FM bandspread reduced?

[Robsradio]

Evening,
I have a set I'm tinkering with that covers the FM band 88-108 on the dial, but in reality I have R2 at 87.5, usually 88.5, and I have Capital 103.2 sitting at just over 108 on the scale.
This must have been OK new, but any thoughts on how to bring it back into line?
I’ve shifted the FM osc on previous sets to just be able to receive 105-106 on a dial only going to 104, but not sure how to tackle this one as it’s the spread and not the offset.
Any help, suggestions appreciated.
Thanks
Rob

[Graham G3ZVT]

When doing alignment, a typical oscillator circuit will have (at least) two ways to alter the frequency. An inductor, and a trimming capacitor.
The procedure would typically be
A) Tune the dial and the sig gen to 88MHz then adjust the coil for a signal,
B) Tune the dial and the SG to 108MHz and adjust only the trimmer for the signal, now go back to A, then B, rinse & Repeat until no further improvement can be made.

So it's C at the low end and L at the high end, the same applies when tracking the RF and mixer stages if you're doing a full alignment.

Please let us know the make and model and we can give you more specific instructions

[Robsradio]

Hello
Thanks for your message, it is a Vega Selena 215.
Thank you
Rob

[TonyDuell]

Interesting....

Let's assume the IF is 10.7MHz (it nearly always is in such FM sets).

If the oscillator is supposed to be 'high' -- that is above the signal, then for 88-108MHz, the oscillator should go from 98.7 to 118.7/MHz. A ratio of 1.20 or so.

But suppose somebody has mis-aligned it so the oscillator is 'low' and got it right at the 88MHz end, that is with the oscillator at 77.3MHz. Now if the ratio from high to low remains about the same, the high end would correspond to a signal frequency of a bit over 103MHz.

Coincidence?

It'd be worth checking the oscillator is the right 'side' of the signal.

[Robsradio]

Originally Posted by TonyDuell

Interesting....

Let's assume the IF is 10.7MHz (it nearly always is in such FM sets).

If the oscillator is supposed to be 'high' -- that is above the signal, then for 88-108MHz, the oscillator should go from 98.7 to 118.7/MHz. A ratio of 1.20 or so.

But suppose somebody has mis-aligned it so the oscillator is 'low' and got it right at the 88MHz end, that is with the oscillator at 77.3MHz. Now if the ratio from high to low remains about the same, the high end would correspond to a signal frequency of a bit over 103MHz.

Coincidence?

It'd be worth checking the oscillator is the right 'side' of the signal.

Hmm, that’s food for thought, i will certainly take a look as that would account for exactly what it is doing.
Thank you
Rob

[Robsradio]

Originally Posted by TonyDuell

Interesting....

Let's assume the IF is 10.7MHz (it nearly always is in such FM sets).

If the oscillator is supposed to be 'high' -- that is above the signal, then for 88-108MHz, the oscillator should go from 98.7 to 118.7/MHz. A ratio of 1.20 or so.

But suppose somebody has mis-aligned it so the oscillator is 'low' and got it right at the 88MHz end, that is with the oscillator at 77.3MHz. Now if the ratio from high to low remains about the same, the high end would correspond to a signal frequency of a bit over 103MHz.

Coincidence?

It'd be worth checking the oscillator is the right 'side' of the signal.

Hello
Sadly this is not the case, the oscillator is high and pretty much spot on frequency, thanks for your suggestion, it sounded as if that was going to be the problem.
Thanks
Rob

[paulsherwin]

It was a plausible suggestion, as the quality control for these sets was very variable.

As explained in #2, the tuning range is determined by the interaction of the oscillator coil and the tuning, trimmer and possibly padder capacitors. It's quite possible that this wasn't set up properly in the factory.

[Graham G3ZVT]

I looked at the circuit of the VHF tuner, and it seems there is no trimmer, they are all fixed in that area.

Then I found this video.
https://www.youtube.com/watch?v=jABh9PyhyXQ
The TL; DR is basically he improved the scale calibration and linearity by carefully bending the veins of the tuning capacitor.
I think what I might have tried, is replacing C18 with a 150 or 200pF trimmer, and doing the alignment procedure I described earlier.
Incidentally, the varicap diode is there to apply AFC, I expect he knew that, but he didn't explain it.

[Robsradio]

Originally Posted by Graham G3ZVT

I looked at the circuit of the VHF tuner, and it seems there is no trimmer, they are all fixed in that area.
Attachment 321407
Then I found this video.
https://www.youtube.com/watch?v=jABh9PyhyXQ
The TL; DR is basically he improved the scale calibration and linearity by carefully bending the veins of the tuning capacitor.
I think what I might have tried, is replacing C18 with a 150 or 200pF trimmer, and doing the alignment procedure I described earlier.
Incidentally, the varicap diode is there to apply AFC, I expect he knew that, but he didn't explain it.

I watched the video, seems a similar condition to mine.
I tried adjusting the vanes of the tuner gang, didnt make any difference on mine.
I have decided to lose Radio 2 and gain a couple of stations at the top of the band that i listen to regularly by adjusting the oscillator, seems ok, but scale calibration is out as expected, I'll live with that.
Mine is a very late set and bears no resemblance to any diagrams I have.
Thanks for help offered by all.
Rob

[paulsherwin]

The various B21x models are notorious for undocumented circuit changes during the production runs.

[Robsradio]

Originally Posted by paulsherwin

The various B21x models are notorious for undocumented circuit changes during the production runs.

The PCBs only have a fraction of the component count of the earlier models, it’s also the version without a signal meter, but still has the legend on the metal front.
This is marked ‘made in Belarus’ as opposed to the earlier sets being marked as ‘made in the USSR’.

[Mr 1936]

Hi

I'm going to start by assuming that it's just a case of the tuning range being "out", and not something odd like picking up image frequencies.

A typical set will have a bit of overrange, and will tune from 87.5 to 108.5 MHz. With a high LO that's an LO range of 98.2 to 118.2 which is a ratio of 1.21 The max to min capacitance ratio will be the square of that, i.e. 1.47. Putting some typical values on, that would be from 22 to 32 pF

From your description, your set seems to have a restricted range, something like 88 to 103 MHz. That implies a max to min capacitance range of only 1.34 e.g. 30 to 40 pF

The cure is normally to reduce the fixed capacitance in the LO tuned circuit to get the top end of the range right. This will throw out the bottom end to a smaller extent, so you then need to increase the inductance. These two adjustments interact to some extent (more than for an AM radio), but it is feasible.

In the absence of a convenient LO trimmer cap and adjustable inductor, you may be forced to get out the soldering iron to reduce the fixed capacitance. Based on my example above, you need to "lose" about 8 pF. The obvious first target is the 4p7 capacitor. I estimate the inductance increase at about 25%, which can be achieved by squashing turns together or even adding an extra turn or so.

[John_BS]

Assuming the radio once worked correctly, I'd be inclined to look at the AFC varactor. They generally exhibit high capacitance at very low voltages, so it might be worth checking the radio with the afc set to a high voltage, or with C23 lifted if the varactor has failed- s/c ?

[Robsradio]

Thank you for the suggestions, i will look into it when i next have some spare time.
Thanks
R