The paddle on the right is 1”. You can look at my QSL page for more information.?
--
Colin Kaminski - K6JTH
https://www.qsl.net/k6jth/

On Mon, Aug 22, 2022 at 12:00 AM, KK4ITX John wrote:

Go over your frequency settings, they too are just as important and as you know be aware of the 1500 offset.

I have had several discussions with users who want to TX at the higher end of the band and with these typical descriptions, they go "above it".
They hear 1500 Hz off set and 200 Hz receive band and end up using a value >1600 fro the dial.?
I think we should make a point of saying 1500 Hz +/- 100 Hz, rather than just 1500 Hz offset.

Others always set their TX frequency RIGHT ON 1500 Hz and get caught in the large pileup there.

I am not saying that is the issue here, but since no TX frequency was talked about, something else to consider.

FYI:? WSPRNET frequency page (), it shows a table - could have been spaced better (IMHO).
???????? This version is clearer: .

--
Rob KB8RCO

On Mon, Aug 22, 2022 at 01:39 PM, bojan.naglic wrote:

I disconnected the T1 primary and used a 10k to bias the IC3 output and took the BPF output via a 100n cap.

Sorry, T2 primary obviously. My mistake.

I can see that Hans quite did the job with BPFs, quite...? ?I saw that I still have to adjust L12 for 30m BPF.? But after I change the IC3.? Hot air station rules here.

73 Bojan S53DZ

On the QDX rev 3a PCB board, capacitor C41 is completely shorted to GND, because of a PCB board fault. The latest assembly manual (rev 1.12) shows on page 13 how to fix this.
My question: How will the behaviour/performance be different after fixing C41? Yes, the related continuity to GND will be gone, but what else will change?
73 de HA7RJA

Hi Geoff,?
Yes, as I measured that resistance between pin 10 to 13, I removed C30.?

I left unassembled the 80m LPF, so I just bypassed it and band config it to all bands during measurement of BPFs.? I disconnected the T1 primary and used a 10k to bias the IC3 output and took the BPF output via a 100n cap.
I made a VNA measurements on 50 Ohm so there is also one thing that prevent me to measure the precise attenuation.?
I know also that connecting BPF and LPF in series makes difference however I have tried this in Elsie first and seen the difference is quite low.? Larger difference is probably by connecting back T1 due to impedances and not the L.

I hope to get better response on the 20m BPF after I change IC3.??

73 Bojan S53DZ

开云体育

Those are beautiful I. Their simplicity and size, assuming the paddle on the right is the size of a nickel (?).??

I'd be interested?


72,

Ted
k3RTA

How did you measure the bandpass filter and low pass filters separately ? Remove diodes D1-6 to get access to LPF but receive is a bit more difficult.
When I did a simulation with Elsie, I found that the response was not correct until the LPF components were included. Others have commented on this and Hans has admitted that it was a grope and hope solution. Did you see the 100R between IC3 pins 10 and 13 with C30 removed? With C30 removed, the receive path will be open circuit. My receive response on all bands other than 20m are all greater than 60dB down which is what I would expect. The LPF's have been bypassed with wire links so that in the event of an accidental transmit, the PA would see the dummy load.
I have only fitted the filter components for 20m leaving the other band slots clear for future expansion for 15m, 17m, 12m and 10m. The receive filter response peaks at about 17MHz with the 14MHz being about 7dB down. Adding extra inductance and capacitance ( 4 turns on L12 and 30pF to C29 ) made very little change to the response.
As I only fitted 20m for testing and it works OK, I shall leave it as it is. It receives WSPR fine ( not transmitted as 4 watts is much too high ) and I have had a few contacts on FT8 ( had to lie down in a darkened room afterwards to relieve the stress )

On 22/08/2022 08:16, Jon Pike wrote:

But, get the following in the log:
[RIGCTRL][2022-08-01 05:17:34.128432][00:00:00.447861][info] Hamlib version: Hamlib 4.5~git Sat Jan 01 23:05:51 2022 +0000 SHA=18548e
[RIGCTRL][2022-08-01 05:17:36.335333][00:00:02.653966][error] kenwood_open: not the right driver apparently (found 2002, asked for 2028, checked TS-440S)
I believe I get the same or similar if I try TS-480.

Jon

I do not know, set the polling to 99 seconds in WSJT-X, might reduce the number.
Can't think why Hamlib wants a driver.
This is the latest general release of WSJT-X??

