Cobra 138XLR low TX power

[TM86]

Yet another Cobra 138XLR. This one has extremely low, as in less than a watt, output power. On a scope I can see that the driver seems to be doing its job. Meaning the output is bigger than the input. The final, not so much.

I've already tried swapping out the final. About to try the mv1y from a donor carcass. Any other suggestions for when that doesn't work of should I just shotgun this puppy and be done with it?

 

[Stapler]

If not done already I would definatly replace the 2.2 tant at C17 [Sometimes marked as C117] This tant is directly on the PA stage voltage and almost always go short, Also check the power regulators traces as vie seen these crack if bumped, If the bias voltage is good @ around 0.5 - 0.8v then the MV1Y wont need replacing

 

[TM86]

If not done already I would definatly replace the 2.2 tant at C17 [Sometimes marked as C117] This tant is directly on the PA stage voltage and almost always go short, Also check the power regulators traces as vie seen these crack if bumped, If the bias voltage is good @ around 0.5 - 0.8v then the MV1Y wont need replacing

Thanks! If you're referring to C179 as the 2.2 uF tantalum, it's already been replaced. The one in this radio was already cracked so it had to go before I'd apply power. Not sure if it had blown or was about to, and just didn't want to take a chance of it rapidly self-disassembling while my face was over it.

I think I was looking at this the wrong way around. I started thinking that quite possibly if the input was low enough it could like the final wasn't doing much when in fact it was working with what little I gave it. So I pulled out another 138 (I call it the growler due to a hum on LSB) and sure enough, the input levels to the driver and final were substantially higher on that radio.

So I busted out the bronze screwdriver (I can't afford a golden one) and started tweaking cans I knew are in the TX chain. L17 was the magic coil, and I'm now up to just under 2 watts out.

So, still not good, but better.

 

[Hawkeye351]

How bout sideband output?
I'm sure you've already recapped (or had it recapped) right?

 

[TM86]

How bout sideband output?
I'm sure you've already recapped (or had it recapped) right?

Sideband output is way down as well.

Radio has been recapped.

What I suspect may have happened is that something quit working the way it was supposed to and someone started tweaking to see if they could make it go again.

Only they didn't keep track of where things were when they started and just left them wherever.

 

[Ghetto]

As an idea, I would check bias of driver and final to see if they are in ballpark ma. Especially the final from what you said. I had an blown (read open carbon track carbon trimmer) in the bias circuit of Realistic Navaho TRC490 (pc385) that was acting similar and everyone helped me work it out after seeing 0ma bias on the final. the final had the correct collector voltage etc and radio had very low output but could be heard on second radio placed nearby.

 

[nomadradio]

I need to shoot a pic of a 10mm Toko IF/RF coil with the skinny slug dead even with the rim of the hole. Would save some typing, just asking "do any of these cans look like this when you peaked them?"

That dead-even position isn't a peak. It's the slug's end of travel. Turning it clockwise from the dead-even position reduces inductance. Likewise turning it counterclockwise causes the slug to rise above the hole's rim, also reducing inductance turned in that direction.

That dead-even position isn't a resonant peak, just the slug's end of travel, inductance maxed out.

Failure of the tiny tubular ceramic capacitor inside the can causes this.

And if they are all peaked with the slug visibly below the rim, this isn't causing your problem.

73

 

[Onelasttime]

While your in there replace all 3 voltage regulators they are identical. You normaly upgrade them.

I am not saying that is the issue either just saying while you are there!

 

[Hawkeye351]

VCO Buffers, carrier oscillators, driver/final bias drifting, RF stage not aligned properly. All these will affect output. Do a complete alignment, paying close attention to results.

Like nomad said, if any of those silver cans are topped out (core flush with top) or bottomed out (core flush with bottom) then check solder joints for that can or replace it. Those little silver cans (IFT, Intermediate Frequency Transformer) have a tiny tiny tiny capacitor tied to its coil on the inside, over time those caps will come loose or a sudden voltage spike can zap it loose. When adjusting those silver cans, if any of them show a sudden spike and then drop back, it's bad, replace it. These cans should show a smooth transition from top to bottom when adjusting, no sudden spikes.

