K2: KPA100 Removal
Locked
 
|
I would like to put my K2/100 back to a basic K2 as part of debugging some
problems I've had around power and RFI distortion. I have the K2, KPA100, KSB2, and KNB2 installed. My plan is to only remove the KPA100 at first to see if that corrects the problems I've been having. As I understand it, removal of the KPA100 involves only the following: 1. Disconnect the ribbon cable that goes between the KPA100 and P4 on the control board 2. Disconnect the Aux 12v connector that goes between the KPA100 and the Aux 12v jack on the main board 3. Disconnect the Aux RF connector that goes between the KPA100 and the Aux 12v jack on the main board 4. Disconnect the SPKR connector that goes between the KPA100 and the SPKR jack on the main board 5. Install the SPKR connector from the original top cover to the SPKR jack on the main board and screw on the original top cover (in an uncharacteristic fit of planning, I did keep the original top cover). 6. Uncover the BNC RF out jack on the K2 base and use that in place of the SO-239 on the KPA100. My original top cover did have the KIO2 installed for interface to the KAT100, and it is my understanding that I can simply plug that into P4 on the control board. Is there anything else I need to do? For reference, my problems at the moment are as follows: 1. I get RFI distortion on digital and SSB on the lower frequencies (40 and 80 meters). I get the distortion when feeding the K2 directly into a dummy load and I get it when feeding the K2 into a Buddipole about 100' from my shack. I do not get the distortion when running an FT-897 in similar circumstances. 2. The output power is not consistent. On tune, my MFJ 941E and Elecraft W1 usually read about twice what the K2 says it is outputting (about 20w). On CW, the readings per the MFJ and W1 disagree with what is dialed into the K2 although the amount and direction of the difference varies by band. 2. Going straight into a 50 ohm dummy load, (no MFJ, no W1), the K2 is reading SWRs of 5: or 6:1 in tune mode. When I put the MFJ and the W1 in the line, they both read 1:1 even though the K2 continues to read 5: or 6:1. I believe that the MFJ and W1 are reasonably accurate based on testing against other devices. Thanks, Jon KB1QBZ ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
Jon,
You have the KPA100 removal steps correct, but I suggest that you do not do that yet. Instead do a quick test - remove the APP connector from the KPA100 and power the K2 from the coaxial power jack. Then set the power at 5 watts and measure the power output with your external wattmeter (connected into a dummy load). Do the same but with the power set to 2 watts. Did you get 10 or 12 watts in both those tests? Or did you get something close to the requested power? If your actual power was much greater than the requested power, the most likely cause is that diodes D16 and D17 in the KPA100 wattmeter have been damaged. Remove the KPA100 physically and repeat the test with the coax to the wattmeter connected to the BNC ANT jack on the base K2. If you have normal power control, then the KPA100 wattmeter diodes are the culprit. Those diodes are used for power control when the KPA100 is installed, and if damaged, the K2 and KPA100 will go to full power output which will likely put it into a non-linear condition. The other thing that will cause distortion in SSB is the KPA100 ribbon cable and speaker cable - if it is close to the KSB2 board, some coupling can occur. To easily position the cables, remove the right side panel and tuck the cables into the space behind the heat sink and at the top of the control board. Keep the cables as far away from the KSB2 board as possible. 73, Don W3FPR On 3/27/2011 2:11 PM, Jon Perelstein wrote: > I would like to put my K2/100 back to a basic K2 as part of debugging some > problems I've had around power and RFI distortion. I have the K2, KPA100, > KSB2, and KNB2 installed. My plan is to only remove the KPA100 at first to > see if that corrects the problems I've been having. > > As I understand it, removal of the KPA100 involves only the following: > > 1. Disconnect the ribbon cable that goes between the KPA100 and P4 on the > control board > > 2. Disconnect the Aux 12v connector that goes between the KPA100 and the > Aux 12v jack on the main board > > 3. Disconnect the Aux RF connector that goes between the KPA100 and the Aux > 12v jack on the main board > > 4. Disconnect the SPKR connector that goes between the KPA100 and the SPKR > jack on the main board > > 5. Install the SPKR connector from the original top cover to the SPKR jack > on the main board and screw on the original top cover (in an > uncharacteristic fit of planning, I did keep the original top cover). > > 6. Uncover the BNC RF out jack on the K2 base and use that in place of the > SO-239 on the KPA100. > > My original top cover did have the KIO2 installed for interface to the > KAT100, and it is my understanding that I can simply plug that into P4 on > the control board. > > Is there anything else I need to do? > > For reference, my problems at the moment are as follows: > > 1. I get RFI distortion on digital and SSB on the lower frequencies (40 and > 80 meters). I get the distortion when feeding the K2 directly into a dummy > load and I get it when feeding the K2 into a Buddipole about 100' from my > shack. I do not get the distortion