K2 unadvertised feature ?
Locked
 
|
Does the K2 have a feature where a sufficiently high V RFDET voltage causes the 8R line to go low, thus switching the PIN (1N4007) diodes and protecting the receiver from damage if high power is applied to the antenna socket while receiving ?
If so this would explain the behaviour I have been getting due to RF getting into the K60XV, which I posted at http://mailman.qth.net/pipermail/elecraft/2009-January/104241.html . I've done some more measurements with everything disconnected from the K2 except for 12v power and an alligator clip lead used as an antenna connected to the transverter out socket. Transmitting on another rig on 6m still causes the transverter key out line to be keyed, and I have confiirmed that the 8R line in the K2 is going low. It seems to take around 100 microwatts to make this happen. The only explanation that seems to make sense is that this power is detected (V RFDET) and the K2 microprocessor reacts to this by setting 8R low. Does this theory make sense ? 73, Steve VE3SMA __________________________________________________________________ Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your favourite sites. Download it now at http://ca.toolbar.yahoo.com. ______________________________________________________________ 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 |
|
|
Steve,
While something is happening, I don't think it is happening the way you surmise. If the 8R line in the K2 goes low, it is because the K2 has entered the transmit state. As far as I know, the presence of voltage on the VRFDET line will not cause the K2 to enter transmit state. As a further thought, the driver for VRFDET on the K60XV is a tri-state driver and must have an enable input to drive the VRFDET line. I do not doubt that something is happening in your K2 due to the RF input to the low power TX line, but if you do not have a transverter band selected on the K2, the input should be grounded by the K2 relay on the K60XV option. If you *do* have the K2 set to a transverter band and the low power output is selected, then yes, the RF input on the output line could be coupled back into the K2 bandpass filter - the obvious result should be receiver overload, but not changing into transmit state. If your RF field is strong enough, it could cause the K2 to do 'strange things', and going into transmit state could be one of those 'strange things', but I doubt the mechanism is through the VRFDET signal - that line only controls the power output, it does not place the K2 in transmit state by itself - there is another cause. Without measurement, I can only speculate about it, but I would suspect something is creating confusion on the AUXBUS signaling line which causes the K2 microprocessor to enter the transmit state. I would suggest that you work to reduce the stray RF energy in the shack, and if that is not possible, try ferrite cores on all inputs to the K2 including the power connection. Try running the K2 and the 6 meter transmitter from different power supplies, and/or try bonding the other transmitter to the K2 to see if there is any difference. Often slight ground potential differences can cause "strange happenings". 73, Don W3FPR Steve Kavanagh wrote: > Does the K2 have a feature where a sufficiently high V RFDET voltage causes the 8R line to go low, thus switching the PIN (1N4007) diodes and protecting the receiver from damage if high power is applied to the antenna socket while receiving ? > > If so this would explain the behaviour I have been getting due to RF getting into the K60XV, which I posted at > > http://mailman.qth.net/pipermail/elecraft/2009-January/104241.html . > > I've done some more measurements with everything disconnected from the K2 except for 12v power and an alligator clip lead used as an antenna connected to the transverter out socket. Transmitting on another rig on 6m still causes the transverter key out line to be keyed, and I have confiirmed that the 8R line in the K2 is going low. It seems to take around 100 microwatts to make this happen. The only explanation that seems to make sense is that this power is detected (V RFDET) and the K2 microprocessor reacts to this by setting 8R low. > > Does this theory make sense ? > > 73, > Steve VE3SMA > > > __________________________________________________________________ > Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your favourite sites. Download it now at > http://ca.toolbar.yahoo.com. > ______________________________________________________________ > 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 > ------------------------------------------------------------------------ > > > No virus found in this incoming message. > Checked by AVG - www.avg.com > Version: 8.0.233 / Virus Database: 270.10.16/1926 - Release Date: 01/30/09 17:31:00 > > 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 |
|
|
Thanks for taking the time to consider my problem, Don. Here's a few more observations.....
