Some days the bear gets you. But not every day.

A year or so back I finally got this gadget developed to version 6. It replaces the 5-pin mike socket in a Cobra 2000, 142, Uniden Madison/Washington radio with a 4-pin socket. It was version 5 of this thing, but it had a fatal flaw. Had to scrap them.

First version from 2003 or so used a relay to keep the two ground circuits on the 5-pin mike jack properly isolated from each other. Hooking those two separate ground circuits together to use a 4-wire mike cord tends to cause feedback squeal problems on both transmit and receive.



Just one problem. A lot of radios would make a loud "POP!" out the speaker when you unkey the mike. Figured out it was a timing issue. Your thumb on the PTT button delays turning the speaker back on. A single relay was just too fast.



Finally made a prototype of the 2-relay solution and built a production batch of fifty.



Just one problem. The relay I used sucks.

Oops. I chose it because it's the smallest one I could find. Making two relays fit behind the mike socket calls for a small relay. But it had a fatal flaw, a seam that holds the top of the plastic case. Fujitsu used glue, and the cap would just pop right off.



Oops. Built all fifty of them before I figured out I couldn't sell them. It was just too fragile.



Some days the bear gets you.

Finally made two slightly-larger relays fit on a board that fits in a Cobra 2000, 148, Madison, etc.




So far no mention of just why. Sure, your old 4-wire D104 will work on this radio now. We're only using three wires out of the 4 pins. No need to plug in a mike to get your receiver audio now.

What really goosed this idea was the demand for roger beep/roger-K/5-tone beeps that have no relay on them. They'll work just fine in RCI-made radios that receive without the mike plugged in. Jut hooks right up.

But the Uniden 40-channel 5-pin SSB radios are different. You need a way to shut off the speaker while the beep is holding the transmitter keyed. The mike can't do this, since you just released the PTT button. Got tired of hand wiring the whole thing, and put it onto a pc board with the 4-pin socket.

But those are the only two problems that this toy will solve. Also lets you put a bump-bump into a Cobra 1000, 29 or President Zachary/Dwight D or other radio with the mike-needed-to-receive issue.

The extra holes in the board allow it to be adapted in one more way. Haven't tried this revision yet, but the 23-channel Johnson radios required a six-wire cord. The transmit/receive wires were not connected to ground, but to the hot side. The mike would switch 12 Volts to the reciever, or to the transmitter when you keyed it. That sixth wire was to shut down the receiver speaker in transmit. That's where the useless yellow wire in your six-wire Astatic and Turner mike cords would connect. By cutting a couple of traces and adding some wires, this version should let you put a 4-pin socket in a Johnson 250 Base, and the others that used the 5-pin metal lock-ring mike plug. But wait a minute, where do you connect six wires to a five-wire plug? Simple. The metal body of that plug was the connection for the sixth wire. Odd but true. Not sure how long until we'll have pics to show this hookup, if ever. Not a lot of demand for that mod.

Gotta get this batch tested. Pricing the thing will be a problem. Probably ought to just pre-wire it to a Bump-Bump and sell the package that way.

But at least this version won't puke after it's been in the radio for a week or three.

73

Thanks for the link, but that part is the same size as a 14-pin DIP IC, 8/10 of an inch long. The new board is just that wide. Two of them wouldn't leave room for anything much more. Besides, at 7 bucks apiece, the cost for two of them per board would be out of hand. It's a reed relay, and they're never cheap. I shopped for a much-better deal on the Omron relays in the pics. And it only has a SPST contact in it. Gotta have the normally-closed side for the receiver speaker. Close, but no cigar there.

The tiny Fujitsu part that caused the headaches is 4/10 inch long. Half that length.

The Omron relay in the new version is a half inch long. That extra tenth of an inch required some layout changes. And the pinout is not like the previous smaller part. Got the wiring figured out, though.

Size matters. This board leaves about a 1/16-inch gap between one side and the front-panel edge installed in a Cobra 2000. It JUST fits in some of the side-mike 5-pin mobiles. The larger it gets, the fewer radios it will fit without bumping into nearby stuff.

Got all fifty of this batch built. First few of them test just fine. Now for the joy and excitement of checking the rest of them.

73

The problem we had with the simple solution was with audio ground-loop feedback, not from RF.

The pin 2 ground leads first to the CCW end of the mike gain control, and from there to the circuit board at the audio input to the mike amplifier. The ground foil where it attaches is diagonally opposite the rear corner where the pin 4 ground wire attaches. Sometimes a ground connection "over here" is not quite the same as a ground connection "over there".

Connecting the pin 4 ground wire to the pin 2 ground wire caused two problems. When the AM carrier is turned down, audio feedback would require turning down the mike gain.

The other problem we had was with an audio-feedback squeal in SSB receive mode. If you turned down the volume control, that would stop it. Never came up with a good analysis of why that happens, but keeping the two ground circuits separated stops it.

When the 5-pin radios first appeared in 1979 customers wanted us to wire their 4-wire D104 mikes to them. Found that hooking pin 2 to pin 4 inside the mike plug for the mike's single ground caused audio-feedback issues on some radios, but not on others. It was clearly a "marginal" problem, where the unwanted feedback is just-barely enough to cause trouble on some production batches, and 'almost' enough on some others. When the customer's radio would squeal on transmit with his 4-wire D104, I would pull out a mike with the added switch contact inside it, plug into his radio and show him, "This mike doesn't make the radio squeal, your mike does. What else do you need to know?". When Turner and Astatic began selling the mikes with 6-wire cords, we assumed this was to meet the market demand for a mike that worked on your Cobra 2000,140,142, President, Teaberry and other 5-pin SSB radios that all used the new MB8719 pll. The six-wire mikes had a separate SPDT switch section not found in the older mikes. That permitted keeping the two ground circuits separate.

What originally prompted this strategy began when customer wanted the first roger-beep boards installed in 5-pin Uniden-made 40-channel radios. The older models that came with a 4-pin sockets were no big deal. The board installed just fine.

But the 5-pin radios would have the same ground-loop audio-feedback issues we saw with the 4-wire D104, since the beep board had only one ground connection. Connecting the two separate grounds together. What fixed this issue was to cut the foil trace on the beep board at the common pin of the beep's SPDT relay. Hooking the pin-4 ground wire to the now-isolated relay pin cured the feedback issues. This led to the conclusion that keeping the mike-amp ground separate from the T/R switching ground made that difference.

I didn't dream this toy up from scratch. Started telling a customer who wanted to use his 4-wire D104 on a Cobra 2000 that we could make it work by putting in a roger beep. Before long my idiot light came on. Why bother with the beep if all the radio needs is a relay?

Naturally once that's done the 5-pin socket has two pins wasted. Only need 3 of them. More and more customers would have multiple radios, and want to interchange mikes between them. Swapping the 5-pin socket with only three pins active for a 4-pin socket with only three pins in use made life easier for a lot of them.

But hey, there's always more than one way to skin a cat. Any radio that works okay with the simple solution is naturally cheaper. Never did keep a record of how many of them would. Might have been one out of three. Might have been three out of four.

Doesn't matter. Any solution that I can't rely on to work EVERY time is not something I can sell with a straight face.

And that's what this toy is about. A way to provide a customer with a sure thing. We found the simpler method to be a "maybe" thing.

Just don't like selling things that "might" work.

73