|
My first post to the group never arrived, so I'll try again today.
I bought and assembled a QMX a couple of weeks ago to beacon in the Eclipse challenge. It arrived in just a few days and worked fine at first power-up. I brought it in to the HQ of the "local amateur radio manufacturer" to show it off and they were suitably impressed, as expected.
One tech there connected the QMX to a KX2 battery (12.5V) and was investigating the menus, hand-holding the radio with the battery dangling. It shut off by itself. He turned it back on and wiggled the 2.1mm connector. It shut off again.
Now it will not turn on.
I was standing right there watching, so I know this is how it happened. Either I am unusually "lucky" or this problem is not as rare as has been postulated.
The 5.6V zener was blown and was replaced. Now the 3.3V supply is clamped to about 1V and the 5V supply does not start. Thermal camera images show one of the QFPs getting warm while the rest of the board(s) are cold. The boards only draw current while the power switch is held closed, and the display is dark. Radio is completely unresponsive.
I like the idea of trying separate CURRENT LIMITED 3.3 and 5V power supplies to test the QMX after replacing the bad QFP. Anything else I should watch out for?
Meanwhile, I purchased another QMX board set. It also arrived quickly and was immediately built--the second one is so much easier than the first. It also worked the first time and survived 12 hours of WSPR beaconing, much of which was transmitting at nearly 100% duty cycle (in every WSPR cycle), during the eclipse. I wound this one for 9V (the first was wired for 12V) and has only been used with a current-limited supply. It has QSOd on CW and FT8 besides its time on WSPR.
FWIW, I'm seeing what appears to be a firmware bug not yet mentioned on this list. WSPR transmissions are squelched if a GPS is connected--whether or not CW Protection is enabled. This is with 1.000.10 firmware.
Anyway, any advice for troubleshooting my dead boards is welcome.
--
73 de Bob, K6XX
--
73 de Bob, K6XX
|
When Zener diode gets blown, it's a good idea to stop and think what caused it, and if there's anything else that might have died. Zener diodes, like tantalum caps, rarely die but when they do, they very often die in short mode. Testing the conduction state of the dead Zener can help you narrow down the areas when you think what else might have been affected.
|
Well documented. ?
The power supply safe-starts at start-up but loses the plot if there are large steps in input voltage (Hans has verified this can happen). ? The diode died holding the ?regulator output down (to some degree) but other components in the PSU or rig may be damaged. ? I think the other thread identified a few things to replace ?but if available I¡¯d get a new PSU board instead and check it out from there?
|
Looks like your original post came through after this, your re-post. I replied to you there. See /g/QRPLabs/message/112504
I'd like to add ... that despite my experience with my first QMX, I love this thing so much, that I too built a 2nd one. I'll take better care it and won't let it out of my hands  .
73!
Roy - KI0ER
|
Ryuji,
Yes, I did. It was shorted. Typical zener death.
Checked the larger more obvious tantalums. They seem OK.
--
73 de Bob, K6XX
--
73 de Bob, K6XX
|
The problem is that, unless the board was designed to allow you to isolate the Zener diode, all you may know is that SOMETHING somewhere has shorted to ground.? It might be the Zener diode, or it might be something else.
Many, many years ago, Will WD5GJA/SK and I were working on a Radio Shack SSB CB transceiver.? It had no output, as I recall.? Will took a voltage measurement, saw zero volts, and said "Uh-oh.? Something, somewhere in there has shorted to ground, and it is going to be a real pain in the ass to find it."? A moment later, he said "I wonder..." and, I don't remember how, tested the Zener diode that supplied that particular internal power supply line.? We got lucky.? The Zener had failed shorted.
I remember talking with a guy at the Dallas Sidewalk Sale, many years ago.? He had a well-known rig, a good one (ham transceiver, I think), with no activity in one stage.? We were looking at the schematic.? The power for that one stage was provided through a dedicated Zener diode, that was actually inside the shield for it, and only supplied it.? I told him the story, and said something like "I'd probably check that Zener diode first, just because I've seen this kind of failure before, and THIS rig makes it easy to isolate the Zener."
--John WB5YOO
On Wed, 18 Oct 2023 10:48:22 -0700, "Ryuji Suzuki AB1WX" <ab1wx@...> wrote:
?
When Zener diode gets blown, it's a good idea to stop and think what caused it, and if there's anything else that might have died. Zener diodes, like tantalum caps, rarely die but when they do, they very often die in short mode. Testing the conduction state of the dead Zener can help you narrow down the areas when you think what else might have been affected.
_._,_._,_
-- --John R. Strohm WB5YOO
|
Unplug PCB#1 and check it on the bench.? Should be easy to find the short, but possibly not determine if the failure is D108 (most likely) or C107
|
Regarding what caused the Zener to blow, this sounds like the voltage spike issue that Hans is already aware of. Jiggling the power connector likely caused a fast-rise voltage spike that is known to blow that Zener. The question then becomes what else blew as a result?