when running an FT-897 in similar > circumstances. > > 2. The output power is not consistent. On tune, my MFJ 941E and Elecraft > W1 usually read about twice what the K2 says it is outputting (about 20w). > On CW, the readings per the MFJ and W1 disagree with what is dialed into > the K2 although the amount and direction of the difference varies by band. > > 2. Going straight into a 50 ohm dummy load, (no MFJ, no W1), the K2 is > reading SWRs of 5: or 6:1 in tune mode. When I put the MFJ and the W1 in > the line, they both read 1:1 even though the K2 continues to read 5: or 6:1. > > I believe that the MFJ and W1 are reasonably accurate based on testing > against other devices. > > Thanks, > Jon > KB1QBZ > ______________________________________________________________ > Elecraft mailing list > Home: http://mailman.qth.net/mailman/listinfo/elecraft > Help: http://mailman.qth.net/mmfaq.htm > Post: mailto:[hidden email] > > This list hosted by: http://www.qsl.net > Please help support this email list: http://www.qsl.net/donate.html > Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
Don,
Thanks as always for your reply. But you don't seriously believe that my problems with electronic equipment could be THAT simple, do you?? After all, I'm the guy whose dog ate one of the ICs while I was building the K2 (no joke). I ran 5 watts and 2.5 watts, and ran each test five times (first 5, then 2.5, then 5, then 2.5, etc.). With 5 watts commanded, both power meters usually read about 12 watts, the K2's RF display read 5, and both the K2's ammeter and an external WattsUp meter read about 2.7 amps. HOWEVER, on one try, both power meters were indicating 15 watts (with about 2.8 amps on the ammeters and 7 on the RF display). With 2.5 watts commanded, the MFJ consistently read 7 watts, and the W1 consistently read between 6 and 8 (it doesn't have a display for 7). The K2's RF display flickered between 2 and 3, and both the K2's ammeter and the WattsUp read about 2.1 amps. You had me run this test sometime late last year because I was reporting excessive heat when running in digital modes (e.g., PSK31). At that time, the numbers were a lot closer to what was being commanded -- although back then I was running it with the KPA100 in use (trying 100 watts, 50 watts, and 25 watts). This started today because I was trying to tuck up the ribbon cable and the speaker wire, which you had suggested to me a while back. I hadn't actually tried that until today, and I noticed the power and SWR problems after I buttoned everything back up and was testing the change. I've put the ribbon cable and the speaker wire back where they were to begin with, but am still having the power and SWR problems, so my guess is that I managed to mess something up while I had the case open, but I don't see anything offhand when looking at it now. Jon ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
John,
I believe after you replace diodes D16 and D17 in the KPA100 (both are 1N5711), you will find the problems go away. What appears to be happening is that they are not fully open nor shorted, but are at some state in between, so you do get some control, but it is just not right. Replace the diodes, then re-balance the wattmeter (adjust C1 with a non-metallic tool), and re-calibrate the forward power at about 80 watts into a dummy load. You might just want to have 4 diodes on hand in case the two in the KAT100 have also been damaged. 73, Don W3FPR On 3/27/2011 4:30 PM, Jon Perelstein wrote: > Don, > > Thanks as always for your reply. > > But you don't seriously believe that my problems with electronic > equipment could be THAT simple, do you?? After all, I'm the guy whose > dog ate one of the ICs while I was building the K2 (no joke). > > I ran 5 watts and 2.5 watts, and ran each test five times (first 5, > then 2.5, then 5, then 2.5, etc.). > > With 5 watts commanded, both power meters usually read about 12 watts, > the K2's RF display read 5, and both the K2's ammeter and an external > WattsUp meter read about 2.7 amps. HOWEVER, on one try, both power > meters were indicating 15 watts (with about 2.8 amps on the ammeters > and 7 on the RF display). > > With 2.5 watts commanded, the MFJ consistently read 7 watts, and the > W1 consistently read between 6 and 8 (it doesn't have a display for > 7). The K2's RF display flickered between 2 and 3, and both the K2's > ammeter and the WattsUp read about 2.1 amps. > > You had me run this test sometime late last year because I was > reporting excessive heat when running in digital modes (e.g., PSK31). > At that time, the numbers were a lot closer to what was being > commanded -- although back then I was running it with the KPA100 in > use (trying 100 watts, 50 watts, and 25 watts). > > This started today because I was trying to tuck up the ribbon cable > and the speaker wire, which you had suggested to me a while back. I > hadn't actually tried that until today, and I noticed the power and > SWR problems after I buttoned everything back up and was testing the > change. I've put the ribbon cable and the speaker wire back where > they were to begin with, but am still having the power and SWR > problems, so my guess is that I managed to mess something up while I > had the case open, but I don't see anything offhand when looking at it > now. > > Jon Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
OK, thanks Don.