> If the 8R line in the K2 goes low, it is because the K2 has > entered the transmit state. You'd think so, but since the display does not show a frequency change between the receive state and the anomalous state when the RIT is on (and offset) this is not entirely the case. > As a > further thought, the driver for VRFDET on the K60XV is a > tri-state driver and must have an enable input to drive the > VRFDET line. > > if you do not > have a transverter band selected on the K2, the input should > be grounded by the K2 relay on the K60XV option. > > If you *do* have the K2 set to a transverter band and the > low power output is selected, then yes, the RF input on the > output line could be coupled back into the K2 bandpass > filter - the obvious result should be receiver overload, but > not changing into transmit state. I am thinking of the VRFDET detector in the K60XV, not the one on the K2 RF board, as the source of my problem. Its enable signal is the 6LP power supply to Q2/U2, which is on in both receive and transmit states. The VRFDET output from Q2 (P1-pin7) is about 0.4 V in receive and goes up to over 2 V when I transmit on 6m, so there is certainly RF being detected. In any case, when I switch to a transverter band which does not use the low power output (thus turning off the 6LP voltage), the problem goes away completely. I tried connecting the cliplead "antenna" to many points on the K60XV (with the cover off the K2), including XVTR IN, all 3 pins of Q1, P1-pin1 (currently fitted with an extra 4.7nF bypass to pin2), and the junction of D1 & D2. The only one which resulted in the anomalous state was XVTR OUT. I'm thinking I should try the VRFDET detector diode on the K2 RF board, though. > I would suggest that you work to reduce the stray RF energy > in the shack, and if that is not possible, It really isn't practical in my apartment station or in the rover vehicle. Space limitations force operation close to the antennas. > try ferrite cores > on all inputs to the K2 including the power connection. In this simplified K2-only test configuration the only connections to the K2 are the pickup antenna to the XVTR OUT and the power connection, and the power lead does in fact have a clip-on ferrite choke in use for this testing, which ought to do something at 50 MHz. > Try running the K2 and the 6 meter transmitter from different > power supplies, The power supplies (at the home station) both plug into the same AC power bar, but otherwise the 6m setup is completely separate. > and/or try bonding the other transmitter to > the K2 to see if there is any difference. Often slight > ground potential differences can cause "strange happenings". Agreed. I have tried various different ground connections, and different sensitivities to the RF fields result, presumably due to different RF potentials on the K2 case compared to that on the XVTR OUT jack. But there was no magic cure, so far. And this situation will be completely different in the car, anyway...I need to find a more robust solution. I am thinking of perhaps some sort of bandpass or low pass filter in the TX IF line, though the design is complicated by the fact that the terminating impedances aren't all 50 ohms. 73, Steve VE3SMA __________________________________________________________________ Instant Messaging, free SMS, sharing photos and more... Try the new Yahoo! Canada Messenger at http://ca.beta.messenger.yahoo.com/ ______________________________________________________________ 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 |
|
|
Steve,
The VRFDET is secondary. If the 8R voltage rail is going low, the K2 is either going into transmit state or the microprocessor is dropping its RX output. The control for 8R is on the K2 Control Board - Q2 and Q4. Check the gate of Q4 to see if the voltage there drops - that causes 8R to drop. Also check the state of the TX signal from the microprocessor (gate of Q3) - if that signal is above 3 volts and the RX signal is zero, the K2 is definitely in transmit state, whatever the cause. The K60XV uses the 8R input dropping to switch its circuits from receive to transmit, and will produce energy at the output - it is operating into an open load (your wire antenna), and as a result, VRFDET will become quite high because the output is measured assuming a 50 ohm load on that K60XV low power output. Put a piece of coax on the transverter output terminated in 50 ohms and see if the behavior continues - that should simulate normal operating conditions better than your wire. If the behavior continues with the coax attached, you will have to find how the signal is being coupled into the K2. If you find the microprocessor TX output on and RX off, then I would look at possible coupliing from the PTT input or the paddle input (the PTT and DOT paddle lines are the same inside the K2), and RF coupling into them could induce 'strange happenings' even though the DC voltages may be close to the normal 5 volt level. My guess that the PTT input may be the culprit is only a guess, but it is possible for severe noise (high RF levels) on the PTT line to cause the microprocessor to become 'mixed up'. 