Grundig 80U

[ppppenguin]

This example uses numbers from the USA but the principles are the same here: http://www.bristolwatch.com/sr/buck_boost.htm

[Amie]

Ah. Got it!

Primary across the mains, secondary in series with it to the set

Thanks guys.
Amie

[ppppenguin]

Amie, not only do you learn fast, you also have the benefit of your late father's collection of testgear and other stuff. You'll be teaching us things soon!

If you've got a big "continental" set, Grundig or suchlike with lots of piano key controls, its mains TX may well run hot on our mains. A buck TX is an ideal way to sort that out. To buck 20V, that's 1/12 of 240V so the transfomer rating need only be 1/12 of the power consumption of the set. In other words tiny for for most sets.

The subject of isolating transformers and RCDs have been worked over heavily both here and elsehwere. Look for Mark Hennessey's contributions on the subject for a really clear exposition. If you stop to work out where and when currents can flow through your body then you'll be a long way towards staying safe. Draw out the circuits, with you as part of those circuits. Most of us here have made mistakes, sometimes rather nasty mistakes, and we know what a mains shock is like. I've had worse than that - about 1200V DC - and it £$%^&* hurt. You will make mistakes and with a very modest amount of luck you'll survive and learn the lessons. Just like we have.

[Terry]

Of course, all of this discussion assumes that Amie really does have a 240V supply!

Towns are usually pretty close to this but, in more remote areas, there can be significant variations, both up and down!

[Mike Watterson]

Primary across the mains, secondary  in series with it to the set

But measure voltage BEFORE connection!
If it's 260V instead of 220V you swap EITHER the 20V connections OR the 240V connections on the small transformer.

If you've got a big "continental" set, Grundig or suchlike with lots of piano key controls, its mains TX may well run hot on our mains. A buck TX is an ideal way to sort that out. To buck 20V, that's 1/12 of 240V so the transfomer rating need only be 1/12 of the power consumption of the set. In other words tiny for for most sets.

Unless they are push pull (rarer) or have a stereo decoder (very rare, really only export to USA models in Valve era) they are all about 40W to 45W max. Irrespective of how many buttons, speakers, size of cabinet or number of bands.
My "everyday" German model is about 38W as it has EL95 rather than more common EL84 and only one scale lamp.
https://www.radiomuseum.org/r/aeg_bimby_luxus_1461.html
Perhaps the big motorised one I gave away was 55W max, it had 2 x EL84 for single ended stereo from tape/Gram and push pull mono on radio.
Some German and French models do have voltage settings.

[Amie]

Jeffrey.
Dad kind of brow beat me into learning electrical safety. He said a shock is almost inevitable, but staying safe by not receiving a shock accros the chest or one that makes the hand grasp the source would be less serious, but that most people die from secondary mechanical injury, falls, hitting the head, etc. My memorable belt is from a Pye TV that still sits up the corner, I took the eht cap off the tube long after switch off, just to look. I learned about  how a tube is its own smoothing capacitor.

Terry.
Mains here is usually 235/245v

Mike.
Check first, OK, got it. Guess the only way is to try, there's no way of knowing phase?
Jeffreys suggestion may be better. Transformer in a box in line with the mains lead, no mods inside the set?

Just one more thing to try on the 80U is see if the AF to the triode grid is distorted or if that section of the valve is causing it.
The set also won't work on FM......

Thanks all,
Amie.

[ppppenguin]

Amie, sounds like you've had your electrical safety apprenticeship

Totally agree that mechanical hazard is usually worse than a brief mains shock.

[Terry]

Yes, I was once putting my hand under the CRT of a Pye VT4 (probably to reach a valve at the front) when the dusty scan coil harness tickled the hairs on the back of my hand. My hand came out so fast that I had some very deep cuts where my hand had caught on the tag strip that the harness was connected to. Your brain doesn't have time to work out what caused the sensation!

Fortunately they missed the veins but the cuts took quite a time to heal.

As for your EHT 'experiment', I once swung a chassis round on the bench on a Monday morning - it had last been powered up the previous Saturday - and was surprised when I touched the CRT top cap by accident. It still packed quite a punch!

[Amie]

Hey!
That 200K resistor I replaced shouldn't have failed!

Thinking about it, even with a dead short coupling capacitor, and the pentode section passing grid current, (it glowed blue briefly until I switched off!), I tested the old 200K from the bin again, it wasn't completely open but near 700K, the first new one WAS dead though. Red black black orange space red? 200K 2% right? (From an RS development kit) tested a new one, 200K, OK, tack that in. Second replacement measured 17.5K....

Hmm, one failed, one wrong value. Checked the price of that kit on RS...not what you would pay unless you are expecting a quality product! Possible that it was near its maximum working voltage with the dead capacitor maybe?

Never trust new parts either?

Hey presto! Good volume and clear sound. No need to get a new valve just yet.

First question. I've to test why the FM band won't work now, the EBF80 is working on the MW. IF, so....is it reasonably safe to assume it's working on the FM. IF?

Second question, do I need to test the FM. IF with the SIG generator set at 10.7MHz with modulation?

Third question. If Q2 answer is yes, what deviation should I set to modulate the 10.7MHz test signal, (is it the same as the transmitted deviation, or is it a proportion of it, never thought about that before?)

Enough electronics and creative writing for now!

Thanks,

Amie.

[Old Sparky]

Regarding alignment, I'd suggest injecting 10.7Mhz at about 30-40% deviation as a start. The transmitted deviation will vary with the program content, so a setting around the estimated average level will suffice.