Cobra 148GTL-DX/SS360-FM SWR meter/Calibrate issue

[Christophe Barbe]

I still did not replace any cap in the driver and final area ; only from the DC in area (1000 uF caps) down to the mike socket area and not all of them (14 pcs if I remember well). Still, I may need to revisit the solder and reflow them and as I still have the OEM grey ones, I can put some of them back to the PCB (the 010AB) ; I did not replace C44, so no worries on this one.
I do have MICA wafers and thermal compound (OFF topic : as I am willing to insulate an MRF477 on my Lincoln Mk1 which makes some weird SWR reading on TX (antenna too close) ; I disconnected the 2 ceramic caps on the board side, which works great but losing 3-5W off of 22-23W (not that it can be noticed on the other side) but just to experiment). So I may replace these too on the 148GTL-DX in case they got worn.
I had a tough day at the workplace with an extended meeting late afternoon, so i am exhausted to say the least, I postponed to tomorrow afternoon what I wanted to do on the PC010AB today.
Good info, all this which not only will serve me in my research but next ones looking for a solution.
Actually i had this thinking that : indeed old caps aged together, all having some loss in capacitance, getting "used and matched to each-other", maybe, in my "frenglish" way to explain myself and maybe i thought that mixing new caps and old caps could lead to a strange behavior, so it may make sense according to your explanations.
So I'll be back to you soon with my comments on how things go on.

 

[Handy Andy]

= Relax, you still have a radio!

 

[Christophe Barbe]

Hi Andy, i solved the "self-oscillation" kind of symptom by reflowing solders related to ground on lots of places on the PCB. It helped, i had a nice 15 mn QSO to France flawlessly.
SWR issue: still on going even after reflowing the solders of all the SWR-related circuit even the vertical board and having cleaned-up the pot with contact 61.
I am afraid some diode may have died on the SWR PCB.
Is there any way to have accurate readings without desoldering the whole thing and having the diodes in parallel with resistors?
If yes, do you have measurements of your copy or could you do so that we can make a comparison?
So that's it, even if it's a simple circuit, SWR fault not found yet.

 

[Handy Andy]

Ok, will need time -

This is what I'll will do, keep SWR board in place, and measure (with no antenna connected, power off) to see if the diodes conduct - and if only one way...

 

[Handy Andy]

Initial settings...

CAL on,
SWR CAL - Mid point





Now,
Meter was set to 200 ohm setting...
NOW SET FOR 20K, 20,000 ohms
When Third Trace from left to Board ground is checked...shows infinite...

This gets interesting...
4th Trace from LEFT (or 3rd from RIGHT)
shows 3.53...
AND IS VARIABLE!
Adjust your SWR CAL knob to verify this...


Turn front panel switch to SWR...
SAME 4th Trace from LEFT - SHOWS INFINITE


However 3rd trace from LEFT - shows Reading!
And also is variable!

All midpoint readings are in the 3.5K range...




​

 

[Handy Andy]

If one diode is bad - your resistive variable reading would change between FWD and REF and if you reversed the leads on those same two spots the variable would be infinite - so if the Diode is open, it would show infinite

If Diode is shorted, the variable would NOT BE EQUAL in CAL or SWR settings.

If diodes are good, they would show variable at mid point to be equal within 1K of tolerance. in my test 3.5...

 

[Christophe Barbe]

Oh great, thanks a LOT Andy, I'll do that today (Friday evening) after work see what readings I get and I'll report you what I find.
So i'll be back soon...

 

[Christophe Barbe]

Hello Andy,
I did the exact same way you did,
-> 200 ohm outer solder joints as on pic and i have a short too
-> I do have very similar readings (3.53-ish pot mid way) with testing ground and 3rd or 4th from left and depending on the position of the CAL pot on the front of the rig and according to same setting of CAL or SWR on switch.
--> Traces from resistors to the other side of rig PCB (front) : same as you, i have short reading for same trace.
so, all SAME as yours... but,

--> BUT, i do actually have differences when it comes to the testing of together BOTH the middle REF and FWD solder pads (the red and blue arrows on your picture) at the SWR board.
And depending on the position of the switch :
SWR switch position:
on 20 kohm range and depending on the pot position:
max 9.43 (fully clockwise)
mid ca. 11.55 (middle or so, as the front plate and buttons not on the rig)
min 13.20 (fully counterclockwise)
CAL switch position:
i have a short aka open (on your radio, you have open on both switch positions).