73, Don W3FPR Steve Kavanagh wrote: > Thanks for taking the time to consider my problem, Don. Here's a few more observations..... > > >> If the 8R line in the K2 goes low, it is because the K2 has >> entered the transmit state. >> > > You'd think so, but since the display does not show a frequency change between the receive state and the anomalous state when the RIT is on (and offset) this is not entirely the case. > > >> As a >> further thought, the driver for VRFDET on the K60XV is a >> tri-state driver and must have an enable input to drive the >> VRFDET line. >> >> if you do not >> have a transverter band selected on the K2, the input should >> be grounded by the K2 relay on the K60XV option. >> >> If you *do* have the K2 set to a transverter band and the >> low power output is selected, then yes, the RF input on the >> output line could be coupled back into the K2 bandpass >> filter - the obvious result should be receiver overload, but >> not changing into transmit state. >> > > I am thinking of the VRFDET detector in the K60XV, not the one on the K2 RF board, as the source of my problem. Its enable signal is the 6LP power supply to Q2/U2, which is on in both receive and transmit states. The VRFDET output from Q2 (P1-pin7) is about 0.4 V in receive and goes up to over 2 V when I transmit on 6m, so there is certainly RF being detected. In any case, when I switch to a transverter band which does not use the low power output (thus turning off the 6LP voltage), the problem goes away completely. > > I tried connecting the cliplead "antenna" to many points on the K60XV (with the cover off the K2), including XVTR IN, all 3 pins of Q1, P1-pin1 (currently fitted with an extra 4.7nF bypass to pin2), and the junction of D1 & D2. The only one which resulted in the anomalous state was XVTR OUT. I'm thinking I should try the VRFDET detector diode on the K2 RF board, though. > > >> I would suggest that you work to reduce the stray RF energy >> in the shack, and if that is not possible, >> > > It really isn't practical in my apartment station or in the rover vehicle. Space limitations force operation close to the antennas. > > >> try ferrite cores >> on all inputs to the K2 including the power connection. >> > > In this simplified K2-only test configuration the only connections to the K2 are the pickup antenna to the XVTR OUT and the power connection, and the power lead does in fact have a clip-on ferrite choke in use for this testing, which ought to do something at 50 MHz. > > >> Try running the K2 and the 6 meter transmitter from different >> power supplies, >> > > The power supplies (at the home station) both plug into the same AC power bar, but otherwise the 6m setup is completely separate. > > >> and/or try bonding the other transmitter to >> the K2 to see if there is any difference. Often slight >> ground potential differences can cause "strange happenings". >> > > Agreed. I have tried various different ground connections, and different sensitivities to the RF fields result, presumably due to different RF potentials on the K2 case compared to that on the XVTR OUT jack. But there was no magic cure, so far. And this situation will be completely different in the car, anyway...I need to find a more robust solution. I am thinking of perhaps some sort of bandpass or low pass filter in the TX IF line, though the design is complicated by the fact that the terminating impedances aren't all 50 ohms. > > 73, > Steve VE3SMA > > 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 for the quick reply ! I'd reply off list with the nitty gritty details but I did have a request from another interested ham to make the results public. I have now put together a much more controlled test setup for the K2 that doesn't involve high levels of RF all over the shack ! I now have a low power crystal oscillator/multiplier source at about 146 MHz (which puts about 0.5 mW into a 50 ohm load) connected to the XVTR OUT jack connected in parallel with a 50 ohm load using a BNC Tee just behind the rear panel. This reliably reproduced the observed effects (KEY OUT going to ground, receiver quiet and S-meter to maximum), and simulates the loading effect of having a transverter connected. > Check the gate of Q4 to see if the voltage there drops - > that causes 8R to drop. Yes it does drop when I key the 146 MHz test source, and the collector of Q2 (8R) does too. > Also check the state of the TX signal from the > microprocessor (gate of Q3) That voltage remains unchanged at approximately zero volts, as does the collector of Q1 (8T). So clearly the K2 is switching into an odd state. > My guess that the PTT > input may be the culprit is only a guess, but it is possible > for severe noise (high RF levels) on the PTT line to cause > the microprocessor to become 'mixed up'. If the K2 is in CW mode it does not start sending dots, so the PTT line is not going low. Which doesn't mean it doesn't have RF on it, but we are now talking about a pretty low level (500 microwatts) of generally shielded RF so I would be surprised if stray coupling to the PTT line would be the culprit. Back to head scratching..... 