Grundig AM/FM portable "Transistor 305" troubleshooting

klem kedidelhopper · 2012-03-05T03:47:09+00:00

I would like to ask for help once again with the repair of my ownpersonal old radio. I started this project almost a year ago, and it'sbeen on the back burner since then. But I had a little free time andso I put it back on the bench. I would really like to see this radioworking again.This is a Grundig "transistor 305", circa 1970 I think. The radio isintermittent. I have been working on this set on and off for almostthe past year. I recently took some voltage measurements and there aresome serious differences in base and emitter voltages in the last IFstage and the detector stage when this radio fails. I have tried manythings, heat and cold, including substituting all the transistors inthe RF and IF sections one at a time and have studied the schematicuntil I felt a migraine coming on. l think that by now I could almostdraw it from memory, and I'm still coming up empty.I considered trying to inject a signal and/or signal tracing, howeverwith the circuit voltages so far out of whack when it fails I thinkthat approach would be futile. The base and emitter voltages in thesethree stages increase substantially when the radio quits. Somevoltages are jumping from for example 1.2V up to 4.0 V. Withconditions like this you would think that the cause would be obvious,and perhaps I've been staring at this this too long and it is. Howeverthe remedy has stubbornly eluded me thus far. The intermittent natureof the problem is what makes it so difficult.This is typical of what is going on: If the radio has been off for along period of time it will typically work for several hours and thenquit. In the beginning of this Quitting sequence, just after it firstquits it will try to come back on intermittently for a short period oftime, crackling etc, (as though something was intermittently breakingdown). Eventually it will just remain silent. There was anelectrolytic in the audio driver section, C57 a 100uf/3.0 volt withslightly high ESR. Replacing it improved the low frequency...

K

klem kedidelhopper 14 years ago

n

Given the intermittent nature of this job so far I have trouble actually saying that it's repaired, However I now have two periods of time wherein it has operated without a problem for at least 14 hours continuously, with a few hours powered down in between. It's never done that since this problem began. So I am starting to feel a bit more certain about the validity of this repair.

I reset the bias to 1.18V yesterday and last night I noted that it had drifted up to 1.23V. It's not a big excursion but I will keep an eye on it though comparing initial turn on with after a few hours running over the next couple of days or so.

There are a few things I still have to do. All the can grounds are.currently unsoldered and disconnected from the circuit. After I log some more time on this radio I'll need to restore those and see if that makes any difference. I don't expect that it will but I want to take this one step at a time.

The other thing I want to do is remove FIV one more time and put back the transistor, (AF126 II) that was originally there. In actuality that transistor is an AF124 which I purchased from Grundig service in Manhattan about 35 years ago. The AF126 II was the problem with the radio when I got it from a friend back around 1975 and not having a

126, Grundig substituted the 124 for me. It never seemed to affect the alignment so I see no reason to not use it over again. The replacement transistors that I have been using through out this repair came from a friend in Sweden and are likely as old as my originals anyway.

So as it turns out I didn't need to replace any of them. Therefore the alignment shouldn't have been affected, nor should the replacement of C38 have affected the alignment much if at all either.

It occurred to me that if this were a customers radio, I'd likely own their home at this point...

Lenny

Vote

W

Winston 14 years ago

at ground level, but it's not. That's the

the base at the voltage level seen.

Yup. You're right.

It almost must have been be an intermittently leaky C38.

--Winston

Vote

W

Winston 14 years ago

Were it me, I would move on to other things.

If the radio repeats the symptoms, then it wasn't an intermittently leaky C38. At that point, I would open the investigation once more.

Hey, it's working. Be happy. :)

More likely, they'd own your shop. :)

--Winston

Vote

K

klem kedidelhopper 14 years ago

re

So I've replaced all the original transistors, resoldered the transistor cans, (grounds) and reset the bias with R22 to -1.18V with

7.50V applied to the battery connector. That bias voltage however rises to -1.33V when the supply is boosted up to 9.0V. So as I expected, it seems that the supply voltage does make a small difference in bias. Does anyone else have any theories as to what Grundig may have had in mind when they specified that bias as well as the other circuit voltages? Was it with 7.50V applied or 9.0V? Lenny

Vote

W

Winston 14 years ago

(...)

The way I read Grundig's schematic notes, they want us to set the lab supply so that 7.0 V is seen on the power rail if in 'Medium Wave' mode or 6.75 V on the rail if in 'FM' mode.

If that requires 7.8 V on the actual battery connector or 12.7 V on the actual battery connector, it does not matter as long as the rail voltage is set correctly for the function chosen.

They want the radio to function well with brand new batteries or almost dead batteries.

I would not sweat over bias drift when a real battery is connected. The documentation is pretty clear on their requirement.

