No FM stereo on a Tandberg receiver

[stevehertz]

A Tandberg Huldra 12 is a receiver that is pretty much a Tandberg 2025 (FM + MW) plus added long and short waves. This one exhibits the following faults on FM:

1) On a normally strong station the ‘stereo’ indication LED does not illuminate.
2) I'm pretty sure it is only reproducing mono (mono/stereo switch has no effect) and I cannot hear any stereo effects in ‘stereo’ mode.
3) Operating the mute button mutes all stations, even the ones that I receive strongly on other sets.
4) The signal strength meter does not move.

Other points:

The set works fine otherwise, all other bands and modes etc
The signal strength meter works on AM and also on FM only when it is used as a frequency meter as a visual aid to set the presets.
The aerial input panel comprising a ‘click in’ plastic moulding with with FM 75 and 300 ohm inputs and an AM input has been Araldited into the back of the chassis as one of the ‘click in’ end lugs is missing. The person who did it got the stuff everywhere and the moulding is literally plastered in place all around it and over it. In places, the Araldite bridges antenna connections and the chassis..

Taking cognisance of all of the above, it seems to me that there is insufficient signal strength on FM to ‘trigger’ stereo. That would explain the lack of stereo and the signal meter not working. However, on my normally strong stations the sound quality seems fine, not excessive background noise. I have to do more investigations today with a meter, but I’m wondering if that bodge job on the antenna panel is robbing the FM signal strength? And or screwing up the impedance match sufficiently to stop the set operating in stereo on FM?

Otherwise, the stereo decoder chip?

[orbanp1]

Hi Steve,

I take you have the service manual for your radio.

hifiengine.com has the schematics for the Huldra 12:
https://www.hifiengine.com/manual_li...uldra-12.shtml

manualslib.com has the service manual for the 2025:
https://www.manualslib.com/download/...g-Tr-2025.html

The service manual has detailed measurements for FM sensitivity and signal levels at the various stages.
Hopefully you have the test equipment to check those out, that will give you the answer why is the signal strength not satisfactory.

Good luck, Peter

[stevehertz]

Thanks Peter, I have those documents. Yep I do have a scope and various pieces of test gear, but frankly we're getting to the limits of my ability at that level ie checking FM sensitivity etc. I'm more of a 'prod it and check it' man I'm afraid, lots of use of my eyes - that kind of thing! Well, maybe that's doing me down a bit, but I have my limits as we all have I guess. Anyway, I'll look into that aspect, thanks.

[Radio Wrangler]

No, you don't have limits. There are plenty enough people on here will talk you through it.

Having a scope and a sig gen will get you going.

I've just downloaded the circuits from Electrotanya (they're nice generous people and don't demand membership) It looks to be a nice FM tuner.

Dual gate MOSFET RF
Dual gate MOSFET mixer,
THree ceramic filters with an IC buffer amp between them
TDA1047 doing most of the work.

So the TDA1047 datasheet will tell you what its sensitivity should be at 10.7MHz at its input. So that's the first place to check. If OK, you just inject signals in stages working back to the antenna.

What would help a lot is a good sig gen with a decent output attenuator.

David

[stevehertz]

I've been doing some more checks. These are my findings and observations.

* The set seems to lack both sensitivity and for sure, selectivity.
* The aforementioned signal strength meter does move, albeit an extremely small amount when tuned to a strong station.
* There's an obvious noise floor between stations.
* The voltage on pin 8 of the stereo decoder chip should vary (switch?) between (as best as I can read the values on the fuzzy schematic) 0.02V and 2.4V when stereo. However it is always at about 0.015V.
* I have substituted both the MC 1310 stereo decoder and the TDA 1047 FM IF amplifier +demodulator chips with no joy.

It all points to low gain at the front end doesn't it?

[roffe]

I've had a couple of Tandbergs with low sensitivity.
Turned out to be faulty ceramic filters.Easy to find which by bridging with say a 10nF cap across one filter at a time.

[stevehertz]

Talk me through this in more detail please. 'Ceramic filters' they're the three lead devices right? There's three of them in this set that I can see. What do I observe when the cap is used to bridge it? Where do I bridge the ceramic filter - end to end? end to middle? Thanks.

[Nuvistor]

If you find a faulty ceramic filter I seem to remember that they all need to be the same group. The groups used to be colour coded, to denote a slightly different centre frequency.
Not sure how much it matters but it was noted in service manuals or service updates.
Bridge the filter input to output with the capacitor.
Frank

['LIVEWIRE?']

Ceramic Filters were, as Frank says, grouped by colour code according to their frequency. which could be, IIRC, between 10.6 & 10.8Mhz in something like 250kHz steps. Obviously all three filters need to be the same frequency, unless otherwise specified by the manufacturer. Not sure if they're still available, but I may have a few in one of my spares boxes if needed. They're in Blaupunkt branded packs, as their car radios used them, but are the same types as used elsewhere.

[stevehertz]

Ok, so I bridge each ceramic filter in turn with a 10nF cap, and what should happen in the cases of one being faulty, and not faulty?