I'll order the diodes (local Radio Shack doesn't carry them anymore). In the meantime, I think I will remove the KPA100 anyway to see what impact that has on the RFI problem. Do you think that this is also impacting the SWR reading that I'm getting in tune? Both the MFJ and the W1 say that I'm good (I would hope so into a dummy load), but the K2 is reporting 5 or 6:1 Jon ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
Of course it is Jon, one diode rectifies the forward power and the
other the reverse power. BUT - I hope you do not have the KAT100 in line between the KPA100 and the external wattmeter when you make that evaluation - if it is, you should first put it in CAL mode using the K2 menu. 73, Don W3FPR On 3/27/2011 4:55 PM, Jon Perelstein wrote: > OK, thanks Don. > > I'll order the diodes (local Radio Shack doesn't carry them anymore). > In the meantime, I think I will remove the KPA100 anyway to see what > impact that has on the RFI problem. > > Do you think that this is also impacting the SWR reading that I'm > getting in tune? Both the MFJ and the W1 say that I'm good (I would > hope so into a dummy load), but the K2 is reporting 5 or 6:1 > > Jon > > Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
Duhhhhh! Sorry, I wasn't thinking when I asked that.
The KAT100 is out of the circuit completely. All my testing has been with the MFJ-941E. I just removed the KPA100 and tested. My power levels are as commanded (usually a little better, but not double) AND my PSK31 signal looks like a PSK31 signal again. While I was testing on 17 meters (5 watts PSK31), a station in Slovenia came back to me and gave me a 559 with a very good IMD. Given that digital modes are continuous duty, what power level should I maintain with the bare K2. My general rule of thumb is 1/3 of max power (which is about 15 watts on my bare K2), so I'm thinking 5 watts for digital. What do you think? Jon ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
Jon,
Your 1/3 rule is right for continuous duty data modes like RTTY, but I understand (and have not tried to figure out the details for myself) that plain PSK31 does not come up to a full duty cycle - something about the coding, so you might go up to 1/2 the normal rated power - 6 or 7 watts. If you push it more than that, put your hand on the bottom rear cover once in a while. If it is too hot to hold your finger on it constantly, then you should back off the power a bit. 73, Don W3FPR On 3/27/2011 6:23 PM, [hidden email] wrote: > Duhhhhh! Sorry, I wasn't thinking when I asked that. > > The KAT100 is out of the circuit completely. All my testing has been > with the MFJ-941E. > > I just removed the KPA100 and tested. My power levels are as commanded > (usually a little better, but not double) AND my PSK31 signal looks > like a PSK31 signal again. While I was testing on 17 meters (5 watts > PSK31), a station in Slovenia came back to me and gave me a 559 with a > very good IMD. > > Given that digital modes are continuous duty, what power level should > I maintain with the bare K2. My general rule of thumb is 1/3 of max > power (which is about 15 watts on my bare K2), so I'm thinking 5 watts > for digital. What do you think? > > Jon Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
In reply to this post by Jon Perelstein
A couple of other points to consider:
1. I find that my K3's finals get hotter in PSK31 at 50% power than they do in RTTY at 100% power, but that's very likely due to the fact that my RTTY QSOs are in contests, with very short transmissions, whereas my ragchew-style transmissions in PSK31 last a lot longer before giving the finals a break to cool off between transmissions. Of course, the breaks also last longer during ragchews, but the peak temperatures will still be higher with longer transmissions. Depending on your operating and conversational style, this may be a factor. 2. PSK31 is very intolerant to non-linearity. The "crest factor" of PSK31 can be up to 2, i.e. the peak power can be twice the average power. Since it is clipping of the peaks that causes splatter, with most radios you need to keep the output power below one-half the rated maximum in order to ensure linearity. Linearity is not a problem in RTTY, but it matters in PSK31. 