--Winston

Vote

J

josephkk 14 years ago

own

and it's

and

radio

radio is

almost

there are

last =A0IF

many

transistors in

schematic

almost

however

think

these

obvious,

However

nature

for a

then

first

period of

breaking

with

You can

it

that

well.

after a

instance

or so

again

before it

almost

check

improvement

listed

I am,

=A0look

during Go

stages

I'm

unit

voltage

ride

since I

for any

=A0 =A0 =A0 =A0Voltage readingso

=A0Radio not working

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-1.20V

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-1.50V

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-1.20V

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-1.44V

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-0.90V

=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0-1.30V

=A0 =A0 =A0 =A0 =A0 =A0-3.70V *

=A0 =A0 =A0 =A0 =A0 =A0-4.00V *

=A0 =A0 =A0 =A0 =A0 =A0 =A0-3.70V *

=A0 =A0 =A0 =A0 =A0 =A0 =A0-3.60V *

=A0 =A0 =A0 =A0 =A0 =A0 -0.80V

=A0 =A0 =A0 =A0 =A0 =A0 -2.95V

=A0 =A0 =A0 =A0 =A0 =A0 -0.62V

=A0 =A0 =A0 =A0 =A0 =A0 -3.10V

(hereinafter

so will

would

being a

current

current of

base

were 42,

be

of, pin

126 I

50, so

but

resistor,

supply to

admittedly

whose

the

used.

resoldered

8 of

After reading this thread through to this post i have a question. Can = you find the traces where R24 and R27 meet? I suspect a crack between there and the + rail.

?-)

Vote

S

Sylvia Else 14 years ago

That wouldn't have the observed effect on the base and emitter voltages.

Sylvia.

Vote

S

Sylvia Else 14 years ago

There's a note on the schematic that they're measured with a battery voltage of 7.5V. Using a vaccum tube volt-meter, which you might have some difficulty obtaining now ;)

I'm surprised at the variation in bias with supply voltage. I would have expected it to be quite insensitive.

Sylvia.

Vote

W

Winston 14 years ago

(...)

of 7.5V.

That looks like a typo.

R46 will drop 100 mV for every milliampere of battery current.

The only way to get the specified rail voltage of

7.0 V at 20 mA (Medium Wave operation) is with a battery voltage of 9.0. Similarly, it takes 8.95 V of battery voltage to get 6.75 V at 22 mA on the rail, specified for FM operation.

I'm willing to believe that a modern multimeter is more than sufficient for these measurements. My Fluke 189 has an input impedance of 10 M. I'm not too concerned about the additional 82 nA it costs to use, for a 9 V measurement.

:)

--Winston

Vote

W

William Sommerwerck 14 years ago

I'm guessing that the 7.5V specs are for the lowest voltage the designers felt gave acceptable performance.

Regardless, the important thing is the /relative/ voltages on a device's terminals.

The base-emitter junction is forward-biased, about 0.3V for Ge devices, 0.6V for Si. The base-collector junction is reverse-biased.

Vote

W

Winston 14 years ago

I feel that Grundig assumed the service person would connect their bench supply between rail ground and 'battery +',

*not between 'battery -' and 'battery +'*.

Unfortunately, they call 'power rail voltage' and 'battery voltage' the same thing. They are different by 2 V in MW mode!

--Winston

Vote

W

William Sommerwerck 14 years ago

While we're at it... It's perfectly normal for amplifier stages -- tube or transistor, in any kind of device -- to be fed through a small resistor, with a largish capacitor to ground. This "decouples" the stage from the power supply, to prevent feedback. The resistor /is not/ present to reduce the power-supply voltage.

Vote

W

Winston 14 years ago

[the same thing. They are different by 2 V in MW mode! ]

Thanks for fixing that William, but that's not what I said or meant.

The MW and FM models are the same radio model.

When testing in Medium Wave *mode* the radio requires 7.5 V between system ground and the power rail and is specified to draw ca 20 mA. In FM *mode*, the radio is specified to draw ca 22 mA from that same 7.5 V rail.

This is *not* the same as a 'battery voltage' of 7.5 V. Our friends at Grundig have the same phrase to describe 'rail voltage' and 'battery voltage', which are normally ca ~2 V different on the 'return' side.

I now understand that it makes perfect sense to attach our service supply between 'system ground' and 'power rail' so that the 'system ground' can be bonded to earth ground via the power supply, for safety. This is not the same as attaching the supply across the battery, however.

Yup. In this case, they are decoupling the final audio stage from the rest of the radio.

Yet it does, by about 2 V from the battery to power rail.

--Winston

Vote

W

Winston 14 years ago

(...)

Mia Culpa! The specification on the schematic is not a *range* of acceptable rail voltages, it is a declaration that the rail voltage must be +7 V in MW mode or +6.75 V in FM mode, (An inch to the right, they re-declare both at 7.5 V on the rail!)

