Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
#QMX Failed to turn on today after working perfectly before
#qmx
|
My 60-15m QMX has been working really well since I built it. This evening, however, when I pressed the encoder - I got nothing but a brief click in the headphones!
?
After checking the obvious - battery, cables etc, I opened up the radio to investigate. There were no signs/smells of overheating parts or magic smoke. There was no short from either VCC or VDD to ground. 12V was present on the V_IN pin of the 5V supply board and pressing the encoder dropped the PWR_ON pin to 0V, as expected. I did notice that the 3.3V regulator on the 3.3V board was getting slightly warm if I repeatedly pressed the left encoder. Figuring that its output might be being dragged down by something, I pulled the 3.3V board out of the radio and briefly applied 9V to the input pins of the 3.3V regulator chip. D103 instantly got hot so there was something pulling a lot of current on the VDD line. From the schematic, the main suspect was D109. Removing that from the board cleared the fault.?
?
As I don't have any 3.6V zeners handy, I popped the 3.3V board out of my Low-Band QMX and fitted it to the Mid-Band radio which brought it back to life.
?
Having now had a search on this forum, it appears that there are quite a few mentions of D109 failures. Would it make sense to order a slightly higher rated replacement part or would it be wiser to stick to the 500mW surface mount component? Why did it fail in the first place? Any thoughts?
?
Cheers,
Ronan
MM0IVR |
|
Same here - after one year of SOTA usage I had the same fault on the summit. When I switched the QMX a 2nd time on I got a puff and holy smoke out of the case. But I didn‘t get an smoke answer for the summit activation.?
I ordered new power boards and hope that the QMX will be alive again.? Steve --
Let your fingers talk - CWFE ?
|
|
On Wed, Sep 4, 2024 at 11:57 PM, Ronan Cantwell wrote:
Would it make sense to order a slightly higher rated replacement part or would it be wiser to stick to the 500mW surface mount component?Ronan, I remember a Post from Hans G0UPL a while ago. He told NOT to change the type. It was something about the voltage characteristic. 73 Ludwig |
|
The 5.6V Zener on the 5V SMPS board is now 5W (first PCB revision, it was 500mW. 5W is better).? The 3.3V Zener on the 3.3V SMPS board is 500mW and MUST REMAIN 500mW, do NOT change it to a higher rated part, it won't work! 73 Hans G0UPL On Thu, Sep 5, 2024 at 1:17?PM DH8WN via <DH8WN=[email protected]> wrote:
|
|
Hi Hans, “The 3.3V Zener on the 3.3V SMPS board”, did you mean “3.6V Zener”? I have placed an order of 3.6V 500mW from digikey yesterday. 73,? Wei On Thu, Sep 5, 2024 at 3:56?AM Hans Summers via <hans.summers=[email protected]> wrote:
|
|
Hi Wei yes, sorry, I meant 3.6V 500mW.? 73 Hans G0UPL On Thu, Sep 5, 2024 at 4:19?PM Wei AG6AQ via <wcheng95=[email protected]> wrote:
|
|
On Thu, Sep 5, 2024 at 11:56 AM, Hans Summers wrote:
The 3.3V Zener on the 3.3V SMPS board is 500mW and MUST REMAIN 500mW, do NOT change it to a higher rated part, it won't work!Thanks Hans. Good to get confirmation directly from the Morhership... I'll order up a few 3.6v 500mW posts and get the other rig up and running again. ?
Cheers,?
Ronan
MM0IVR? |
|
Steve,
If you measure the output pins of the 5 volt and 3.3 volt SMPS boards and see a low resistance (~2 ohms) you may have damaged the CPU. I had a similar fault and the CPU and IC402 were destroyed. I sent my QMX to Jeff Moore, who brought it back to life. You might want to make some measurements and contact Jeff and get his feedback.
?
Dennis
KK5DB |
|
Farnell delivered some bits and bobs today. In amongst the parts were some replacement 500mW? 3.6V zeners for the faulty mid-band QMX and some 10V tantalum caps to replace the lower voltage parts in the Rev.1 low band QMX.?
?
Having fitted the parts, my low band QMX is up and running. The mid-band QMX powered up with the new zener fitted and ran for a while but then popped it's clogs. There was a brief noise from the unit and then it stopped receiving and transmitting. The CPU is ok as the radio still powers on but it's completely deaf and there is zero output power when keying down.
?
