Sunday, October 27, 2019

So Near, Yet So Far

For a few brief, shining moments, I thought the lightning episode in the ham shack was over.  Ignoring the 60 pound doorstop of an amplifier, and I thought I could see the light at the end of that tunnel last Tuesday. 

You see, I got my main station radio (rig - in ham speak) back from the service center on Friday.  It had been gone nine weeks, and it was a relief to get it back into place as the center of the operating position.

As refresher, I sent it off because the part that I had to replace was in a package I'm just not set up to work on; among the worst of the surface mount packages, called a QFN.  It has no discernible legs, just tinned areas that are reflow soldered onto the board, with the worst being a large ground tab in the middle of the bottom.  In manufacturing, these are soldered down and removed with hot air.  I have a paint-removing heat gun, but nothing I'd feel comfortable using for this.  (I know - I need to practice on some)  Here's the datasheet drawing of the part's package upside down.

While it was at factory service, I asked the guy in charge if they had a way to do a thorough checkout of the rig before they sent it back.  He said they did and I said "do it."  

For the last several months, I don't think I'd have had a voice mode contact on the radio.  I either operate Morse code (what hams call CW) or one of the many digital communications modes that are run with a computer sound card and software that supplies audio to modulate and demodulates the receiver audio.  Setting up the radio and computer to do these things is something you do once or twice and then forget about, so it's always good for some headaches and anguish until you remember the secrets from last time.  I got everything running by early afternoon yesterday, and for few happy hours, I was just a ham again doing what I had been doing before being so rudely interrupted by the lightning.  

Late in the afternoon, I noticed my station microphone where I had put it aside and thought that I haven't tried verifying the voice circuity since the lightning strike.  A voice in my head said, "ya know, people talk through these radio things - you should try that."  I grabbed the microphone and plugged it in, then found an open area where no one was likely to be interfered with to give it a test.

Nothing.  No output signal at all.  Worse, the radio has an audio feedback to headphones that allows you to hear your audio and there was nothing there.

The obvious question was whether it's the microphone or the radio that's bad.  I didn't have a "known good" microphone handy yesterday; the other radio's microphones have a different connector.  Today, I dug out the microphone that came with the radio and tried it, assuming it's good.  Same result, so I'm 99% positive it's the radio. 

Frankly, the thought of sending it back to get them to fix something that they should have found (in my opinion) makes me sick to my stomach.  Another two months?  Another $100 in shipping and hour of labor?  If everything were to go perfectly, the parts are about $1.50 from Mouser.  They're surface mount SSOP (shrink small outline package), which would be much easier to remove and replace than that QFN36. 

Then there's the thought of whether the radio is just going to fail one part at a time, from being overstressed in the lightning's electric field.  I feel a bit overstressed just thinking about it.

Edit to add 10/19 at 0915 EDT:  This in the comments, but I called the service center and they verified they had tested the mic input and it was good.  I explained what was going on and he quickly led me to a setting screen in the radio that had been unintentionally set which disabled the microphone.  The radio is fine, and it's back to normal life.


  1. Have you done a complete reset of the radio? Yes, they should have done that before they shipped it, but....

    I've had a "Factory Rest" clear up problems that had me scratching my head. It almost sounds like the "monitor" function is on, and the gain is set too high.

    Man, if that were my radio, I'd be somewhere on the far side of "Pissed Off"!

    1. I didn't think of doing that because it looked they had done so. A lot of my settings were different.

      I called the service center tech and asked if they had tested audio through that jack and he said, "absolutely!" I explained what was going on and he pointed me to a setting that might be the issue and it was. I can use SSB now.

      I didn't even order parts I'll never use ;-)

      I'm working through some other things now that involve changing over from using the audio interface box I've been using since "The Dawn of PSK31" to using the USB port from the computer to drive everything. The audio box is super deluxe high-tech: two miniature audio transformers and an optoisolator for PTT.

    2. Cool! Good the hear it was a setting that was causing the issue.

      I built a couple of those interface boxes in the past. I used a little circuit to rectify some of the tx audio and trigger a FET for PTT. Wound up using the same thing at Sea Launch to interface the rack mounted aircraft radio to our internal system so the people in the Launch Control Center could talk directly to the helicopter crew when the chopper was flying.

  2. That is a huge bummer... I am listening to the shortwave after tuning around on 40m and hearing some decent stations. It's been really noisy around here lately and blowing out everything including commercial SW... I haven't had the time to go looking for my local noise source though.

    Good luck getting your radio sorted out.



    1. Meant to comment before that this past weekend was one of the few really big contests that gets groups to travel around the world and set up temporary stations to activate rare (or semirare) countries. That's probably why you were hearing a lot of activity in the 40m band. It's called the CQ (magazine) World Wide DX contest or just CQWW. The phone contest is the last weekend of October and the CW contest is in the last weekend of November.

      They also run another contest, called the WPX to work all the different prefixes; phone is the last weekend of March and CW is the last weekend of May - virtually always Memorial Day weekend.

  3. QFNs aren't that bad, in themselves. But if it's a lead-free assembly, the solder is likely SAC305, and... yeah, that's a "make it be somebody else's problem" situation right there.
    If all the pads are accessible (this obviously doesn't help with a QFN), I find that alloying a bunch of lead- or bismuth-alloy solder into the SAC305 joints makes rework a whole bunch easier.

    1. Well, yeah. Nothing's that bad if you have the tools setup to work on it.

      I was going to say it's one of the worst packages but then thought of at least a dozen more, like 512 pin BGA packages or micro BGAs. The place I retired from had scrapped $10,000 boards with custom Gate arrays because of one bad solder ball.

      None for me, thank you.

  4. Check out Louis Rossman on youtube to see some down and dirty board level repair. He fixes macbooks, and I've learned a lot about troubleshooting and board level repair from watching him. His goal is to teach and educate so other people can raise themselves up by starting businesses fixing things. He's also testified or commented several times to legislatures on "Right to Repair" issues.


  5. I don't get any kickbacks for this, and I KNOW it is NOT the best solution. But I've had some success with this. 3 in 1 853D SMD Soldering Iron Hot Air Heat Gun Rework Station.

    I haven't tried to do a large surface mount chip, but the small chips are manageable, even if you have to try a couple of times. (Desoldering vacuums are your friend.)

    1. That's an interesting looking station. I have a good Metcal soldering station, and it does well desoldering with wick and liquid flux, but its forte isn't soldering 36 pins at once, it's doing one joint well. When it looked like I'd have to replace a 14 pin TSSOP, I think that's right up its alley. Lifting or replacing that 36 pin QFN, not so much. That's a job for a hot air gun like the one you link.

      One thing I can be sure of, parts aren't getting bigger. Some can't get much smaller but some will. For dissipation reasons, some 0402 resistors can't go to 0201 or smaller. I can't imagine replacing a really big ball grid array or something that requires lots of area reflowed.