73, Rich VE3KI Don W3FPR wrote: > Your 1/3 rule is right for continuous duty data modes like RTTY, but I > understand (and have not tried to figure out the details for myself) > that plain PSK31 does not come up to a full duty cycle - something about > the coding, so you might go up to 1/2 the normal rated power - 6 or 7 watts. > > If you push it more than that, put your hand on the bottom rear cover > once in a while. If it is too hot to hold your finger on it constantly, > then you should back off the power a bit. > ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
In reply to this post by Don Wilhelm-4
Don,
I'm trying to replace D16 and D17 in the KPA100 but I'm having trouble removing the heat sink so that I can get at them (it looks to me as if I have to remove the heat sink to get at the bottom of the PC board to unsolder them). I have removed the three black pan head screws that attach the rear panel to the heat sink. I have removed the seven standoffs that screw onto the set screws that come out of the heat sink and hold it to the PC board. I have removed the four (2 each) 4-40 x 1/4" screws at Q1 and Q2 (2 each) that hold Q1 and Q2 to the heat sink. I have removed the two 4-40 x 1/4" screws that hold Q3 and Q4 to the heat sink. When I try to remove the heat sink from the board, I'm getting resistance from the center of the board, near Q1 and Q2. I don't want to just pull in case there's something that I've missed removing. Did I miss something, or is this one of those cases where a sledgehammer is appropriate? Well, okay, maybe not a sledgehammer... Thanks. Jon WB2RYV (ex KB1QBZ) On Sun, Mar 27, 2011 at 4:50 PM, Don Wilhelm <[hidden email]> wrote: > John, > > I believe after you replace diodes D16 and D17 in the KPA100 (both are > 1N5711), you will find the problems go away. ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
|
|
John
If the board has not been removed a few times, it often sticks on the screw threads - I suspect that is the problem. I usually use a flat-blade screwdriver between the heatsink and the board to carefully nudge it up. Watch where the screwdriver is going (peer between the board and the heatsink) so you do not damage any of the capacitors mounted under the board. Usually when you lift the board, the standoff near the fan is the troublesome one, but the other center ones can be troublesome at times. 73, Don W3FPR On 4/10/2011 2:51 PM, Jon Perelstein wrote: > Don, > > I'm trying to replace D16 and D17 in the KPA100 but I'm having trouble > removing the heat sink so that I can get at them (it looks to me as if > I have to remove the heat sink to get at the bottom of the PC board to > unsolder them). > > I have removed the three black pan head screws that attach the rear > panel to the heat sink. > > I have removed the seven standoffs that screw onto the set screws that > come out of the heat sink and hold it to the PC board. > > I have removed the four (2 each) 4-40 x 1/4" screws at Q1 and Q2 (2 > each) that hold Q1 and Q2 to the heat sink. > > I have removed the two 4-40 x 1/4" screws that hold Q3 and Q4 to the > heat sink. > > When I try to remove the heat sink from the board, I'm getting > resistance from the center of the board, near Q1 and Q2. I don't want > to just pull in case there's something that I've missed removing. Did > I miss something, or is this one of those cases where a sledgehammer > is appropriate? > > Well, okay, maybe not a sledgehammer... > > Thanks. > > Jon > WB2RYV (ex KB1QBZ) > > On Sun, Mar 27, 2011 at 4:50 PM, Don Wilhelm <[hidden email] > <mailto:[hidden email]>> wrote: > > John, > > I believe after you replace diodes D16 and D17 in the KPA100 (both > are 1N5711), you will find the problems go away. > Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:[hidden email] This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html |
«
Return to [K2]
|
1 view|%1 views
| Free forum by Nabble | Edit this page |