Premier Midget FM Oddity

[cathoderay57]

Afternoon folks. I have just acquired a couple of Premier Midgets; one in a wooden case looks pretty much original with a standard octal valve lineup for this set. The other one is an oddity. The white Bakelite case looks almost certainly like a Premier Midget, although I've never seen one with the black plastic carrying handle so that is maybe an addition. The innards comprise an FM-only set with an on-off toggle switch at the rear of the chassis and the centre cabinet hole with a blank fitted. There is a metal rectifier (different to the type used in the Midgets that I've seen) fed directly from the mains, a heater transformer, and the valve line-up is 4 x EF91 with an ECL82 output valve. The chassis has the look of a kit about it; probably not commercial production but not home-made either, although the dial looks home-made from a thin paxolin sheet. I didn't think Premier offered an FM kit?? Anybody seen anything like it before? Cheers, Jerry

[cathoderay57]

I'm going to see if I can get any useful signals out of this FM oddity just for the fun of it. I'll probably end up transplanting the standard AM chassis from its wooden cabinet into the white Bakelite one. The wooden cabinet is OK-ish bit will need refinishing. I'll post the FM circuit later when I've figured it out. J

[kalee20]

Looking forward to it! Seems a strange line-up, waiting to see if it's a conventional superhet or something else!

[Sideband]

I wonder if it's a 'Jason' kit that's been built with an output stage and then built into a cabinet as a self contained radio.? The layout does bear some resemblance to the Jason kit using four EF91's.

[cathoderay57]

Thanks for the comments. Here's the circuit sketch. I've assumed that there are fixed capacitors across all of the IFT coils but I didn't open the cans to verify. I'll add a few notes concerning what I do with it. Cheers, J.

[TonyDuell]

This is not really my field, but shouldn't there be a chassis ground connection to some point in the discriminator circuit? Otherwise there doesn't seem to be a complete circuit for the audio signal/volume control.

[cathoderay57]

Top marks to Sideband. The circuit is very close indeed to that of the Jason FMT1; thanks for the tip J.

[Gabe001]

Does it work jerry?

[cathoderay57]

Yes Tony, looks like I missed a couple of resistors across the diode pair whose centre point is common and grounded to chassis. I did my best but nobody's perfect! No point correcting it seeing as the official circuit for the Jason FMT1 is the same. Cheers, Jerry

[cathoderay57]

Hi Gabriel, not tried it yet. I need to check out the metal rectifier, test the mains and OP transformers, and change a couple of capacitors first. Since the FMT1 was, I think, designed for local reception I'm not expecting spectacular results here, although we aren't far from the Wenvoe FM transmitter. J

[cathoderay57]

I've attached a snip of the ratio detector circuit showing the 2 resistors that I missed (R13 & R14). I was quite surprised to see R16 though. This is a 470k connected between the HT line and the ratio detector. It is missing from my circuit. Would someone please explain what is its purpose? J

[cathoderay57]

OK, I've answered my own question (by reading the Jason manual). The 470k is for muting inter-station noise. J

[kalee20]

Thanks for the circuit!

It's a reasonably conventional superhet with an RF amplifier then.

The AVC is curious, grid rectification by V3 which is then applied to the RF amplifier (both of which are sharp cutoff valves).

And the ratio detector is incomplete - there's no return path (from the midpoint of the 8uF capacitor?) to chassis... nor is there any injection of the primary-winding voltage to the secondary centre-tap, so no phase variation with frequency that gets converted to an amplitude variation. Maybe something you just missed!

[G8UWM-MildMartin]

The metal toggle switch mounted on a metal chassis connected to one of the unmarked mains leads is, erm, remarkable!

[Sideband]

Originally Posted by cathoderay57

Top marks to Sideband. The circuit is very close indeed to that of the Jason FMT1; thanks for the tip J.

I have an FMT1 and recognised the basic layout. It looks like someone has made a custom chassis using the same basic layout as the Jason and then extended it to take a power supply and output stage. It should work quite well although it could do with another IF stage! My Jason works well enough for R2,3 and 4 plus a couple of locals but it really struggles with selectivity at the 'top end' of FM where stations are only 400khz apart. Having said that, it sounds really good through the Leak amplifier when I tried it. I had to re-align it when I first got it....clearly it hadn't been near a signal generator!

[TonyDuell]

Originally Posted by cathoderay57

Yes Tony, looks like I missed a couple of resistors across the diode pair whose centre point is common and grounded to chassis. I did my best but nobody's perfect! No point correcting it seeing as the official circuit for the Jason FMT1 is the same. Cheers, Jerry

I apologise if my message came across as a flame, that was certainly not my intention. And I don't want to know the chap who never makes errors as he never does anything.

Having done some reverse-engineering in my time I know only to well how difficult it can be and how easy it is to miss a component, etc.

That said, if I see a circuit here, I try to understand it. And if something seems wrong, I will comment. It's quite possible I've misunderstood things, of course.

[cathoderay57]

No offence taken Tony, thanks for picking it up!

[cathoderay57]

Thanks Kalee, yes, you're correct that I also missed the small extra winding providing the phase injection into the ratio detector. This is clear in the snip attachment in Post #11. J

[cathoderay57]

The single pole toggle switch was maybe a bodge by a later fixer; reason being the 500k volume control is switched but one set of contacts is o/c as so often happens. Whoever did the original build appears to have been cognisant of the live chassis risk since both poles of the aerial socket have 1000pF isolating capacitors, albeit not Y-class. Insulation from the metal control spindles being provided by the plastic knobs. J

[cathoderay57]

Using an Avo on the x10k range the forward resistance of the metal rectifier measures 70k and reverse 750k. I've not much experience of these devices so is that reasonable? Since it is connected directly to the mains I could foresee a very loud bang if it fails and puts mains onto the capacitor... I might try it but I'll insert an in-line 100mA fuse. Or I could just use a 1N4007 and a 100R series resistor. Thoughts? J