FM line-up (Dynatron Atlantis)

[Mark Hennessy]

Hello All,

This might be one of those embarrassingly simple questions

I've had a Dynatron Atlantis on the bench for a few days now. OK, I've had a lot of other stuff going on too, and it's possible that I've not been thinking 100% clearly, but things are getting back to normal now (and I will respond to some PMs soon!)

This set had myriad problems, but one by one they've all been exorcised. I'm left with strange behaviour on FM; as you tune to a station, rather than tuning through the usual strong peak at the centre of the range, the set actually mutes in the centre. Tuning off-centre gives audio, albeit rather weak and a little distorted...

To me, this sounds like a problem with the discriminator. I've checked all the components in the area, and everything seems fine. There is healthy amounts of 10.7MHz going in to either diode, and a DC potential on the "output" of each corresponding to what is going in. One observation, however, is that the AFC drive voltage - the DC offset on the recovered audio - doesn't go negative as you tune through a station.

The next step is to attempt a line-up, which is what I plan to do after tea. Unfortunately, I've been chasing other problems with my test gear, but I think I'm good to go. But before I dive in, am I missing something obvious? That said, it's fair to say that someone has had a good go at this set, so perhaps all the FM IF cores have been randomly twiddled...

I do have a couple of these sets myself, and they are awaiting attention. But, I have dragged them downstairs because I needed a reference to compare the patient to when reassembling the FM tuner (these things are really unpleasant to work on - nothing like a Hacker!). I did note that the FM discriminator diodes have been disturbed in my sets as well - it's almost like this is a common problem...

I've never had to disturb FM IF sections of sets in the past, RF alignment, certainly, and AM (RF and IF) too. I've got a Marconi TF2008 that can be swept, so will set up for a visual alignment in a bit, but very interested in the experience and wisdom of others in the meantime...

Thanks in advance,

Mark

[Steve_P]

Do a line up and make sure that the IF is exactly on 10.7 Mhz. Change the diodes and check that the resistors are of the same correct value.

Cheers,

Steve P.

[Mark Hennessy]

Hi Steve,

I've checked both diodes for Vf and reverse-leakage, but will check them by substitution if all else fails (I'll need to borrow a pair from a working set, as I don't have many germanium diodes in stock). As for the resistors, I did find that RV1 had changed value (track measured 3K6 rather than 1K), so replaced that. I forgot to mention that by turning this control, I can make it null out perfectly - I presume that I'm balancing the detector for maximum AM rejection at this stage?

I also changed C66 and C69, which made no difference. The AGC capacitor had high ESR, and replacing that improved AM, and as it's the same type, I suppose I ought to check C56...

Will attempt line-up once the kids are in bed.

Cheers,

Mark

[Mark Hennessy]

Well, finally made some progress...

I tried new diodes - no change. I tried a few other things, then it seemed to start working. I wasn't happy, because I wasn't sure about what had changed. I want to see cause and effect in action!

So, did the complete FM lineup - IF and RF, and find that things were fairly far off, but soon got it working reasonably well. But, it was insensitive, and the tuning meter only reaches about 25% deflection... Definitely not right. But then it started sulking again, and we were back to the start.

I checked/changed lots of components, but nothing really helped. So, that only left the IFT. And that was where the fault was. However, I can see why I was so confused...

After a careful visual examination under a x7 eyeglass failed to reveal anything obviously amiss, I checked the continuity of the various coils and found an O/C. By carefully scraping away enamel, I was able to prove that thankfully the O/C was at the joint, rather than within the coil itself (which I guessed would be unlikely - the wire wasn't all that fine). The bad joint was at pin 8, where the centre-tap of the secondary is picked up, and the O/C section was to the winding that terminated at either pin 4 or 7 (can't remember which). The joint looked fine, but I was able to unwind a small section of wire and tin that, just to be sure, although I'm sure that the good sweating I gave it would have burnt of the rogue enamel buried deep inside.

But, the reason this was confusing was that when in fault mode, there was a healthy sine wave present on both pins 4 and 7 (and a correspondingly good DC potential on C66 and C69). But of course, this was coming via C58, inside the can. At 10MHz, 68pF has a pretty low reactance (around 200 ohms). Of course, if I'd put a 'scope across pins 4 and 7 simultaneously, I would have seen that the sine waves were in phase - and annoyingly I did consider doing this, but ruled it out because it would have meant going upstairs to get a better 'scope; my DM63 is only 15MHz officially, and the triggering isn't brilliant at 10MHz.

Now, the set seems a little more sensitive, and the signal strength meter reaches 75% deflection with strong signals fed in from the FM dist amp. I guess that means that when it working before, it wasn't a zero-ohm connection.

As mentioned, my examples have all been worked on in the discriminator area - in fact one of them has had R53/54 shorted to increase the meter reading. I wonder if this is a common fault? I suppose that now they're all in pieces, I should spend some time fixing them before I forget everything I've leant about them...

Hope this helps someone in the future!

Mark