequency it will be such that this mixing product will produce exactly the IF frequency-- a birdie is born. Of course will this certain VCO frequency correspond with a RF frequency (the frequency which is on the display of the K2). Now we need to do some maths:-) So the calculation is as follows: (1) X*Fxtal + Y*Fvco = Fif (never mind the + sign , minus is also taken care for by the negative Y values). Fxtal = KPA100 xtal freq Fvco = K2 VCO freq Fif = K2's IF (4.9136 MHz) X and Y are integers from e.g -4 to + 4, i.e. -4,-3,-2, etc to +4 (number of mixing products you would like to consider) Also we do know that Fvco = Frx - Fif (minus will be a plus depending if you are below or above 17 mtrs) (2) Fvco = K2 VCO freq Frx = K2 displayed freq Fif = K2's IF With (1) and (2) we can solve this equation: (3) Frx = Fif + (Fif-X*Fxtal)/Y and (below 17 mtrs) (4) Frx = -Fif + (Fif-X*Fxtal)/Y (above 17 mtrs) This is basically what my spreadsheet is showing, it will loop through the matrix of X and Y values, and calculates Frf depending on the chosen Fxtal and (fixed) Fif. That is will show this matrix for every band and perform a conditional format (red) for in band products. The other colors have been an attempt to show that mixing products with a lower order of X and Y will be stronger than those with larger numbers, although I did find any proof of that (somethimes higher order mixing are even stronger than lower..) And you are correct ! 15.873 would be perfect... Would really appreciate if someone could take the time and verify above paradigm. Gl, Carel, pc5m -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Ben Hofmann K1NT Sent: Wednesday, December 20, 2006 8:17 PM To: elecraft@mailman.qth.net Subject: Re: [Elecraft] 10M birdie and other assundries If PC5M's calculations are correct, and these birdies are caused by mixing products of the 4.9136 MHz IF crystals in the K2 and the 18.432 MHz bias supply crystal in the KPA100, and assuming his spreadsheet calculations are correct, it seems that a crystal at 15.8735 MHz (+ or - about 9.5 kHz) would not produce mixing products in any US ham band (including 5.3305 through 5.4065 MHz for 60m). Since this is not a standard crystal frequency, I wonder what the cost would be of a batch of custom crystals at this frequency? It seems that even within general tolerances for calibration and temperature stability for both values of crystals this should be feasible. Like I said, I don't know what the cost of a run of custom crystals would be, but it may be something for the folks at Elecraft to look into for future outgoing KPA100's, and then those of us who wish, could purchase one as a spare part to retrofit our existing amps. Another alternative would be a separate individual purchasing a batch of custom crystals for sale to Elecraft users. Just a thought. What do you all think? -Ben K1NT ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
Hi Jules! I have a temporary workaround that I use...albeit a pain in the butt one. That birdy seems related to an oscillator in the KPA-100 (HV PS?)..although Don can be probably be more specific. I've found that by turning the power level down that the birdy shifts a bit in frequency...maybe a few kHz. Enough to make a difference between hearing a station and not. If a new country is sitting on 29.020 I'll twiddle the power level between transmissions...otherwise I avoid that frequncy during normal operation. 73 de Tom K2TA - Original Message - From: J F [EMAIL PROTECTED] To: elecraft@mailman.qth.net Sent: Tuesday, December 19, 2006 8:21 AM Subject: [Elecraft] 10M birdie and other assundries Just out of curiousity, after spending some time on 10M during ARRL 10, I noticed a birdie around 28.019/20. Is there a way to eliminate this or move it below 28.000? As to my high current draw on 160, the KPA100 works fine into a dummy load and straight into the antenna. It does seem to draw more with the KAT100 in the circuit. I can't imagine much in the tuner that would cause such a problem (although I could be wrong). My thoughts drift towards a ground loop issue and/or rf getting back into the rig's control circuit. Guess I will be adding some further current chokes in the shack... Best wishes to all for the holiday season. May 2007 be good to you and yours... Cheers, Julius n2wn ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
