148GTL SSB Strange Drift

[Redbeard U812]

My 148 transmits great on AM, but on SSB I am told it drifts or is off freq. Latest report by friend is that it sound garbled when I first key up but then levels out. 148 is a Taiwan from about '78, has the "poor man's" channel mod, 4 watt dead key, clarifier is unlocked (rolls 8 each way from center) Astatic NC Mic, freq counter on output. The freq only registers when keyed on AM, oscillates a little when modulation swings in AM. Freq acts like a digital slot machine on SSB, but with a tone it will register close to AM carrier.

My guess is the DC power supply will not keep up with power when modulation swings in SSB. Could be leaky Caps. Could be the AM regulator is failing and not keeping up. on and on and on........we go.

It sounds great on AM but SSB just wont perform in trans.

anyone else have ideas or suggestions?

 

[kaos513]

Maybe needs a recap job and good alignment.

 

[kopcicle]

Re-cap can be a placebo with no other effect than having your wallet emptied for no good reason.

Because of part number and wire color variations over the years I will generalize.

Locate the regulated 8v from the MB3756 and check it in rec and TX . It should not budge.

Again , locate the voltage input to the "clarifier" The top of the voltage divider. It should be predictably just less than at the MB3756 and not budge a bit. If it does it will change frequency in receive if unmodified and TX as well if modified. Then follow the output of the MB3756 to the "master frequency adjust" (TX adjust) and see if it is stable at the top of the voltage divider as well as the output. Note: the closer you get to the varactor the more you increase the possibility of affecting the voltage just by measuring it.

The 2% solution. Occasionally a 148 et -all will drift due to lack of lock. It isn't so much a voltage regulation issue as the VCO tuning slug being on one edge or the other of adjustment. This is usually seen as an issue with reduced supply voltage or temperature.

All I got for now

 

[Robb]

Re-cap can be a placebo with no other effect than having your wallet emptied for no good reason.

Because of part number and wire color variations over the years I will generalize.

Locate the regulated 8v from the MB3756 and check it in rec and TX . It should not budge.

Again , locate the voltage input to the "clarifier" The top of the voltage divider. It should be predictably just less than at the MB3756 and not budge a bit. If it does it will change frequency in receive if unmodified and TX as well if modified. Then follow the output of the MB3756 to the "master frequency adjust" (TX adjust) and see if it is stable at the top of the voltage divider as well as the output. Note: the closer you get to the varactor the more you increase the possibility of affecting the voltage just by measuring it.

The 2% solution. Occasionally a 148 et -all will drift due to lack of lock. It isn't so much a voltage regulation issue as the VCO tuning slug being on one edge or the other of adjustment. This is usually seen as an issue with reduced supply voltage or temperature.

All I got for now

I'd agree that replacing caps can be a waste of time sometimes. We've seen SRC replace caps in new radios, which we all agree that is a waste. But not in a radio that is over 30 years old. Then it is just being prudent and saving time.

 

[Martian]

+1 on replacing caps in 40 year old radios like a 78 vintage 148.

The counter isn't going to be able to count AM modulated signals properly.
It should be fine with just a carrier.
Sideband has no carrier so that's why it seems to work with a tone but off frequency.
I am guessing this counter samples RF at the antenna. If so, counter is behaving normally.

 

[nomadradio]

Drifts?

Off frequency?

Garbled?

Could this be three separate problems?

Or just one, with a different description by each operator?

There is no substitute for hearing the problem yourself. You really need a second radio to hear this problem with your own ears.

Never underestimate how many different ways a sideband transmit signal can be distorted.

If the radio's frequency is changing in step with the modulation, it won't affect the sound of AM mode all that much. Sideband is different.

Quick way is to put your steady tone into the mike. Transmit with the mike gain at zero to start. Slowly turn up the mike gain. The pitch of the tone you hear in your monitor receiver should not change. If it does, this is what's garbling your voice audio.

Or just one part of it.

Modulating the radio with your voice, start with the mike gain at zero. Turn it up so the wattmeter shows around 1 Watt, no more. Should sound nice and clear. Now turn up the mike gain and compare.

If it sounds rough and gravelly when turned 'way down, this is a bias problem. If it sounds okay until you turn the mike gain all the way up, that suggests the limiter (ALC) is disabled or turned too high.

You don't need a room full of fancy equipment to narrow down the root of the problem.

But there is no substitute for hearing it yourself.

Relying on the reports of other operators is not reliable.