I think ideally the QDX needs a separate hamlib definition with no polling.
Maybe we should see if there is a response in the WSJT-X group, tests and mods to Hamlib are made there if picked up.

I see large numbers of these errors talking about missed audio samples:
[SYSLOG][2022-08-01 05:17:59.407122][00:00:25.725703][warning] Detected dropped audio source samples: 720 (0.015 S)

This is normal and is not important unless the dropped samples rise much higher and the delay gets up to 0.5 S>

I made a few notes about syslog here

I also sometimes notice a loud pop, in my pc speakers,? when the QDX goes from TX to RX.? Maybe more noticeable on 20m than the lower bands, or gets quieter on 80m? Possibly related?

Maybe the RF switch-on/off?
My QDX does not switch cleanly, two bursts at TX start, I've not noted any clicks but I think it could cause them. There could be a similar abrupt switch at the end?
Or, something to do with the PSU?

73 Alan G4ZFQ

Hi Marc,

I would hold off on installing the BS170s.? You can test the receive function without them; their failure can cause more severe problems during testing if the rig goes into transmit.?

Have you verified the audio input selection in WSJT-X?? Note that the Terminal program only uses the serial connection to the CAT functions in the QDX.? There is no testing of the audio interface to the PC.

Does the input meter of the WSJT-X software change when you disconnect and reconnect the antenna?? I wonder if there is an issue with C2 or the connection from the BNC to the LPF selection diodes.? If you have leaded 2uF or so capacitor, can you parallel C2 to see if the level meter in the WSJT-X software increases and maybe starts to decode?

Did you check the LPF selection?? Verify that the band selected has the diodes on for the correct filter and that the other 4 are off.? For example, if in the RF Sweep function of the terminal program, diodes D2 and D5 should read close to zero on the cathode (banded) side.? D1, D3, D4, and D6 should read close to 5volts.? I apologize if you have done this already.? I am losing track of who has done what as others jump into the thread.

A brief review of what I think you have tested and the current state:
- You have pulled the BS170s and found them all to be OK
- You do not know the current in the Receive mode as your meter has malfunctioned.? I base this on a .19ma reading as that is almost an order of magnitude off.? Unless that was meant to be 0.19 amps, then it is too high.
- The Audio and RF Sweeps look normal.? You have not listed the magnitude of the sweeps on the band's center frequency.? It should be above -10db
- WSJT-X does not decode any signals, while another rig can decode from the same antenna
- IC5 does put out 2.5volts on both sets of gates when transmitting

If the above is correct, verifying the antenna to the LPF selection connection is the only thing I can think of to explain the lack of decoding.? I base this on your reporting the sweeps look normal.? The sweeps would test the signal path from the LPF side of C2 through to the audio input into the MPC (IC8).


73
Evan
AC9TU

Hi Ed, I had same problem with my QCX Mini 20 meter. The number of spots I was getting was very low and all showed a drift of 4 Hz. I think anything greater than 4 Hz results in a no receive. I tried replacing the 27 MHz crystal which improved things but was not perfect. In the end I fitted the TCXO and problem solved. Now all spots show 0 Hz drift. I think you will find the same. 73 Dave M3PMG

Hi All...
?
Am enjoying my recently built QDX, awesome little rig!

Have been playing around with a few things, and in looking around,? noticed a couple of error message like things in the WSJT-X Syslog.

First thing:
Error regarding the CAT rig type.? Note, I seem to have a working CAT situation, get the green button, and seems like everything is working.? But, get the following in the log:
[RIGCTRL][2022-08-01 05:17:34.128432][00:00:00.447861][info] Hamlib version: Hamlib 4.5~git Sat Jan 01 23:05:51 2022 +0000 SHA=18548e
[RIGCTRL][2022-08-01 05:17:36.335333][00:00:02.653966][error] kenwood_open: not the right driver apparently (found 2002, asked for 2028, checked TS-440S)
[RIGCTRL][2022-08-01 05:17:36.335333][00:00:02.654054][error] kenwood_open: not the right driver apparently (found 2002, asked for 2046, checked TS-440S)
[RIGCTRL][2022-08-01 05:17:36.335333][00:00:02.654069][error] kenwood_open: not the right driver apparently (found 2002, asked for 2049, checked TS-440S)
[RIGCTRL][2022-08-01 05:17:36.335333][00:00:02.654078][error] kenwood_open: your rig (ID020) did not match but we will continue anyways

I believe I get the same or similar if I try TS-480.