J F wrote: Just out of curiousity, after spending some time on 10M during ARRL 10, I noticed a birdie around 28.019/20. Is there a way to eliminate this or move it below 28.000? Yes, but even I haven't had the guts to implement it. Replace the diode t/r switch with a reed relay such as the Matsushita RSD12V.Reduce the power level of the bias oscillator and remove the high voltage multiplier section so you can get -12v for the RS232 interface with less stray RF floating around. Alternatively, find a way to power it from the computer (and some RS232 ports will apparently work without the -12v at all), so the bias oscillator can be completely disabled. I tried various forms of shielding, filtering and bypassing without success. A friend pointed out to me that the birdie is a result of mixing between VHF harmonics of this and another oscillator in the K2, so that explains why it's so hard to kill. -- 73, Vic, K2VCO Fresno CA http://www.qsl.net/k2vco ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
Vic K2VCO wrote: Replace the diode t/r switch with a reed relay I should have specified that I was talking about the t/r switch in the KPA100, not the QRP K2, of course. -- 73, Vic, K2VCO Fresno CA http://www.qsl.net/k2vco ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
Anyone know if remoting the KPA100 to an EC-2 case reduces the birdie? --Mike At 10:32 AM 12/19/2006, Vic K2VCO wrote: J F wrote: Just out of curiousity, after spending some time on 10M during ARRL 10, I noticed a birdie around 28.019/20. Is there a way to eliminate this or move it below 28.000? Yes, but even I haven't had the guts to implement it. Replace the diode t/r switch with a reed relay such as the Matsushita RSD12V.Reduce the power level of the bias oscillator and remove the high voltage multiplier section so you can get -12v for the RS232 interface with less stray RF floating around. Alternatively, find a way to power it from the computer (and some RS232 ports will apparently work without the -12v at all), so the bias oscillator can be completely disabled. I tried various forms of shielding, filtering and bypassing without success. A friend pointed out to me that the birdie is a result of mixing between VHF harmonics of this and another oscillator in the K2, so that explains why it's so hard to kill. -- 73, Vic, K2VCO Fresno CA http://www.qsl.net/k2vco ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
Hi Julius, If the birdie at 28.019/20 MHz changes frequency when you change the transmitter power demand from 10w to 10w, then the KPA100's HV Bias oscillator is involved. Vic, K2VCO, and I had an off list discussion some time ago about birdies including this one, and though Vic found the birdie I did not because his HV Bias oscillator runs at 18.434 MHz while I have opted to keep mine at 16.293 MHz for reasons other than birdies. If I remember correctly this particular birdie involves not only the HV Bias oscillator but also another oscillator in the K2 which mix to produce a signal at a spurious response frequency of the receiver when tuned to 28.019 MHz. I'll have to look up my notes. You can move this 10m birdie to another frequency by changing the HV Bias oscillator frequency although the options for all band freedom from birdies involving this source are limited. In my K2/100 the bias oscillator signal enters the K2 via the ribbon cable, which can be 'choked' with a clamp on ferrite, and the loudspeaker leads which can be removed from the KPA100 board where the trace acts as a link, but be advised that these last two mods might not work with all K2/100s. Later on I repackaged the HV Bias oscillator and voltage multiplier in a shielded box, with pi configuration filters on all DC lines to kill the problem at source. 73, Geoff GM4ESD - Original Message - From: J F [EMAIL PROTECTED] To: elecraft@mailman.qth.net Sent: Tuesday, December 19, 2006 1:21 PM Subject: [Elecraft] 10M birdie and other assundries Just out of curiousity, after spending some time on 10M during ARRL 10, I noticed a birdie around 28.019/20. Is there a way to eliminate this or move it below 28.000? snip ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: [Elecraft] 10M birdie and other assundries
I don't know how reduced it is because I haven't compared it to the KPA100 as the top lid but it is barely noticeable in my K2 with the KPA100 in an EC2 unless I have the preamp on. Even then, it is more a nuisance than a problem unless the signal is very weak. 73, Mike WA3KYY At 11:58 AM 12/19/2006, you wrote: Anyone know if remoting the KPA100 to an EC-2 case reduces the birdie? --Mike At 10:32 AM 12/19/2006, Vic K2VCO wrote: J F wrote: Just out of curiousity, after spending some time on 10M during ARRL 10, I noticed a birdie around 28.019/20. Is there a way to eliminate this or move it below 28.000? Yes, but even I haven't had the guts to implement it. Replace the diode t/r switch with a reed relay such as the Matsushita RSD12V.Reduce the power level of the bias oscillator and remove the high voltage multiplier section so you can get -12v for the RS232 interface with less stray RF floating around. Alternatively, find a way to power it from the computer (and some RS232 ports will apparently work without the -12v at all), so the bias oscillator can be completely disabled. I tried various forms of shielding, filtering and bypassing without success. A friend pointed out to me that the birdie is a result of mixing between VHF harmonics of this and another oscillator in the K2, so that explains why it's so hard to kill. -- 73, Vic, K2VCO Fresno CA http://www.qsl.net/k2vco ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: Elecraft@mailman.qth.net 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: Elecraft@mailman.qth.net 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: [Elecraft] 10m Birdie K2