[Nuvistor]

Bridging the bad one should should result in a much more sensitive radio, there may be an increase across a good one but I would expect a big difference.
Frank

[Radio Wrangler]

If the signal strength indicator isn't showing much, then you've lost your signal a long way before the stereo decoder, so it most probably didn't need changing. But it is responding a little, that tells me that most of the signal path is operating, that the LO is oscillating and the mixer is mixing.

A cracked ceramic filter is a possibility. If so, you'll need a set of three matched ones. Rather than leap into trying to find them, let's find out whether you need them first.

You haven't said what sort of sig gen you have. Knowing would make it a lot easier to give guidance.

An old trick is to use a times-ten scope probe. connect its BNC plug onto the output of a signal generator. If it's a posh one set it up for 10.7MHz, with some FM applied and a level of maybe 10mV. If it's not a posh generator and hasn't got FM, the signal will still be noticed and ought to lift the signal strength meter. Having some audio to hear just makes things quicker if you don't have to keep eyeing the meter. If it's a very basic sig gen, then you may have to probe pin 18 of the TDA1047, and swing the generator tuning knob around a bit to find a response.

The scope probe will carry a signal in the reverse direction to its intended purpose. It will look like 10 megohms and a few pF to the point being prodded - a very gentle load on the circuit. The amount of signal injected won't be terribly precise, but it's a great way to find where the break in a signal path is..

Prod pin 18 on the TDA1047, you should get a strong response (after tuning the generator around a bit if necessary)

Then work your way backwards...

output pin of CF3
input pin of CF3
output pin of CF2
input pin of CF2

These places will be more selective, you may need to tune your generator around a bit again to find the peak

Now try the output of CF1It should be more sensitive here if U202 is working. You may need to start reducing the output of your sig gen. (The uA703 does not seem to have a good reputation for reliability)

Then probe the input of CF1.
Then the IF input to the board (+pin 3 of U201) It should be even more sensitive here.

Now we get onto the front-end proper.

Probe R121, either end to see if there is a 10.7Mhz path from the output end of the mixer (Q103)

To continue probing, you'll need your signal generator tuned into the 88-108 Mhz band. If your generator isn't well calibrated, set it against a spare radio. Tune the Huldra to the sae frequency while prodding gate 1 of Q103 or CR104 or the hot end of L104. You may need to tune around a bit to find your signal.

The scpe probe backwards trick isn't flat with respect to frequency, so you won't be able to gauge sensitivity between the input of the mixer and the output.

Now go backards to the drain of U102, then its gate 1.. and finally the input.


Somewhere along this sequence, you should find the sensitivity plunge. Then start looking around that area.

Once you've been through this process, you'll have a tool in your armoury for rapidly finding where to start investigating further. It can also save a lot of swapped parts and the risk of introducing new faults.

David

[stevehertz]

Many thanks David. I have a scope and I do have a (yet to be used by me) Taylor model 62A Mk2 AM/FM sig generator. I've done AM 'alignment work' before but never FM. I'll see how far I can get with your lovely detailed destructions and then holler for more help if that's ok.

BTW, the stereo decoder chip I swapped was out of another set and that still works with the Tandberg one in there, so nothing lost.

[Radio Wrangler]

With repairing sets, there is never an obvious optimum ploy. You can test the living daylights out of everything and eventually replace only the dud bit, but this can be inefficient in time and effort, though minimalist as far as parts are concerned.

At the other end of the spectrum you can test nothing and just start swapping things. If the parts are cheap, it's not terrible, but it eats up just as much time and effort. You save on the test gear, though.

The optimum ploy lies somewhere between these two. Certainly, quick tests that lead you to the area of the fault is a win.

Then there are two ways of testing.

In one case you put a signal in and try to follow it down the chain. The signals are in mst areas too small to see on a scope even if you have a 100MHz scope. Spectrum analysers will do the job, but they're in kilobuck territory. However, you could use another receiver. Pick up signals down the chain and feed them into the antenna socket of the test receiver. Once past the mixer, use a modified test receiver with an IF input connector added to it.

The other case is you work backwards up the chain. Shove some audio into the AF amp at the volume control. Did you hear it? Shove some modulated IF into the detector... ditto?

Just making the expected signal for the input to a stage and shoving some in. You use the remainder of the set as part of your test-gear.

There are many ways to do this, you bend the approach around what you have, and maybe you put some test gear on your grab-if-you-see-an-affordable-one list.

Certainly you can fix many faults with just a multimeter, but with a bit more test equipment you can see what's going on and probably learn more.

David

[stevehertz]

Thanks again David for your help and encouragement. Without going into too much detail, I am very busy in many ways at the moment - not just the five vintage items dismantled and part fixed in my workshop that need finishing off - but a huge garden including tree work, an ageing needy mom, indoor fixes, work packages, etc etc. So I'm thinking that I may put the lid back on it until I get more time as I have already stood in front of it for many hours now, testing and swapping parts. I'm dreaming about the darned thing. I'll get back to it for sure. Thanks for now everyone.