Nomad can hook you up, I've followed him for a long time, he knows what he's talking about on these boards.

 

[Stapler]

Out of curiosity what driver and final is it currently using ? Typically they used a 2SC1306 driver and 2SC1969/2SC1307 final, in my experience if the 1306 is replaced with a 2SC2166 the output would be much lower

 

[TM86]

Thanks everyone for your input. I had to sleep last night, so wasted 8 hours there.

No cans have the ferrite slugs at or near the top rim. I'll have to check if one side of L37 is bottomed out, which I believe is just as bad. I also have not observed any cracked ferrite slugs, which is a relief.

Bias is adjustable, but that was a few days ago that I played with that so I don't recall if the final bias was at the proper level or not. I think it was, but I didn't record the results so it's possible I'm wishfully remembering that.

Replacing the regulators isn't off the table. I did notice last night that adjusting VR8 has no effect on AM carrier except to kill it at one end of the range. Now whether that's because it's so low already or not I don't know. Guess it's time to order some TIP31Cs.

Driver is 2SC1306, final is a 2SC1969. It was originally a 2SC1307, but the one that was in there had a busted off emitter lead. Which is probably the reason it was on fleaBay in the first place. That was found when isolating the transistors to check them out of circuit.

And while the details about the FM audio chain are good to have, I'm not sure exactly how that fits except as a reminder to look deeper because it's likely something I hadn't considered. The 138XLR doesn't have FM. I could add it, but I'll worry about that once this thing is working.

Again, thanks everyone. You've given me lots to look at, which should keep me from day drinking before church.

 

[Hawkeye351]

I was thinking of another radio I was helping someone out with on this site, got the threads mixed up. Sorry bout that, my bad.

 

[TM86]

I was thinking of another radio I was helping someone out with on this site, got the threads mixed up. Sorry bout that, my bad.

No worries, it was funny just because of the context.

 

[TM86]

Had a little time to go back into this thing this morning. Second slug in L37 was pretty much bottomed out and hard to turn. Had the bright idea to swap it with the one from the donor carcass. That got me zero output. So kind of the opposite of what I was looking for.

I did extract the hard to turn slug from the L37 I removed. It came out in pieces. The one in the donor coil was still good, so I stole that. Put the "original" L37 back in, now with two good slugs. Fired the radio up, turned the slugs and got the output up to one watt. That's AM or sideband. I realized just now that when I had said two watts earlier in the thread I was reading the wrong scale.

I am able to set transistor bias. 40 mA driver and 15 mA final.

I am not able to get VR8 to vary the output power more than a little. I did have to replace TR26, but TR25 checked OK with the cheap parts checker. But even if the AM regulator is shorted and passing the full supply voltage I should be getting more output, not less. Well, that or smoke.

Just a thought. Is it possible that someone modified this thing to only put out a watt to drive an amp?

 

[Cable Guy]

Had a little time to go back into this thing this morning. Second slug in L37 was pretty much bottomed out and hard to turn. Had the bright idea to swap it with the one from the donor carcass. That got me zero output. So kind of the opposite of what I was looking for.

I did extract the hard to turn slug from the L37 I removed. It came out in pieces. The one in the donor coil was still good, so I stole that. Put the "original" L37 back in, now with two good slugs. Fired the radio up, turned the slugs and got the output up to one watt. That's AM or sideband. I realized just now that when I had said two watts earlier in the thread I was reading the wrong scale.

I am able to set transistor bias. 40 mA driver and 15 mA final.

I am not able to get VR8 to vary the output power more than a little. I did have to replace TR26, but TR25 checked OK with the cheap parts checker. But even if the AM regulator is shorted and passing the full supply voltage I should be getting more output, not less. Well, that or smoke.

Just a thought. Is it possible that someone modified this thing to only put out a watt to drive an amp?

15ma seems way too low to me for a 1969. Just saying, I haven't worked on a 138 but that doesn't seem right. I bias a 1969 at 50-55ma.

Edit, sorry, I see it calls for 15ma. That's assuming the 1307 final. There are quite a few service bulletins, have any of them been applied?

Cobra 138XLR