Cobra 29gtl/ltd upd 2816 channel mod ?

[Cable Guy]

Actually, I can't even see how this worked on the 25 gtl at all (tc9106), much less the 29 (upd2816). Let's break it down.

Standard, unmodified 25 gtl scheme, channel 1 26.965...
When rx, the vco is putting out 16.270, which gets subtracted from incoming 26.965 leaving 10.695 the first if, which the reference 10.240 is subtracted from to get .455 the second if, then is demodulated.
In tx, the pin is flipped causing an upward shift of the vco by .455 resulting in a vco of 16.725. That gets added to the reference of 10.240 for 26.965 and out it goes into the world.

Now, inverting the t/r pin...
When rx, vco is putting out 16.725, which subtracts from incoming 26.965 leaves 10.240, which gets the reference of 10.240 subtracted leaving 0... uh oh, so when inverting that pin, we're not listening to that frequency, which is expected. But what frequency are we hearing? 16.725 + 10.695 = 27.420, not the 26.510 I'm looking for. Seems we need the vco to put out 15.815 to rx on the same channel we TX on, maybe.
In tx, vco is putting out 16.270 and adds to reference of 10.240 giving 26.510, the desired frequency.

Inverting that pin seems to put us 45 and a half channels below 1 on tx, and 45 and a half above channel 1 in rx.

This inverting of that pin worked on the 25, I promise, and I don't have it here anymore to see why, but using the CB PLL DATA BOOK as reference, I don't know why it did work.

Now we didn't tweak the reference to get back on channel while the t/r pin is inverted. That won't be a problem, just trying to figure out why this inversion worked on the 25gtl at all. What am I missing?

 

[brandon7861]

What am I missing?

27.42MHz-16.725MHz does indeed create 10.695MHz, which can then mix with 10.24 to generate 455kc. But we don't want 27.42 on receive. We want 26.51MHz

If you nix the first IF filter and depend solely on the second IF filter (lol), you can do this:
26.51-16.725=9.785 (instead of 10.695) and 10.24-9.785=455kHz

Silly to do IMO, but I was just trying to make the math work lol.

 

[Cable Guy]

27.42MHz-16.725MHz does indeed create 10.695MHz, which can then mix with 10.24 to generate 455kc. But we don't want 27.42 on receive. We want 26.51MHz

If you nix the first IF filter and depend solely on the second IF filter (lol), you can do this:
26.51-16.725=9.785 (instead of 10.695) and 10.24-9.785=455kHz

Yes, I figured that much out, just not sure why it worked without nixing the first if filter as that was not evident with the 25. I'm going to get that radio back and on my bench and see what it's doing. I'm telling you, it rx and TX on same low frequency when I fixed it, and no xtals were swapped for a different value, although the ref xtal was bad. I took some pics and scribbled down what I thought was the mod but I must have missed something. Or I'm going nuts.

 

[loosecannon]

So i was going to post a link to a kit on ebay, and then i realized they are YOUR KITS for sale! LOL
LC

 

[Cable Guy]

Okay, got the 25 back and looking it over. Looks like I expect. Swept it from 26.510 to 27.420 and it does receive on both those frequencies, albeit stronger on the 26mhz signal.

I imagine the lower channel receiver magic comes from tuning the receiver to the opposite IF image but I could be wrong. But if so, does that come at a cost of receiver performance when switched back to the normal cb band?

Also, I verified that the vco is running as expected with t/r pin inverted so there is no magic there. Both IF filters are in place, although the first, FL1 is a ceramic filter instead of a crystal filter.

 

[Cable Guy]

I got it, there is a 220pf cap strapped across the first filter FL1 and another strapped across L4, I guess to widen them up so the other image can be tuned and not rejected. Forgive me if I don't have all the facts, I'm learning. But, with a test 29, 2816, strapping those places, I can now rx where I tx. But I feel doing this will dump selectivity. Pushing onward.

 

[brandon7861]

Nevermind, I was looking at it wrong.

 

[Cable Guy]

Yeah, this is a crap mod, you have to compromise the 1st IF filter making it wide enough and tune to the other side, the "negative" 455khz image if you will. You will receive signal from both frequencies and while you can tune either, there will always be interference from the other half. That explains why when tuned for the lowers, the normal range was weaker and vice versa. I would say a xtal kit would be more appropriate without sacrifice. But, at least we know now. If I got anything wrong, sorry. I could make a guide, but I don't think it's worth the effort.

 

[brandon7861]

I was drawing out a better synthesizer block diagram trying to figure out another way, but I didn't see one. Forgive my post 22, I was wrong and edited it. The pins of IC2 are correctly labeled. The block diagram had me thinking that 16 was the mixer output lol. How dumb is that? lol

I'm coming to the same conclusion. The only (good) way I can figure is a PLL swap or taking over the 15.36MHz signal with an external oscillator.

 

[dss56]

Yeah, this is a crap mod, you have to compromise the 1st IF filter making it wide enough and tune to the other side, the "negative" 455khz image if you will. You will receive signal from both frequencies and while you can tune either, there will always be interference from the other half. That explains why when tuned for the lowers, the normal range was weaker and vice versa. I would say a xtal kit would be more appropriate without sacrifice. But, at least we know now. If I got anything wrong, sorry. I could make a guide, but I don't think it's worth the effort.

I would say dont bother and like u said just use an Expo S kit for the 25s channels or an Expo A kit for the 29s or an N kit for the Cobra 29s but the N kit only gives you the high channels ch 41 to 28.045. Im sure some where this mod is documented and was used way back.
Thanks for trying to find out how it was done.