What do you think? one of the diodes?
Thanks again for the pics and your support Andy,
Cheers, Chris

 

[Handy Andy]

If you can - measuring from the two point back at the SWR card - FWD and REF -

Remove the 2 wires going to the pot on the front of the board, but keep ground...to it...

Then re-measure at those two points to eliminate the switch causing this issue.

If it still shows a resistance (not open) - then the card and the parts on it need to be checked.

Be extremely careful with the card, those traces have been there since the beginning - it can tear open or lift on you...

 

[Christophe Barbe]

Hi Andy, yes this is a great idea which will be next tomorrow morning.
I just had time today to re-mesure the 2 solder joints at the SWR PCB (red and blue arrows on your drawing) : my measurements on 20K multimeter position were false and good at the same time, I explain: it depends on polarity, you have a short on one way, but you have the values I wrote for the opposite polarity, and this, is working for both the SWR and CAL positions on the switch.
I guess you have the same on your radio, and to me it means the diodes are good. I tried to measure them in circuit and got around 0.405ish mesurements, which tells me they are good as Schottky diodes.
The 4 resistors (100, 150 and 2x 1k) are good too.
I can de-solder tomorrow the 2 leads going from the PCB to the switch to see what I get, but i am even more puzzled at where the problem can be :/
Any opinion Andy?
Thanks a lot by the way,
Chris

 

[Handy Andy]

This may be due to the switch or if you look at the board I have, these long runs are next to other traces that carry information (IF and Audio signal) so the issue of isolation comes to mind.

This is what makes this difficult.

You may have been working in the rear portion of the board, but as with any device of ages such as these, the issue of "dust rings" and Corrosion comes into play. IT can b e you "shifted" the case off kilter, and a key postiton is now no longer a part to ground.

At least the KIND of ground it should be.

This is why I tend to go back to the switch, and even the pots own mounting location.

The Case to backplane is no longer, or at least acts like, it's not grounded or BONDED for RF - it simply needs cleaning up and reseating ALL the controls.

Why would a meter show the same needle position while you operate a switch or the Calibrate knob? Realize that the Meters "sense" is all the way back to the rear of the board - if RF has found a different path, you will see it bypasses all the circuitry and works with what it knows - being the easiest return path (or patch) in the circuit - could be the side panels - (measure using your DVM on the Side panel to Foil ground for this effect you may see it)

You had a radio working fine until you started to recap it, now the SWR meter is goofing up, yet you also noticed POWER rising up when you didn't want it too, you seem to have tamed that problem...so you may need to look further and into more details about Ground plane effects and ground loops.

This WIKI article ( Search for: Capacitance instability ) talks about some of the problems with AGE and Disc caps - they fail gradually but then when they do, they act like isolators, insulation - for a greater range of frequencies - meaning they are nothing but dirt in fancy ceramic packaging - you'll need to replace them.

I showed you something earlier, and you didn't seem ot have read it very well, there are TWO types of 148DX models - two board, 010 and 879 - one uses 8719 while the other use 145106 - ok? do you see a pattern, you have NOT told me if your board is the same as the one I've sent you in pics...

Why does this matter?

Review the posts above, there are two types of RF grounding present in these two revisions of the same boards - you need to pay attention to their differences because what works for my radio may not be the same for yours due to a revision/Provisional change to the layouts between the two.

The front panel may be the issue - which is why I mentioned the switch - as you work thru the cleanup - even potentiometer shafts and switch housings can lose connection with the main backplane they're mounted to because their outline they were originally assembled and shipped and serviced life'd thru - has now changed.

• work the switches - cause their contacts to do some wiping action - while you are doing this - rotate that CAL knob - see if you can make the needle bounce or shift, if it does, then your issue with body capacitance qUirK revealing this grounding problem and why? because the potentiometer or even the entire case assembly for the front panel has an Open RF grounding problem - you have to locate it.

Just take some alcohol - fine steel brush and some simple oil - find a combination thereof that works for you, and using a hot iron - start carefully reheating some of the slide switches and apply some oil into that assembly to let the heat and detergents in the oil to loosen and liquefy the crud and oxides - don't need a ton of heat, but you will need to get after their "TiteBond" they used in the threaded screws to keep them seated - you have to do this - to remove them and reseat on a clean surface to help with this conduction problem.