On Wed, Oct 18, 2023 at 12:48?PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: When Zener diode gets blown, it's a good idea to stop and think what caused it, and if there's anything else that might have died. Zener diodes, like tantalum caps, rarely die but when they do, they very often die in short mode. Testing the conduction state of the dead Zener can help you narrow down the areas when you think what else might have been affected.
|
On Tue, Oct 24, 2023 at 03:24 PM, Tony Scaminaci wrote:
Jiggling the power connector likely
caused a fast-rise voltage spike that is known to blow that Zener. The
question then becomes what else blew as a result?
That is a small bit alarming. Is it useful to strap a big electrolytic cap in QMX?
At a minimum I'll double check the correct fit of all my DC barrel connectors that go into QMX and DeoxIt them all.
(And that's why I like XT60/XT30 for DC.)
|
Hans investigated this failure mode a while back and confirmed he could make it happen. However, I believe he concluded that this was a rare condition which didn¡¯t require a fix.
On Tue, Oct 24, 2023 at 2:30?PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: On Tue, Oct 24, 2023 at 03:24 PM, Tony Scaminaci wrote:
Jiggling the power connector likely
caused a fast-rise voltage spike that is known to blow that Zener. The
question then becomes what else blew as a result?
That is a small bit alarming. Is it useful to strap a big electrolytic cap in QMX?
At a minimum I'll double check the correct fit of all my DC barrel connectors that go into QMX and DeoxIt them all.
(And that's why I like XT60/XT30 for DC.)
|
This was my QMX. So far the dead include _both_ zeners (5.6 and 3.6V) and one of the 3253s.
Debugging/repair continues. Waiting for a replacement 3253. Will advise once there is progress.
I remain convinced that this is not such a rare occurrence. Testing a replacement QMX fed from a LM317 linear regulator set to 9.0V (and, importantly, with its inherent current limiting) has resulted in a reliable radio with over a dozen hours on the air so far.
--
73 de Bob, K6XX
--
73 de Bob, K6XX
|
Hi Tony, all
I confirmed I could make it happen with a lot of effort. When it did happen, all I had to do was replace the 5.6V zener to return the QMX to full working order. I do think it would be a rare condition. I would like to prevent even a rare failure mode but this one is still on my list, as it's a far from trivial thing to tackle.?
On Tue, Oct 24, 2023 at 10:42?PM Tony Scaminaci < tonyscam@...> wrote: Hans investigated this failure mode a while back and confirmed he could make it happen. However, I believe he concluded that this was a rare condition which didn¡¯t require a fix. On Tue, Oct 24, 2023 at 2:30?PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: On Tue, Oct 24, 2023 at 03:24 PM, Tony Scaminaci wrote:
Jiggling the power connector likely
caused a fast-rise voltage spike that is known to blow that Zener. The
question then becomes what else blew as a result?
That is a small bit alarming. Is it useful to strap a big electrolytic cap in QMX?
At a minimum I'll double check the correct fit of all my DC barrel connectors that go into QMX and DeoxIt them all.
(And that's why I like XT60/XT30 for DC.)
|
Zener diode is a bit unique among semiconductor devices in that it takes some amount of repeated beating as if nothing happened and then it all of sudden dies in the short mode. That often makes troubleshooting tricky and may leave the underlying issue under-appreciated.
I did read the report on QRP Labs website but I thought battery-only operation would make it irrelevant. But then the jiggling of the DC connector is a bit different... those barrel connectors have zillions of minor variations and not all are well made. But at least I can check, clean and use them with caution for now.
|
This problem just occurred with my QMX. I had removed the power connector from the radio and reconnected and the radio was dead. Upon troubleshooting, I discovered the shorted 5.6 zener diode. I could not find the reason for the diode shorting. I'm glad I searched this site and found out it not an isolated case. I'll order a diode and replace it and hope that solves my problem. I love the radio. It's a marvelous piece of engineering. Thank you Hans .
|
I'm now thinking I'll replace the DC barrel socket on my QMX with short wires soldered to the board and power-pole connectors on the other end. XT-60 connectors would be a good choice as well. It'll be less pretty with a wire hanging out of the case, but that should dramatically reduce risk of failure due to a loose connector.
My first QMX died due to this issue, and while testing my 2nd QMX for field operation, I noticed that the DC barrel plug had come loose while the QMX was on my desktop and I was fumbling with positioning the antenna cable. The QMX is small and light-weight so it moves around easily if any of the cords connected to it get moved ... and that seems to happen more often than anticipated, even though I learned from experience to be careful with the power cord.
I caught the issue this time before any damage occurred. But I think the wiser choice now is to eliminate the barrel connector because it's just too easy for it to dislodge.
|
With the original QCX I replaced the power barrel connector with a short PowerPole cable. It has proved solid and reliable. I find the barrel connectors flakey and unreliable.
toggle quoted message
Show quoted text
On Oct 25, 2023, at 11:59, Roy - KI0ER via groups.io <ki0er@...> wrote:
? I'm now thinking I'll replace the DC barrel socket on my QMX with short wires soldered to the board and power-pole connectors on the other end. XT-60 connectors would be a good choice as well. It'll be less pretty with a wire hanging out of the case, but that should dramatically reduce risk of failure due to a loose connector.
