开云体育

Hello group. This is my first post--wow, reading this list is akin to "drinking from a fire hose".

Two weeks ago I purchased a QMX kit to use as a WSPR beacon during the eclipse challenge event. The kit was received surprisingly fast and powered up fine on the first try, making good power on all bands. I brought it in to the local Amateur Radio manufacturer's HQ to show it off, where it was met with the expected fascination.

Unfortunately, one of the techs was playing with it using a lithium battery (12.5V, used with the KX2). While I watched, he turned it on and was investigating the controls with the battery dangling. The radio shut off. He turned it back on and wiggled the 2.1mm power connector. The radio shut off again.

This time it did not power back up.

I was standing right there watching this happen. Maybe I'm just "lucky", but it appears this problem is not as rare as claimed.

The 5.6V zener had shorted. That was replaced, but still the QMX will not boot. With the power switch held closed, the 3.3V supply is clamped at about 1V and the 5V supply is zero. One of the QFPs is drawing power and showing heat using a thermal camera. The rest of the board is cold.

Instead of repair, I purchased another QMX kit, which fortunately also arrived quickly and was built in time for the eclipse. The second build is a whole lot easier than the first! First one took me 8 hours straight, and of course I didn't properly file the "mouse bites" from the PC board panel, so the control board didn't fit properly inside the display window. The second build was accomplished in 5 hours without issue. This second rig went on to beacon WSPR for 12 continuous hours, much of that time transmitting continuously in each time window. Wonderful. The QMX definitely stuffs 4.5kg into a 2.3kg box!

This second QMX has not been out of my hands. It was wired for 9V (vs. 12V for the first one), and has only been used with a current-limited power supply of 9.5V or less. Being new, both of these were initially loaded with firmware 1.000.10. I've had it on the air enough to confirm the firmware bugs/not-yet-implemented issues of the keyer garbling after sending for a minute or two, no sidetone volume control, GPS connection halts WSPR transmission, and, strangely, the audio gain encoder on the first QMX had the non-monotonic and "big gain step" issue, but the second one does not.

Now that there is time to look at the dead one, I'm thinking that using separate, current limited supplies for 8-12V, 5V, and 3.3V is the best plan. The hot IC will be pulled and replaced (where is the best source?). Any other suggestions from the group? What am I likely to kill while debugging?

TU

73 de Bob, K6XX

I would check the main board for shorts or near shorts on 3.3v and 5v to GND. ?If that looks good, replace the SMPS boards. ?They can be repaired but it’s not worth it since Hans will sell you a new set inexpensively. ?I also have some as well if you need a set.
Jeff
W1NC

Hi Bob. I had exactly the same issue with similar results! (nearly so. see: /g/QRPLabs/message/110288).

I have confirmed on my dead QMX that VCC and VDD both now have shorts on the main board, but have not attempted to diagnose for repair. I measured on the main board only 24.3 Ohms VCC to Gnd, that's a short, and 2.57 kOhms across the VDD to Gnd rail; probably a short somewhere there too.

In my case, right after the power supply incident, my QMX would still boot up, but receive was nothing but loud static, and TX still appeared to work via terminal hardware diagnostics. But in attempting to diagnose with enclosure removed, I thought it would be smart and safe to power it up on a current limited supply at 7V (mine QMX was a 12V version), limited initially to 100mA (because if I recall it idles on receive around 60mA or so from memory ... could be wrong).

Anyway, the current limiter killed power, but then resumed power immediately afterwards for a few cycles with me horrified watching the QMX backlight blink on and off as the lab power supply did it's limiting thing before I could cut the power. Ugh. This was before I tested for shorts on the VCC and VDD rails. I think that exercise might have done additional damage.

I've not attempted a power up on that board on since then (because I now know power rail shorts are present), but I do have an infrared heat camera ... wonder if I should power it up again and look for the short culprit via a tell-tale hot spot?? But I only have 1 current limiting ps at my disposal.

Hans has diligently tried reproducing this issue on a QMX at his bench and he has a detailed write-up of his experiment which is much appreciated! But we both know first hand that a QMX can be fried due to an ugly power-input connection.

Recommendation: Power off your QMX before moving it or its power supply and cord.

My 2nd QMX is alive and working well ... I intend to keep it that way.