Topics

#qcx #bpf #lpf 30M QCX Problem Solving #qcx #lpf #bpf

 

I just returned to it's owner a 30M QCX I had on the bench. The symptoms were:
1. C1 tuning was erratic and intermittent as indicated in alignment item 8.7. Scaling factor number jumped wildly when tuning and even when lightly tapping C1.
2. The I&Q alignments 8.8, 8.9 & 8.10 would not show a minimum. All variable resistances were at end of range.
3. Very low TX output. With PS at 13.4 VDC only 0.7W with 280mA including LCD. (in Rx mode without LCD current was 97 mA.) With PS up to 14.39VDC it was putting out 1.03W into the dummy load while drawing 300 mA.

The cause/solutions were:
1. C1 stator pins were not making good contact. I found the layered pins had not soldered properly, with the middle layer not well connected. I removed the var cap, cleaned up the pins and reinstalled it. >> C1 tuning was smooth and solid. Alignment 8.7 adjusted smoothly, but did max out with a 6 on the scaling range.
2. Following the excellent guide in the Troubleshooting document, I found the o-scope showed the test signal with distortion at the output of IC6B and very low signal level at the output of IC6A. The problem was that R22 was a 33k resistor instead of 3.3k. In searching for where the 3.3k parts had found a home, I found both R28 & R29 were 3.3k instead of 33k. With these parts replaced with the correct values, the Alignment of 8.8, 8.9 & 8.10 went smoothly.
3. I removed three turns from each toroid in the LPF (l1, L2, L3) and tested the output. With 14.39V the output was up to 1.5W. I removed two more turns from each inductor and output was up to just over 3W while drawing around 500 mA. The decision was made to not make any further changes as these inductors had been removed before, and the PCB was showing stress.

I reached a couple conclusions regarding the areas of the QCX where I performed this work.When making the initial installation of C1, carefully turn the movable plates to be fully meshed with the stator before starting the installation. When soldering the stator leg try to apply the iron heat to the edge of the pin so as to get heat to all three layers of the pin. If removing or even re-heating the connections, make sure plates are fully meshed as well. This component is very fragile, and the physical construction makes the stator connection prone to damage or at least distortion. When fully meshed, it is more sturdy.

In regard to the LPF inductor behavior, it occurred to me that when the TX is activated into an LPF that is tuned below the operating frequency, the situation is not unlike transmitting into a mis-tuned antenna, or an open feedline with abnormally hig RF voltage present in the FET circuit. This may account for some of the PA failures that have been reported. My suggestion would be to initially set the PS low, perhaps around 6VDC to check and verify the LPF performance. If it looks good compared to the output vs Supply V values shown in the tables in Section 7.3 then carefully increase the supply V and check again. This may help some avoid making PA parts replacements.

Let me say again, this is only a description of what I found in one case, and my own conclusions resulting from these findings.
Happy QCX-ing!!
--
> I finally got it all together...now I can't remember where I put it<

Wes

AE6ZM

Hereford, AZ

 

Liam Kingsmill <lumetters1@...>
 

Interesting post, Wes.  Since I've got a handful of C1 coming, your suggestions for installing will be quite appropriate.  While the C1 repaired with hot glue works here, it's gotta go in favor of unrepaired.

Liam


On 02/24/2018 04:50 PM, Wes wrote:
I just returned to it's owner a 30M QCX I had on the bench. The symptoms were:
1. C1 tuning was erratic and intermittent as indicated in alignment item 8.7. Scaling factor number jumped wildly when tuning and even when lightly tapping C1.
2. The I&Q alignments 8.8, 8.9 & 8.10 would not show a minimum. All variable resistances were at end of range.
3. Very low TX output. With PS at 13.4 VDC only 0.7W with 280mA including LCD. (in Rx mode without LCD current was 97 mA.) With PS up to 14.39VDC it was putting out 1.03W into the dummy load while drawing 300 mA.

The cause/solutions were:
1. C1 stator pins were not making good contact. I found the layered pins had not soldered properly, with the middle layer not well connected. I removed the var cap, cleaned up the pins and reinstalled it. >> C1 tuning was smooth and solid. Alignment 8.7 adjusted smoothly, but did max out with a 6 on the scaling range.
2. Following the excellent guide in the Troubleshooting document, I found the o-scope showed the test signal with distortion at the output of IC6B and very low signal level at the output of IC6A. The problem was that R22 was a 33k resistor instead of 3.3k. In searching for where the 3.3k parts had found a home, I found both R28 & R29 were 3.3k instead of 33k. With these parts replaced with the correct values, the Alignment of 8.8, 8.9 & 8.10 went smoothly.
3. I removed three turns from each toroid in the LPF (l1, L2, L3) and tested the output. With 14.39V the output was up to 1.5W. I removed two more turns from each inductor and output was up to just over 3W while drawing around 500 mA. The decision was made to not make any further changes as these inductors had been removed before, and the PCB was showing stress.

I reached a couple conclusions regarding the areas of the QCX where I performed this work.When making the initial installation of C1, carefully turn the movable plates to be fully meshed with the stator before starting the installation. When soldering the stator leg try to apply the iron heat to the edge of the pin so as to get heat to all three layers of the pin. If removing or even re-heating the connections, make sure plates are fully meshed as well. This component is very fragile, and the physical construction makes the stator connection prone to damage or at least distortion. When fully meshed, it is more sturdy.

In regard to the LPF inductor behavior, it occurred to me that when the TX is activated into an LPF that is tuned below the operating frequency, the situation is not unlike transmitting into a mis-tuned antenna, or an open feedline with abnormally hig RF voltage present in the FET circuit. This may account for some of the PA failures that have been reported. My suggestion would be to initially set the PS low, perhaps around 6VDC to check and verify the LPF performance. If it looks good compared to the output vs Supply V values shown in the tables in Section 7.3 then carefully increase the supply V and check again. This may help some avoid making PA parts replacements.

Let me say again, this is only a description of what I found in one case, and my own conclusions resulting from these findings.
Happy QCX-ing!!
--
> I finally got it all together...now I can't remember where I put it<

Wes

AE6ZM

Hereford, AZ