Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Trying to troubleshot 1960s Panasonic transistor radio

Status
Not open for further replies.

Skogas

New Member
Hello,

I am trying to resurrect an old Panasonic Transistor Radio that has an extremely low volume (and possibly other issues that I'm not aware of yet :) Have replaced all Electrolytic Capacitors but it didn't seem to solve the problem.

When doing some measurements on the final output stage and comparing it to what the schematic says - the numbers seem quite odd (also I don't really know what I'm doing :)

If I read it right, the two 2SB324 power amp transistors should be getting 0.12v at the base, 0v at emitter and 6v at collector.
However then I probe the contacts marked B/C/E on the board for these according to what schematics says, I get:

TR10
B = 5.78v
E = 4 - 4.6v (floats continuously)
C = 0.02v


TR4
B = 4.1 - 4.5v (floats continuously)
E = 4 - 4.6v (the emitter is shared so it's the same as on TR10
C = 0.02v

Would it be right to guess that these two transistors are bust, and if yes what would be an appropriate modern day equivalent, or could the be somewhere else completely?

Thank you

Screen Shot 2022-08-08 at 9.57.24 pm.png


Screen Shot 2022-08-08 at 9.52.51 pm.png
 
On first glance the emitters are supposed to be at 0V and they are not. This doesn't look like broken transistors, but more like a broken wire or trace from the emitters to the common (ground).
 
If I read it right, the two 2SB324 power amp transistors should be getting 0.12v at the base, 0v at emitter and 6v at collector.
However then I probe the contacts marked B/C/E on the board for these according to what schematics says, I get:

Remember its an old PNP transistor design, and positive ground; note the voltages on the drawing are negative.

I suspect you have the meter negative on power negative instead of positive supply? It's an easy mistake when you are used to working on newer equipment :)
 
Thank you very much for your response, it's a great help!
I've traced that line and it's supposed to go back to the "+" line via that 1.5kOhm resistor on the digram.
When I probe the resistance with the multimeter's positive on emitter and negative on that common "+" - the radio "wakes up" a bit and the meter shows something between -040 to -060, and there's no continuity other way around, so it looks like that resistor needs to be replaced.
 
Remember its an old PNP transistor design, and positive ground; note the voltages on the drawing are negative.

I suspect you have the meter negative on power negative instead of positive supply? It's an easy mistake when you are used to working on newer equipment :)
Thank you, yes I did that mistake too :)
Would it be correct to think that it only affects the + vs - on the multimeter reading for voltage measurements, with the values being otherwise correct, or are there any other consequences of this error? Cheers
 
Would it be correct to think that it only affects the + vs - on the multimeter reading for voltage measurements, with the values being otherwise correct, or are there any other consequences of this error? Cheers
If you have the common on the wrong supply terminal, the voltages you read are proportionally inverted compared to the values on the schematic, so you get (supply volts - voltage under test).

So, the output transistor emitters, which should be 0V, read as full supply; the -0.12V base reads as (supply - 0.12V) etc.
 
If you have the common on the wrong supply terminal, the voltages you read are proportionally inverted compared to the values on the schematic, so you get (supply volts - voltage under test).

So, the output transistor emitters, which should be 0V, read as full supply; the -0.12V base reads as (supply - 0.12V) etc.
Oh my, thank you very much for this explanation - it makes total sense now! Really appreciate all the help!
 
As someone who has repaired radios etc. since the valve days, it's usually easier not to get confused over positive and negative - just check which battery terminal goes to chassis, and connect the relevent colour probe to that (so red is common for a positive earth system like this one).

As is typical of this era of radio, the transistors are Germanium and PNP, NPN were still very rare at this point, and much more expensive - hence the almost universal use of PNP devices.

You're also unlikely to be able to easily find replacements, Germanium transistors are like hens teeth!.

For repairing almost anything, you should almost always use the 'half-split' method, and for a radio the absolute obvious start point is the volume control. Either inject a signal at the top of the volume control, or monitor th signal there with an external audio amplifier, or a scope if you have one - even acrystal ear piece would work.

Personally, one of the first 'jobs' I was given when I started as an apprentice, was to build a signal injector - just a simple multivibrator, running at about 1KHz, and powered from a 1.5V AA battery. This allows you to inject at audio, IF, or even RF, due to the harmonics of the squarewave. I used it for many years, far faster and easier than digging a signal generator out :D

I'm also not a big fan of blindly changing electrolytics - fault find first, and see what's wrong - if you should find a faulty electrolytic, change it then. Fault finding isn't randomly testing components either, trace or inject signals, find the suspect area, then measure voltages - then you should be able to isolate which component might be at fault - THEN you can test it.