73

 

[Redbeard U812]

Nomad,

Thanks for the info. I don't have a bunch of fancy equipment. I like the KISS method myself. Have a couple other radios will try your method. I agree relying on what or how my friend describes my TX has varied. Every time we discuss simple ways to diagnose problem he comes up with another problem or variable way off track. Trying to discuss isolating a problem with him sometimes reminds me of trying to hold a discussion with a teenager with ADD. Not really sure his set up is functioning correctly either, though he claims it's "dead on" reported by others. I hear others just fine, but not his. I'll figure out a second set up, and run another antenna some how with good GP and ground.

The Re-Cap is hopefully not what I need. Though radio is old it may be something needed.

My theory, not enough DC power out of the VISTA IV-R (really old power supply), needs addressed.

It's reported that my transmit is garbled at key up then levels out after continued transmit.

 

[kopcicle]

VISTA IV-R (really old power supply), needs addressed.

yup
Let's look at this for a sec...

Higher the voltage the less the current but wait on that a bit.
So the Vista is reported as being a 4 A supply, at 12V. That's 48W
So to produce 12w (peak) output with a 50% efficient (don't try to tell me that circuit is any better than 50%) device that's 24w.
But wait there's more. The 2166 driver has to put out nearly 3w peak so that's another 6w.
Then the 1730L @ 2w total (never measured it out of circuit) or for instance the ECG 340 at nearly 4w total (okay that I have tested and its about 3.7 in a tuned circuit) .
Then the grain of wheat bulb for another 6w.
How about the mb8719 and the mb3756? Only about 20ma each .
The rest of the small signal transistors for maybe a watt.

2+11+24+48=85w / 12 =7.08A@12v
2+11+24+48=85w / 13.8=6.15A@ 13.8v

I didn't even include the TA7222 audio chip

My point is , put a 2A fast blow fuse in there and see how long it lasts at a SSB whistle.
Over and over again I try to explain this and it falls on deaf ears. No matter what Cobra/Dynascan/Uniden say a 148 draws a minimum of 5A TX and 7A single tone SSB.

Don't get me started on the other side of this issue. A 20A fused pair of 1446/455/2290 will not output 200 anything watts. Not Bird not RMS not PEP not magical pixi dust watts.

There just isn't any such thing as a free lunch. You have spend (dissipate) watts to make watts.

 

[Redbeard U812]

yup
Let's look at this for a sec...

Higher the voltage the less the current but wait on that a bit.
So the Vista is reported as being a 4 A supply, at 12V. That's 48W
So to produce 12w (peak) output with a 50% efficient (don't try to tell me that circuit is any better than 50%) device that's 24w.
But wait there's more. The 2166 driver has to put out nearly 3w peak so that's another 6w.
Then the 1730L @ 2w total (never measured it out of circuit) or for instance the ECG 340 at nearly 4w total (okay that I have tested and its about 3.7 in a tuned circuit) .
Then the grain of wheat bulb for another 6w.
How about the mb8719 and the mb3756? Only about 20ma each .
The rest of the small signal transistors for maybe a watt.

2+11+24+48=85w / 12 =7.08A@12v
2+11+24+48=85w / 13.8=6.15A@ 13.8v

I didn't even include the TA7222 audio chip

My point is , put a 2A fast blow fuse in there and see how long it lasts at a SSB whistle.
Over and over again I try to explain this and it falls on deaf ears. No matter what Cobra/Dynascan/Uniden say a 148 draws a minimum of 5A TX and 7A single tone SSB.

Don't get me started on the other side of this issue. A 20A fused pair of 1446/455/2290 will not output 200 anything watts. Not Bird not RMS not PEP not magical pixi dust watts.

There just isn't any such thing as a free lunch. You have spend (dissipate) watts to make watts.

Thank you Kopcicle,

It seems it was what I suspected and you were able put down with math. Using the 50 amp DC power supply in my camper I am able to stabilize on SSB. Ran separate fused line to the radio from the DC output of 13.8 volts. Was able to hold a understandable conversation with my test subject, though Kav claimed to have had to clarify with varied results. He just can't leave a knob well enough alone. Once he stopped fiddling with clarifier on his end we were able to hold conversation. The slide on mine is very touchy, can scroll up or down a complete channel. May be some other quirks that will rear their ugly face, but for now it seems to work. AM dead keys 4w swings to 23, SSB modulation will take it about 30w. Was reading considerably less before.

 

[Handy Andy]

Between both Kopcicle and Redbeard - both of you worked together on this and I only wanted to point out what Redbeard said...

Thank you Kopcicle,

The slide on mine is very touchy, can scroll up or down a complete channel. May be some other quirks that will rear their ugly face, but for now it seems to work.

...whew - glad to know that getting the power supply stable did a lot...

But I also wanted to point out too, in another thread there was talk about the level of voltage used to drive the tuning circuit - so if there were any variations - like you'd have with a touchy clarifier (Voicelock) The voltage drop on heavy modulation peaks can cause problems even with regulation of the 8 volt constant.