I'd go with 7.5 V between ground and the power rail. I'd leave 'battery minus' floating.

--Winston

Vote

K

klem kedidelhopper 14 years ago

or

,

uce

I was using my RCA Senior Voltohmyst in the beginning. It has either a

10 or 11M impedance, however I built it from a kit I think in 1964. So being almost 50 years old it's calibration could have been suspect. I then switched to my Fluke digital. In any case I found that the readings on both were very close anyway. (I don't use the VTVM much anymore unless I get an old tube set in here to work on).

My initial reading on R24 before I made any adjustments, with 7.50V applied to the battery terminals was 1.00V. That voltage rose to

3.60V when C38 would fail. Remember, (after replacing C38) that I noted an approximate .015 V rise in VR24 when I increased the supply from 7.50V to 9.0V. Assuming the engineers meant for this voltage to be set with 9.0V applied, then perhaps I was reading what would be a perfect bias point (with 7.50V applied), as measured on VR24 initially. It would be nice to know for certain after such an exercise, but at this point I doubt very much that even someone at Grundig would be able to shed any further light on this. But then what you're saying makes perfect sense too Winston. I'm just really surprised at the ambiguity of the statement on the schematic. In any event there seems to be an amount of latitude available here, and given (and I'm assuming) that the radio appears to have been working that way for many years, perhaps it's a waste of time to fret over . 015V. Maybe I'll set it somewhere in between, enjoy my radio once again and step back into reality....That would make my wife, who has failed to see the sense in fixing a 40 year old radio very happy indeed. Lenny

Vote

W

Winston 14 years ago

They were great tools in their day. I had a PACO and lusted after a Voltohmyst. :)

formatting link

Yup. Grundig said "Battery Voltage" but they meant "Rail Voltage". The two are different. By 2 V!

I think we have the straight info. I bet the bias when set with 7.5 V *on the rail* would not drift very much at all when the radio is powered from a 9 V battery.

Someone should write a Grundig Schematic Decoder!

They do present a lot of information but it does take time to digest the hieroglyphics in order to translate.

I agree. It is probably as close as it needs to be.

I agree. There are plenty of 40 year old electronics to repair everywhere. You should be very pleased to have your nifty radio back! :)

--Winston

Vote

K

klem kedidelhopper 14 years ago

be or

or,

educe

Well I couldn't resist, so to take this a bit further I checked the power supply supplied with the radio. It is rated at 9.00V .200A. The no load voltage is 9.50V. With it connected to the radio and with the radio playing softly its output drops to 9.25V. In looking at the schematic again and rereading the statement a few more times I now do believe that in spite of the ambiguity, the original intention was to supply 7.50 V to the battery terminals, however reference all adjustments and voltages to the positive side of C59, (system ground). So I did that and set the bias on FM to -1.18V. (AF126 I E) I then recorded the following voltages:

Voltage readings Test point Radio working:

------------------- --------------------

E. AF178 -1.15V B. AF178 -1.48V E. AF124 -1.14V B. AF124 -1.37V E. AF121 -0.90V B. AF121 -1.24V E. AF126 (1) -1.18V B. AF126 (1) -1.47V E. AF126 (II) -1.14V B. AF126 (II) -1.34V

The voltage across C59 is 6.65V So now although the radio has been working fine on both AM and FM, I notice that the first two stages are reading a bit high. I wonder if I may have another leaky cap. I was looking at possibly C12, C19 and C32. Assuming I haven't driven everyone nuts with this project yet I wonder what is the general consensus with this new revelation is? Lenny

Vote

W

Winston 14 years ago

(...)

Here, I added a Vnom column.

You are testing at Vrail = 6.65 V?

Grundig says "don't bother warming up the VTVM unless Vrail ==7.5 V" ( I heard them. Really I did. :)

The little 'K' symbol next to these three apparently denotes a ceramic cap. These 500 V units could become extremely leaky at < 2 V I suppose.

I wouldn't bet that way, however.

There are lots more things that need your attention more than your beloved radio. Personally, I think that at this rate, in another 30 days, you will become borderline obsessive. :)

My advice: Button it up and listen to it while doing one or more:

Clean or replace the sacrificial anode in your water heater

Do an oil change on your car

Flush and fill your car radiator

Rotate your tires

Replace the transmission fluid in your car (Clean your funnel and flush with clean ATF first!)

Grab a basket and fill it with stuff from the back seat and trunk. Neatly put away, toss or donate the contents of the basket.

Shampoo car carpets

Diagnose that rattle in your clothes dryer

Vacuum out your dryer exhaust vent

Clean your gutters

Replace clogged rafter end vents with new ones