My first QMX died due to this issue, and while testing my 2nd QMX for field operation, I noticed that the DC barrel plug had come loose while the QMX was on my desktop and I was fumbling with positioning the antenna cable. The QMX is small and light-weight so it moves around easily if any of the cords connected to it get moved ... and that seems to happen more often than anticipated, even though I learned from experience to be careful with the power cord.
I caught the issue this time before any damage occurred. But I think the wiser choice now is to eliminate the barrel connector because it's just too easy for it to dislodge.
|
Here's what I did:
QRP Labs uses 5.5/2.1mm coaxial power connectors for everything, including for different voltages. The vast majority of amateur equipment uses nominal 12 volts (13.8 V) for power. In North America, the small Anderson PowerPole connectors have become the 12 volt standard. I don¡¯t know if there are any 12 volt connector standards for amateur radio in Europe or other continents, but here it is mainly due to ARES and RACES, among others, adopting it as their standard. This has made emergencies, public service events, field day, during contests, and borrowing equipment so much more convenient and hassle free for everyone.
?
Most of us have switched to PowerPole connectors in recent years. So I¡¯ve been thinking about how to convert all my QRP Labs rigs to PowerPoles and finally came up with a simple, easy to implement solution. There are three different implementations: for the original QCX with BamaTech case, the QCX-mini, and the QDX. Eliminates an extra adapter cable for me.
?
The original QCX used screw terminals for the power connection. The BamaTech case had a very flaky coaxial jack. I modified it to use PowerPoles. I notched out the original coaxial jack hole to a rectangle for a PowerPole jack. A #2 bolt holds the jack fixed to the bottom of the case. This makes field connections much easier and more reliable without an extra power cable interface.
?
The QCX-mini presented different issues. Its coaxial jack is much better and is not prone to failure, so that is a big improvement. I still wanted to eliminate that extra power adapter cable, but the mini is much more compact. There is no extra room inside the case to put the larger PowerPole jack. The original coaxial jack is left unchanged and a 6" power cord is soldered to the bottom of the QCX-mini¡¯s PCB. The other end sports a PowerPole connector and electrical tape was applied to the case bottom to avoid a short circuit from the new connection. As there was no opening in the case for the new cord, a nibbler was used to put a small notch in the end plate of the case to allow for the new cord to exit. The new power cord easily folds up over the QCX-mini and so presents no extra volume for storage.
?
For the QDX, a similar method could be used if there are no modifications to the original QDX. However, I have already modified my QDX to provide an internal 9 volt power supply from a 12 volt input. This meant that the original 12 volts could not be sourced from the bottom of the PCB, but only from the back of the original coaxial connector. Because of this, both connections were approached from above the PCB. The end plate of the case was notched similar to the QCX-mini, but on the top, rather than the bottom, of the end plate.
?
?
Dan
NM3A
?
SPCQ (Society for the Prevention of Cruelty to QRP Labs) statement:
? ? ? No QRP Labs equipment was harmed in this modification.
-- 73, Dan? NM3A There are moments when everything turns out right .... Don't let them alarm you; they pass.?
-Jules Renard
|
_FWIW I have found that there is a wide variety of dimensions for barrel connectors that will fit the QRP labs kits. Many seem to have varying inside diameters of the center connection regardless of the advertised specifications. The pigtailed power?cord available to order along with the QMX has a yellow plastic tip, and fits the power socket perfectly. I just terminated the loose end of the pigtail to power poles. Regardless of how I move the QMX around while powered it remains solid. I would be interested to know if any of the reported intermittent connections occurred?while using the optional QRPlabs power plug Dave VE7AHT
|
Hi All, Something to be on the lookout?for.. When installing the bottom cover check and see if the tabs on the power socket are touching it.? In my case it was. yes I know the cover?is anodized but in time and pressure could cause a short circuit. 73 Ray VE7AVG
_FWIW I have found that there is a wide variety of dimensions for barrel connectors that will fit the QRP labs kits. Many seem to have varying inside diameters of the center connection regardless of the advertised specifications. The pigtailed power?cord available to order along with the QMX has a yellow plastic tip, and fits the power socket perfectly. I just terminated the loose end of the pigtail to power poles. Regardless of how I move the QMX around while powered it remains solid. I would be interested to know if any of the reported intermittent connections occurred?while using the optional QRPlabs power plug Dave VE7AHT
|
Ray, et al, I put a thin nylon washer under the screw closest to the power jack. The screw head now rests on the bottom cover and there is no way things can short out there now.? -- 73, Dan? NM3A There are moments when everything turns out right .... Don't let them alarm you; they pass.?
-Jules Renard
|
Bob Wolbert, K6XX
Ryuji Suzuki AB1WX
Bruce Akhurst