Blindly changing capacitors, and randomly remove components for testing, is very like to cause more problems than you started with - and finding multiple added faults is far harder than finding one original one.

Once you've got something old (like this) repaired, then (if you wish) you could change the capacitors - but do it one at a time - testing if it still works between each change. Then if it stops working, there's only one component you could have messed up with - if you change 20 capacitors, and then it doesn't work, you've little clue where the issue might be.
 
If you see these in your radio, more typically in tube radios, replace them.

Dont wait for failure, they are notorious for failure.

1660149728526.png



Electrolytics back in the day were pretty much all wet types, and the dielectric fluid
simply evaps or leaks over time. Often damaging PCB so replacement prudent. One
can argue that replacing all is a good idea, especially as you can often catch several
bad ones and not have to re-discover one by one their working state. Also modern
caps have better sealing barriers and generally better quality and uniformity of the
cap. Progress in materials and methods have produced better results, generally speaking.

Also some radios back in that era suffered from "silver mica disease", where some
radio IFs used a sandwiched construction cap in the IF cans. If you find a can that
cant be tuned thats a good possibility it has these built in caps, and should be replaced.
When tuning old IFs if the slug does not move I found applying heat with a hair dryer
would soften the wax tube they travel in and slug starts moving. Do not use metallic
screw driver or allen hex wrenches on them, that typically cracks them and they stop
moving and their inductance of the IF coil changes way out of design spec. Use
traditional tools, plastic and nylon.

There is a good forum over here for advice and methods :

https://antiqueradios.com/resources/index.html

Lastly carbon comp Rs, especially ones that worked in a high dissipation environment, will
have shifted the resistance significantly. I cant speak for transistor radios this impact, but
for tube radios first thing I check is all plate Rs, normally high way out of spec. Cathode Rs
next. Other Rs absorb moisture over time (grid Rs) and go low typically.



Regards, Dana.
 
Last edited:
Were they of American origin?, I've never seen any - so they 'probably' won't be in a Japanese radio. Likewise, Wima (German?) non electrolytic capacitors were notoriously unreliable, but again aren't likely to be found in a Japanese radio.
 
As someone who has repaired radios etc. since the valve days, it's usually easier not to get confused over positive and negative - just check which battery terminal goes to chassis, and connect the relevent colour probe to that (so red is common for a positive earth system like this one).
Good tip, thank you Nigel!

You're also unlikely to be able to easily find replacements, Germanium transistors are like hens teeth!.
Luckily it seems like the transistors are ok, so hopefully no replacements will need to be found.

For repairing almost anything, you should almost always use the 'half-split' method, and for a radio the absolute obvious start point is the volume control. Either inject a signal at the top of the volume control, or monitor th signal there with an external audio amplifier, or a scope if you have one - even acrystal ear piece would work.

Personally, one of the first 'jobs' I was given when I started as an apprentice, was to build a signal injector - just a simple multivibrator, running at about 1KHz, and powered from a 1.5V AA battery. This allows you to inject at audio, IF, or even RF, due to the harmonics of the squarewave. I used it for many years, far faster and easier than digging a signal generator out :D
Unfortunately with this being a 'one off' radio fix for me, the multimeter is the only piece of equipment I had available and it didn't seem worthwhile investing into a signal generator unless it was absolutely necessary.

I'm also not a big fan of blindly changing electrolytics - fault find first, and see what's wrong - if you should find a faulty electrolytic, change it then. Fault finding isn't randomly testing components either, trace or inject signals, find the suspect area, then measure voltages - then you should be able to isolate which component might be at fault - THEN you can test it.
This sounds like a logical approach if you have the knowledge and equipment :) With me being an absolute amateur - the approach fault finding was:

1) Find suspects that are easy to eliminate. A few electrolytic caps had their seats visibly bulging / swollen, so it was an easy target to pursue. After replacing the caps - the radio went from silent to quiet, so I think it paid off.

2) Then I measured the voltages and compared them to the values on the schematics, starting with the output stage. This lead me to the suspicion around those two push-pull transistors, but I couldn't make sense of the readings I was getting which lead to point (3)

3) Ask the experts - which was the reason for this post :) And it proved to be really helpful in both - learning and fixing the radio.