73, Steve VE3SMA __________________________________________________________________ Instant Messaging, free SMS, sharing photos and more... Try the new Yahoo! Canada Messenger at http://ca.beta.messenger.yahoo.com/ ______________________________________________________________ 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 |
|
|
Steve,
But I don't think a transverter should normally pump RF back into the K2 transverter output. Are you trying to simulate RF pickup on the coax between the K2 transverter output and the transverter input? If that is what you are trying to simulate, I would feel more comfortable having the high RF field present on the coax shield (with the coax terminated at the far end), and no signal pumped backwards into the K2. You are correct - if both TX and RX are off, both 8T and 8R will be at zero volts (no power to the early stages of the K2 transmitter and no power to many parts of the receiver. You are also correct that noise as I suggested on the PTT/DOT line would cause the K2 to go into transmit - it is not, it is just going 'out of receive'. You will probably have to poke around with an RF Probe or an adequate oscilloscope to determine which lines have RF on them in order to determine what circuits in the K2 (or on the K60XV) are hot with RF. With the information I have received thus far, I would be more suspicious of the AUXBUS signal line, because that is a signal the communicates directly between the various processors in the K2, and clearly the microprocessor is doing 'strange and unusual things'. It may take a K2 firmware expert (Wayne) to figure out what is happening in the microprocessor to cause RX to drop. 73, Don W3FPR Steve Kavanagh wrote: > Don: > > Thanks for the quick reply ! > > I'd reply off list with the nitty gritty details but I did have a request from another interested ham to make the results public. > > I have now put together a much more controlled test setup for the K2 that doesn't involve high levels of RF all over the shack ! I now have a low power crystal oscillator/multiplier source at about 146 MHz (which puts about 0.5 mW into a 50 ohm load) connected to the XVTR OUT jack connected in parallel with a 50 ohm load using a BNC Tee just behind the rear panel. This reliably reproduced the observed effects (KEY OUT going to ground, receiver quiet and S-meter to maximum), and simulates the loading effect of having a transverter connected. > > >> Check the gate of Q4 to see if the voltage there drops - >> that causes 8R to drop. >> > > Yes it does drop when I key the 146 MHz test source, and the collector of Q2 (8R) does too. > > >> Also check the state of the TX signal from the >> microprocessor (gate of Q3) >> > > That voltage remains unchanged at approximately zero volts, as does the collector of Q1 (8T). > > So clearly the K2 is switching into an odd state. > > >> My guess that the PTT >> input may be the culprit is only a guess, but it is possible >> for severe noise (high RF levels) on the PTT line to cause >> the microprocessor to become 'mixed up'. >> > > If the K2 is in CW mode it does not start sending dots, so the PTT line is not going low. Which doesn't mean it doesn't have RF on it, but we are now talking about a pretty low level (500 microwatts) of generally shielded RF so I would be surprised if stray coupling to the PTT line would be the culprit. > > Back to head scratching..... > > 73, > Steve VE3SMA > > 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 Steve Kavanagh
I just tried another test, which I think tends to confirm my theory in some way (or that something is amiss). When I apply my 0.5 mW 146 MHz test source to the XVTR OUT jack, the DC voltage at the cathode of D3 in the K60XV (the detector for VRFDET) increases from zero to about 0.1 volts. So I put together a resistor network, which when powered from a 12V battery produced this same DC voltage, and applied this DC voltage to the cathode of D3, with no RF input signals at all present. The same symptoms were observed (KEY OUT is keyed, receiver goes quiet and S-meter goes to maximum).
73, Steve VE3SMA __________________________________________________________________ Yahoo! Canada Toolbar: Search from anywhere on the web, and bookmark your favourite sites. Download it now at http://ca.toolbar.yahoo.com. ______________________________________________________________ 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 Elecraft
|
1 view|%1 views
| Free forum by Nabble | Edit this page |