Hi Adi, You could be right. But keep smiling, K2/100 is in hospital here and doctors are looking at Ostrich, C.Condor and GORILLAS. Patient doing well. Your information very useful. 73, GeoffGM4ESD K2/100 3255 - Original Message - From: S55M [EMAIL PROTECTED] To: elecraft@mailman.qth.net Sent: Tuesday, January 18, 2005 11:05 PM Subject: [Elecraft] 10m Birdie K2 Hello World! My situation with the bigest bird (called by HAM's B4 me an Ostrich but i think is an California Condor): QRGMode Preamp state Signal RX ant off Signal RX ANT On 28263.11 CW ON S-8 S-2 Disabling KPA100 with turning Power button under 10W just shifts OSTRICH for 8Hz up. With KPA100 removed no such huge birds heard. Cristal in KPA100 is 18MHz. Quiet shure, that the OSTRICH has something to do with HV switching supply in KPA100. I do not worry a lot abt it, but if it's possible to get rid of it, i would be happy, because like California Condor, the HAM's of the world are endangered species. I wait for some mod from You guys with a lot of knowledge about the argument. 73's S55M-Adi K2-3204 ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: [Elecraft] 10m Birdie
Another wierd behavior of the 28020 birdy is that is tunes the opposite of a normal cw signal and changes pitch in large steps. I wonder if it is a harmonic of the HV oscillator in the KPA-100 beating with something related to the main tuning oscillator? - Original Message - From: Russ Tobolic [EMAIL PROTECTED] To: elecraft@mailman.qth.net Sent: Monday, January 03, 2005 10:24 PM Subject: Fw: [Elecraft] 10m Birdie I never operate 10M but just checked my K2 with the KPA100 and found the same thing with a strong birdie at around 28262. The weird thing is that when I turned off the KPA100 (reducing power below 10W) the birdie at 28262 shifted up about 1kc and the birdie at 28019 shifted down about 500c. Russ, N3CO K2 #4016 - Original Message - From: [EMAIL PROTECTED] To: elecraft@mailman.qth.net Sent: Sunday, January 02, 2005 11:19 AM Subject: [Elecraft] 10m Birdie In order to get a match at 28019 kHz as has been reported,= the KPA100 oscillator has to be shifted to 18434.1 Hi also have the 28029 birdie. I've also got the much stronger= 28262 birdie. I've wanted to eliminate this for some time since there a lot of= cw activity between 28025 and 28025. My KPA100 is installed on the K2. The= birdie produces about 1 s unit bar with the pre-amp on. Based on Sverre= comments, how do I eliminate the birdie? Is it merely the freq of the 18= MHz oscillator. Frank, W4FMS ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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 -- No virus found in this incoming message. Checked by AVG Anti-Virus. Version: 7.0.298 / Virus Database: 265.6.7 - Release Date: 12/30/04 -- No virus found in this outgoing message. Checked by AVG Anti-Virus. Version: 7.0.298 / Virus Database: 265.6.7 - Release Date: 12/30/04 ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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: Elecraft@mailman.qth.net 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: [Elecraft] 10m Birdie
-Original Message From: [EMAIL PROTECTED] --- I've wanted to eliminate this for some time since there a lot of cw activity between 28025 and 28025. My KPA100 is installed on the K2. The birdie produces about 1 s unit bar with the pre-amp on. Based on Sverre comments, how do I eliminate the birdie? Is it merely the freq of the 18 MHz oscillator. -- One solution is to shift the 18432 oscillator on the KPA100 to a frequency above the 10m band. The oscillator is only used in a DC-DC converter for producing a voltage for biasing the T/R-switch diodes and the RS232 interface, so the frequency is not critical at all. The only limitation is that the 2N3904 - FT37-61 oscillator may get into trouble as frequency increases. I have a half-built KPA100 so I cannot yet listen for birdies, but I made some tests with two crystals from my junk box: 1. 38.45 MHz: oscillates, but only produces 42 Volts instead of 130 Volts bias voltage 2. 40.00 MHz: oscillates at 1/3 frequency. So it seems as if the oscillator circuit oscillates with less efficiency, perhaps a transistor with a higher fT is required? 73 Sverre LA3ZA http://www.qsl.net/la3za/ ___ Elecraft mailing list Post to: Elecraft@mailman.qth.net 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
