R-392 Power Switch Problem/Repair - How To

[WBear2GCR]

Having recently acquired more than one R-392, I found that the 3 I ended up with - yes three(!!) - all seemed to either not turn on, or exhibited funny behavior when trying to turn on. The power switch has three main settings: Off, Standby, Normal.

On one of these, it seems like it wanted to turn on, but the filaments (front light bulbs run off the fils) flickered but did not stay on. ONE of them "worked" and I've been playing it for a while. It needs some tune up and attention to a few band ranges, but that's another story.

One of the symptoms was that the filaments stayed ON even though the switch was set to Off! Ok, so what, right? Maybe it was a relay with welded contacts?

Finally on Friday, I turned it off, but when I turned it back on it failed to light up. Decided to open the case (no big deal).

Here's the story, plain and simple.
The power switch is a standard rotary switch with a cam on the shaft. The cam activates a large "microswitch". The microswitch is a normally CLOSED switch, so in the OFF position it is pushed IN. This microswitch connects the filaments.

I opened the switch. To open the switch you need to drive out the very small pin that is smaller than the diameter of a paper clip wire from the side of the switch. Very hard to see. I didn't see it. I cracked the case, which is Bakelite trying to pry off the top. Geenie-us!

The problem is that the microswitch sits in the activated position most of the time - OFF. The radio is OFF mostly. So this one was built in 1955, and after 60 years the phosphor bronze snap action mechanism take a "set" and no longer "snap actions" at all!

I was able to repair mine.

I bent the moving contact arm so that it was parallel again, and would "snap" with the very small space it has to move. I also cleaned the contact under a loupe, and switched the NO contact for the NC contact (they unscrew) since the NO contact is not used (except for a place for the arm to land). Then I carefully reassembled the unit, and epoxied the parts back together, filled the one hole with epoxy.

It works. How long? Dunno.
My advice is IF you have an R-392, replace the switch, even if you need to fabricate another mounting bracket for a different switch.

To GET to the switch, remove the front panel side screws, and the nut on the switch shaft, slide the front panel forward and inch or so, and slide the switch assembly back. You have to remove the ANT COUPLING flex shaft.

Fwiw, mini TORX wrenches are a fairly good fit for BRISTOL spline set screws!! I'd not use the Torx on screws that need significant torque or that are corroded, but for these, it was perfect! (Got my set from Harbor Freight, cheap!)

Today I was at the main QTH again, and my switch drawer revealed several old stock microswitches that are perfect fits. One is virtually identical. I'll be into the other two units to replace the microswitches and get them to fire up now.

If I get a chance, I'll post pix of the microswitches, and when I open the other two, I'll shoot some pix of the switch assembly.

As far as I can see the main difference found in the switches that I have is minor variations in body style, actuator lever (if any) and actuation pressure. But this is a very non critical application, and pressure means nothing, physical fit is all that is needed, and the ability to handle ~3-5Amps @ 28vdc.

All in all, not too bad a process for this fix. About 2hrs elapsed, I'd estimate.

                         _-_-

[WBear2GCR]

Forgot to drop in a picture of the R-392s for reference...

And, I just realized that the shot I took some time back of the gear train shows the power switch + the microswitch, it's to the lower right in the photo, you see the microswitch end on, and you can see the cam and the lever!
Bingo!

[N8ETQ]

  Hey Bear,

      Seems all the R-39X rigs use em' and they all
tend to weld up.. I have an R-390 here I've just been
unplugging for years, and 2 R-390-A's. One won't
turn ON, the other won't turn off!

      Maybe this is the year I'll get a chance to
work on my junk..

73

/Dan

[WBear2GCR]

Right, it's not "weld up" - it's tired and bent spring metal material internal to the microswitch. The contacts surely can become arc'd into oblivion, but it seems that mostly the internals don't do the "snap switch" mechanical hysteresis any longer.

Looks like they've exceeded the expected life at 60+ years of service!

Anyhow, the other two are now slated for replacement microswitches.

                  _-_-

[WBear2GCR]

Another R-392 made it to the bench today. At first it lit up fine, but had no audio. Odd. This one was modded by a former user with the ANT input to the back via coax & SO-239, plus power run direct to the fuse holders and then to a screw type barrier strip on the back. Guess he did not have the plugs.

Anyhow, this prompted me to go through the exercise of pulling the bottom modules, and swapping them with the #3 unit, which had audio, but no RF (apparently). That one seems like its either no PTO output or no xtal osc or a bad mixer tube, etc... we'll see later.

Turned out that the previous owner had removed the jumper from the power socket/jack that puts the audio on the two audio socket/jacks (front panel)! Odd that they put audio on the power plug AND two front panel multi pin jacks! They must have had their reasons.

Soldered up the jumper and it makes sound. Yay. A bit deaf though, much less sensitivity than the #1 unit, which is somewhat less than stellar in that department too, and seems rather deaf +/- the 14mc bands... So, I'll swap the modules back. Btw they use different shaft couplers for the bandwidth switch!Thanks guys! Thanks so much for the extra hassle of swapping them.

This #2 unit also has the ratty microswitch issue, so that has to be replaced too.

Looks like I'll need to run the tubes through a tube tester, and go through the alignment to see where the heck the sensitivity went. Scope & sig generator time. Humph.

                        _-_-

[WB6NVH]

The audio is also brought out to the power connector so that the receiver can be used with the matching T-195 transmitter as the AN/GRC-19.

Has anyone determined a part number from someplace like Mouser for an off the shelf new switch?  I think someone still makes it as I see new ones on eBay from time to time. 

[WBear2GCR]

I did not. Since I had a number of exact replacements in the drawer marked "switches" I didn't need to look.

(Yes, yet another case of " ...what do you want that stuff for? why do you keep all that *carp* around, you don't need it or ever use any... why don't you sell it or throw it out?" - those microswitches came in a batch of assorted switches at least a decade or more back.)

But microswitches are still made and unless ur a purist, you could substitute in almost any suitably rated microswitch into the space available. There's a fair amount of space there.

Interestingly, one of the units had a standard rotary switch (with the cam arm, of course) and an aluminum bracket to hold both the switch and the microswitch. The other two had a special switch whose front metal bracket (the part that holds the panel bushing and the detent mechanism) was extended out to hold the mounting screws for the microswitch. Surprised me to see that.

                         _-_-