Anyone who's had to deal with Cobra 148F radios - knows that unless you break the power supply feed into two branches before it goes thru the power choke - you will have a "Warble" in your transmit - on any mode. Because the power supply feed to the MB3756 has to have a minimum of 1.5 volts ABOVE its' regulated output to be regulated - so the design the Malaysian model had, already was strangled from the get go. The draw thru the choke was enough to drop voltage down below the rated minimum input level the MB3756 could work with - and start to drop out - this was a fault on the supply side and design itself - not from the user whom just wanted to use the radio - it happened even when it was stock.

So I thank the both of you to show how to work together and also to help bring up a good issue of clean power in a radio makes a bigger difference than clean power out of it - for you need stable clean input power to get those results on the output side too...

Sorry for the preposition - and dangling participle but - Gosh Darn It! Hi - Hi ! ....

 

[Redbeard U812]

You are saying the 148 by design has a flaw which is a tendency to "Warble" ? I knew when I got this 148gtl it may be a problem child. But I just had to have it, a Taiwan made 148. Old man Charlie who owns Charlie's Airwaves had it and I felt confident that he was old school tech. Man he has all kinds of older radios. After poor results I was starting to wonder about my decision. This 148 turned into a base rather than being installed into my truck due to the touchy clarifier.

Thanks to all here for being forthcoming with information to help remedy problems.

 

[Handy Andy]

Just Malaysian made ones - the Philippines and Taiwan models had SEPARATED the power feeds. One - the power choke one, fed the Audio chip (7222AP) and the Driver and Final - the main feed then went off to MB3756 and TR28 to handle LED brightness. - They branched off from the power feed wire from the Volume On switch back at the board.

So as the audio feed may have "sagged " under load - it only got distorted - the MB3756 never saw the load variations because it did not get it's power going thru the power choke. In this setup the MB3756 did ok - did not suffer such a fate of sagging power load on voice peaks - it hummed right along.

As long as you kept the number of mods simple and not to much audio drive - no radio would have a problem with regulation and line-load variations.

When you put all the eggs into one basket - you tip the basket - you know the rest of the story.

 

[nomadradio]

Ah, a "touchy" clarifier?

If it's just in need of a squirt of control cleaner, that's the best case.

And if it's expanded to cover too wide a tuning range, this makes tuning sideband impractical. I tell a customer who asks for this that he'll make into an AM-only radio if the clarifier covers too wide a range.

I do remember one old guy back in the early 80s who fixed this problem on his Cobra 142 by drilling a 1/8-inch hole sideways through his clarifier knob. Stuck a 1-inch long bolt into the hole and tuned it with his thumb. The longer radius of the bolt had the effect of giving him a larger 'tuning knob' that was easier to fine tune. Looked weird, but it worked for him.

73

 

[Redbeard U812]

Todays afternoon report on SSB was much better for this 148. We got both radios "dialed in" so to speak. Found Kav has a mic connection that needs mended. Installed my Road Devil for extra punch on mine and SSB was doing well, although I did barely adjust my clarifier and it threw us way off. Had to reset and let Kav track back in. LSB 38 right now is coming in here and there from a little skip.

Would like to get some of that sensitivity removed from this radio. Thanks Nomad for the suggestion, but I think drilling a hole in the knob is a little extreme. What Charlie did to this radio is unknown, I tried to consult him about it, but he just could not remember. He had 4 or 5 148gtl's at the time in a case. I had him pick which one would be best, hoping to get the best out of his selection. He's a good guy, don't figure he went to extreme, but do feel it's wound a little tight.

 

[Dmans]

RedbeardU812,
My recent experience with a "Made in The Maylasia" 148GTL has shown me that any mods to the clarifier that have you remove resistors (R174 and R175) and add jumpers is a recipe for "touchiness" and erratic Clarifier operation.
If memory serves R174 is a 22Kohm and R175 is a 2.2Kohm (or vice versa)-one value 10% of the other. On mine, I replaced the clarifier pot, (it had been dropped and the shaft was bent and had a "dead spot" in its travel) I believe I reduced the resistance values to 180Kohms and 1.8Kohms. The clarifier travel is still a bit much for me (a little more than 1 channel up and down) but no touchiness to the operation. My plan is to change those resistors back to stock values to make it even smoother "Sliding". I really only want about 7khz both ways and an even "Slide" both up and down. The clarifier supply voltage (on mine) is tied to the regulated pin of the MB3756 (pin 1 I believe) so I think the voltage is fine.

Another issue has reared its head on this unit first-no bias supply regulation. Until that is solved, the clarifier is on hold.

Good luck

73's
David