Second thing:
I see large numbers of these errors talking about missed audio samples:
[SYSLOG][2022-08-01 05:17:59.407122][00:00:25.725703][warning] Detected dropped audio source samples: 720 (0.015 S)
[SYSLOG][2022-08-01 05:19:44.386176][00:02:10.702751][warning] Detected dropped audio source samples: 768 (0.016 S)
[SYSLOG][2022-08-01 05:20:14.394098][00:02:40.710611][warning] Detected dropped audio source samples: 672 (0.014 S)
[SYSLOG][2022-08-01 05:20:44.403805][00:03:10.719522][warning] Detected dropped audio source samples: 528 (0.011 S)
[SYSLOG][2022-08-01 05:21:14.390693][00:03:40.705813][warning] Detected dropped audio source samples: -576 (-0.012 S)
[SYSLOG][2022-08-01 05:21:44.407385][00:04:10.721992][warning] Detected dropped audio source samples: 528 (0.011 S)
[SYSLOG][2022-08-01 05:22:44.408575][00:05:10.722307][warning] Detected dropped audio source samples: 1248 (0.026 S)
[SYSLOG][2022-08-01 05:23:14.396078][00:05:40.709381][warning] Detected dropped audio source samples: 624 (0.013 S)
[SYSLOG][2022-08-01 05:24:14.383090][00:06:40.695824][warning] Detected dropped audio source samples: -480 (-0.01 S)
[SYSLOG][2022-08-01 05:25:14.370588][00:07:40.683182][warning] Detected dropped audio source samples: -768 (-0.016 S)
[SYSLOG][2022-08-01 05:25:44.380638][00:08:10.692559][warning] Detected dropped audio source samples: 528 (0.011 S)
[SYSLOG][2022-08-01 05:26:44.374144][00:09:10.685269][warning] Detected dropped audio source samples: -528 (-0.011 S)
[SYSLOG][2022-08-01 05:27:44.403403][00:10:10.713315][warning] Detected dropped audio source samples: -576 (-0.012 S)
[SYSLOG][2022-08-01 05:28:44.393982][00:11:10.702943][warning] Detected dropped audio source samples: -672 (-0.014 S)
[SYSLOG][2022-08-01 05:29:14.383875][00:11:40.692371][warning] Detected dropped audio source samples: -576 (-0.012 S)

I also sometimes notice a loud pop, in my pc speakers,? when the QDX goes from TX to RX.? Maybe more noticeable on 20m than the lower bands,? or gets quieter on 80m? Possibly related?

Anyway,? have this also posted on the WSJT-x group forum as well... in case it's more the SW's issue than the QDX hardware...? Interested in what both of you say.

Jon - KA6MOK

I simply found for my 80m qcx plus it had way too much audio. So I added resistance to decrease its audio gain. It works, I can still copy weak signals. If I had to have agc in a plus I would homebrew an agc with LED and photoresistor like I use in ubitx. But I find agc to be less needed for narrow bandwidth cw.?

Curt

John,?
just thought but have you tried a different PSU to your QDX. I did have a few issues at start up on my QDX when I had the current limit too low on mine.?
73 Ken g4apb?

OK John, fight down the panic, think logically, gets you there:-)

73 Alan

Hi Alan,

Our posts crossed in the ether. ?Thanks for your comments. ?Indeed, L9 had failed open - 0V on one side and 4.8V on the other. ?I removed it from the board and it was open. ?I’ll find a replacement tomorrow.?

73
John VK7JB

On 22/08/2022 05:50, John Farmer VK7JB via groups.io wrote:

The voltage on L9 is indeed zero volts with 20m selected (ON) and

Vcc(9.4V) with 20m deselected (OFF), which is also as expected.

So, I don’t understand what’s going on here. ?Why are the cathodes of D3 and D6 at 4.8V and not close to zero?

John,

Then they are not connected to L9. L9 is 0 volts both sides?
Most likely is bad soldering of L4, L10. Is there 0 volts each end of each?

There is a remote possibility of a missing PCB track.

73 Alan G4ZFQ

I might add, this issue is confined to the 20/30m LPF - the other bands and their switching are functioning normally.

An update - a bit of probing around the 20m LPF and I discovered that L9 had failed open. ?With 20m ON, the voltage on one side was zero volts and on the other side of L9, 4.8V. ?In circuit testing of L9 showed infinite resistance across it.?

I’ll replace L9 - I haven’t got a 47uH axial inductor in my junk box, so it will be a trip to the electronics store tomorrow.

Cheers,
John