Restoring a radio is a labor of love and like anything in love or dealing with issues of possession - you'll have to iron out your differences before marriage because otherwise one or both - will wind up in a junk pile somewhere - one rotting away in a landfill while another may wind up rotting in a tavern because it can't get back the time it lost in trying to make the effort work.

 

[Christophe Barbe]

Hi Andy, thanks for your time and your valuable explanations, this all for sure makes sense, and i'll find out, i need to, as i want this radio to work as it should.
I had read and answered if i remember well, it's a PB010AB, so it's got the 145106. Also, the rig started to show again that power increase symptom which tells me there's still an issue with grounding for sure.
So i'll take the time to clean-up things and re-flow the solder pads where i find it useful and especially in the transmit section.
Loss of energy : this should indeed be the explanation when i see roughly the exact same readings on both CAL and SWR positions not being able to have a full deviation of the s-meter to the right, only 3/4 (s9+ sort of reading).
So I'll do the search on the grounding side and on the face-plate side, switch and pot.
I'll be back here once I have any improvement.

 

[Christophe Barbe]

Hi Andy, finally had time to go on with the snake rig.
Short story: it had a cut on the upper trace on the SWR board making a short. It was barely visible and i realized the continuity check on this trace had not been done yet. Actually the only one left, as i checked everything else. This explains why the symptom showed as faulty diodes having continuity on both directions.
I scratched a little the trace to uncover the copper and applied some solder.
Now the functionality is back and works as it should.
Rig works nicely with great receive and transmit, maybe a slight frequency alignment is still left, at least checking on frequency meter how it's performing on this side.
I'll also have a look at the bias setting, it's supposed to be around 50mA if I am not mistaken, on both driver and PA side.
Thanks for your support, I have learned new things from your responses, and it's nice to have a radio working as it should (even if it's the SWR meter which is not the most accurate, but still, it's there, it has to work as it did originally).

 

[Handy Andy]

I'm relieved to know you got it working like it should.

This is kind of bittersweet to me because of not knowing what you had in front of you - it wasn't on my bench to see in more detail. I feel bad that we had to post back and forth in different times and it took weeks to resolve this. IF I had been any form of a counselor, I'd be fired...!

But you helped me answer some stigma's on other radios with SWR issues.

As you know some Cobra 29's have SWR alters on them so you can see a little red light blink away while you talk - usually it's a sign you're antennas' SWR is getting a little high.

But in other radios - this ANT warn light seems to serve more as a Modulation ON-The-AIR light with it winking away on your voice peaks - even though it should be an ANT warn light - some peaked and swing-set radios oftentimes have so much power going thru them the ANT light is more functional as a means for the user to see their voice peaks than to warn of impending doom of blown finals.

What I tried to discuss here, helped me with another thread - here - about SWR Calibration problems with built in metering so it wasn't a total loss on my end, but again for the effort in taking apart my radio to let you see into yours - let me see it in a different way to help me show others why things work, not just the "how".

Because of what I learned - the end result is what they produced to help guide you to use the radio properly - but they never really designed the SWR metering system in these, to offer any true source of 50 ohm matching - they just put resistance and caps in to tone down the power levels to keep the meter from burning up, yet provide some level of scale to show SWR. - they simply tried to make the needle fit the scale of the back-lit decal it used behind it.

 

[Christophe Barbe]

Hello Andi, thanks for all, and i am truly happy you did not waste your time with my issue. I am glad the radio works, i even got a record of its transmit audio and it's outstanding.
I do have another 148GTL-DX from 1981 the second i bought from a friend, literally untouched (with the exception of the lamp replacement) imagine that, it even has the CH9 and this particular one had an awful FM RX, L6 coil had to be tuned to get a normal FM receive (it sounded like i want to listen to an FM signal in AM), so this issue is solved, i'll have to realign the whole machine anyway. While i was at it, i replaced the meter window as it had scratches, and i don't like scratches.
Now I have a last issue before realignment, which is -I know not that big of a deal as lots of folks hate it anyway- the roger beep, and just as the SWR feature, it's not that needed after all, but since it's there i want it to work.
It does work in fact, but it's silent. It's doing it's job, i hear the time it lasts, so it works fine except it does not give a tone, it's a "silent beep".
Any hint ? I guess the audio part of it is not going to the mic path, what do you think ? It's a simple circuit i should find out eventually...