Blindly changing capacitors, and randomly remove components for testing, is very like to cause more problems than you started with - and finding multiple added faults is far harder than finding one original one.
I guess anything invasive poses a certain risk, but having started with dead radio - it didn't seem like there was too much to lose.

Once you've got something old (like this) repaired, then (if you wish) you could change the capacitors - but do it one at a time - testing if it still works between each change. Then if it stops working, there's only one component you could have messed up with - if you change 20 capacitors, and then it doesn't work, you've little clue where the issue might be.
So far I have to thank Dick Cappels who replied the first and was spot on - the resistor from emitters to + line was burned out and was passing the current only one way, so the approach and findings of the diagnostics weren't all that hopeless, they just needed the conclusion of the expert :)
After replacing this resistor the radio came back alive and seems to sound pretty close to what I remember them to be when I was a kid :)

The only thing that bugs me a little is that the schematic seems to have prescribed a 1.5kOhm resistor in that spot, (and I couldn't read off the markings of the actual one since they were melted), however trying the 1.5kOhm replacement sounded too quiet, so I ended up with 220Ohm one.
With the radio now working seemingly fine, I wonder whether there are any downsides in having a lower value replacement in that spot.

Thanks again for your recommendations Nigel, if I happen to come across another such job I'll definitely take it on board, but so far I'm just happy that the radio came back alive even with my amateur troubleshooting.
 
Can you post more of the schematic?, the tiny section you posted shows hardly anyhting, including not showing the resistor you mentioned, or the driver circuit.

As we don't know the model number, we can't even google it?.

I'd like to see where the 1.5KOhm is, as from your description it sounds a fairly unlikely value.

You don't need much test equipment, a meter is fine - just build yourself a multivibrator signal injector - two transistors, four resistors, three capacitors, and one AA battery - a push button switch is good as well.

A quick google finds plenty of examples, such as this one:

 
Thanks Nigel, building a signal injector looks like a great project to try.

Please find the schematic attached - it's a National Panasonic R-1400
The resistor is marked R K1.5, so my guess is that K means thousand?
 

Attachments

  • panasonic-1400.pdf
    223 KB · Views: 350
Is that resistor 100K, 1 followed by 5 "0's" ?

I cant find an EIA standard that covers that marking so wonder ? Or is
that a schematic nomenclature ?


Regards, Dana.
 
what was the meaning of that 'K' before 1.5 so that I know how to read it next time?
Look at Note 6 on the circuit diagram.
Upper case K = 10% tolerance.

JimB
 
Look at Note 6 on the circuit diagram.
Upper case K = 10% tolerance.

JimB
Thank you Jim! It just seems so obvious once you've pointed me to the right answer! :)
Another lesson - always read the notes, they might reveal that there are two capital "K"s used for completely different parameters (my brain just defaulted to Kilo Ohms seeing K next to the resistor value :banghead:)
But now all mysteries seem to be solved, so thanks again!
 
I can recall the turmoil and confusion when troubleshooting these new transistor devices. We were taught with tube circuits, which had wide tolerance, but still easy to troubleshoot because of the wide differences in tube terminal voltages. Only a voltmeter was needed. Then these current devices came out. They were only mentioned in "tube" school. But a few years in the navy cleared that up. They put transistors to work much quicker than commercial electronics. But solid state has always been more difficult to trouble shoot. I've seen EE graduates spend hours and days trying to troubleshoot solid state. No money can be made that way. Then to top it off, we went disposable....just change out the whole circuit. ...( maybe because no one could repair them).

So troubleshooting for circuit repair.....died. A waist of time. Or so I thought for about 15 years.

That's when I learned how valuable Lissajous patterns were. There is a small learning curve and there is some skill or art to it. But no terminal labels, no prints and no power needed. And the state and the crispness of a PN junction can easily be observed. As with the passive components too. So when I troubleshoot, I turn off the power and get out the octopus. It's truly amazing how well this works, and I always gave a 90 day warranty when charging a fee. Very few were ever returned. I have no idea of what the boards were use for.

The boards that no one could repair were call dogs. That was my specialty for quite a while. So if you have some dogs laying around, you should try this out. It might surprise you.

It's just an alternative you might consider. Even fun.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top