I opened the unit up again and could smell that something was getting hot. There were no obvious burned components and nothing was overheating in RX. On key-down (into a dummy load) the bs170's were getting toasty while there was still no RF output. A little isopropyl dabbed on the bs170's sizzled off quickly when the key was pressed.?
?
To be honest, I'm getting a little fed up with this radio. I was dubious about it when I started building the kit as it was supplied without PSU boards and the man board itself looked a pretty grubby. PSU boards were delivered quickly when I raised a ticket but the 3.3v board failed resulting in my initial post above. I've built quite a few QRP labs kits and they've all worked first time and stayed working, apart from this one.... End Of Moan...
?
I've removed the RWTS-wound output transformer and will rewind that and replace the four bs170's so I know those parts are all good. I I'm a bit concerned that the rig is completely deaf though. It does seem like there's something a bit more serious going on... Any suggestions for where to begin fault-finding on this one?
?
Cheers,?
Ronan
MM0IVR? |
|
Ok. I've fitted new bs170's and rewound the RWTS transformer. Further testing shows no sign of any clock signals at all from the 5351 which would explain the complete lack of RX. It may also explain why two of the bs170's were getting toasty on key-down. I had initially suspected a short in the output transformer but, in hindsight, I would have been able to detect even a small amount of RF if that had been the case.
?
Could the failure of the 5351 have been caused by the initial fault with the 3.6v zener on the 3.3V board? If so, are there other parts likely to have also succumbed?
?
I'm not sure how much time I can justify putting into this but I would like to get it back on its little feet again. Any suggestions, pointers or advice would be welcomed at this point.
?
Cheers,
Ronan?
MM0IVR? |
|
Hi Ronan, The MS5351 datasheet shows an absolute maximum Vdd rating of 3.8V. I’ve measured the voltage across D109 under high-current conditions and have seen it consistently exceed 3.8V and even exceed 4V at 100 mA. The primary failure mode of D109 occurs if the PWM_3V3 input gets stuck high or if the duty cycle becomes too large. Based on the measured results across 14 boards, I suspect that the failure of this 3.6V zener correlates strongly with a failure of the MS5351. Tony AC9QY On Sun, Sep 8, 2024 at 11:28?AM Ronan Cantwell via <ronancantwell=[email protected]> wrote:
|
|
On 08/09/2024 17:43, Tony Scaminaci via groups.io wrote:
I suspect that the failure of this 3.6V zener correlates strongly with a failure of the MS5351.Difficult to tell when there are more than one fault. Seen a few times with QDXs was failure of the Si5351 when a BS170 failed, destroyed the driver and presumably fed back voltage to the Si5351.. Don't think that has been reported with a QDM but it must be possible. 73 Alan G4ZFQ |
|
Yes, very true Alan. In this particular case, the 3.6V zener had already failed. Being a 1/2W device, it would take current exceeding 100 mA flowing through the zener to cause it to overheat. Every SMPS board I measured displayed an avalanche voltage of 4.0-4.1V at 100 ma which is still under the half watt rating. So, my conclusion is that the zener was dissipating at least 1/2W or greater when it failed which put at least 4V on the MS5351, exceeding its max rating of 3.8V. Further, if I let the avalanche current sit at 100 ma for more than a few seconds, the zener voltage slowly crept up even higher. One board exhibited 4.3V before I shut it down while the remaining boards hit about 4.2V. On Sun, Sep 8, 2024 at 2:16?PM Alan G4ZFQ via <alan4alan=[email protected]> wrote: On 08/09/2024 17:43, Tony Scaminaci via wrote: |
|
Quick update on this repair:?
I fitted a replacement 5351 and replaced IC501 and IC502. The radio now receives well so I'm happy that the replacement 5351 is functioning. It's only putting out about half a watt so I'll wait till my replacement for IC503 arrives before testing further. I've not checked with the scope to see if IC503 is actually faulty but it's the next suspect in the chain so I'll swap it out anyway as I'm quite enjoying the challenge of the SMD soldering. To be honest, I really wasn't expecting the 5351 to go in so easily ?... Onwards (slowly) and upwards...
?
T.B.C.
?
Ronan
MM0IVR? |
|
Hi. I would like to know if someone could make a mesure over the main board between pin #5 on JP206 (microcontroler Board) and ground. Withuot the 2 SMPS connected.
My QMX suddenly power off, I found D109 shorted. Besides the resistence between VCC and ground over the main board is very low. I think that is not normal.
Thanks in advance and HNY!
Osvaldo - LU3MAM |
Steve
Dan
to navigate to use esc to dismiss