HMV 1356 Aerial Circuitry

[M0TGX Terry]

Good afternoon everyone. I like to try to learn something every day, and today I've learned that I know very little about radio circuitry and how it works!I am repairing a 1948 HMV1356 table radio for a friend, which has been in the family since new but has been neglected in an outbuilding for many years. He'd been told by someone in the trade that "...it'll never work again....", which I foolishly took to be a challenge, so I said I'd have a look. Once out of the cabinet I found a couple of sections of the mains dropper were open circuit, so they were bridged by modern wire-wound resistors. This got the set working after a fashion but with very poor sensitivity, even with a length of wire in the external Ae socket. (It has an internal frame aerial, L1 on the circuit) A copy of what I hope is the relevant part of the circuit is attached to this post. I have replaced C1 (0.001),C2(0.0033), C4(0.1) and C7 (0.047 next to R5 at the bottom rhs) and this has considerably improved the set's sensitivity.I could find no trace of the existence of R1 in the set (cct shows it as 1Mohm) so I have left it off.
At last we come to the bit (one of many) that I don't understand. I am getting what I feel are odd results from attaching a short (approx 2 metre long) wire to the aerial socket and to various points on the circuit. The points marked Yellow, Blue and Green on the circuit diagram appear on the same piece of paxolin as the Aerial socket, and take the form of tags with a hollow fixing rivet which allows a wander plug to be inserted so I can use them as test points with my 2mtr "long wire". When the set is in its cabinet, only the Ae socket is accessible, the other points being totally covered by the rear cover.
I don't want to go in for a wholesale "change everything on sight" approach, so I would really appreciate any pointers as to where I should be looking for the cause of the results shown below. If the results suggest all is well, at least that would be something, but at the moment I am at a loss to understand what's going on.
Results of my tests:
1.A finger on the Ae socket improves the volume of the sound, but only slightly
2.A 2mtr wire in the Ae socket reduces the volume
3.Finger on the green socket improves volume
4. 2mtr wire in green socket improves volume just slightly more than 3 above
5. Finger on yellow socket slightly improves volume
6. 2mtr wire on yellow socket reduces the volume
7. finger on blue socket produces a bigger increase in volume than the above
8. 2 mtr wire in blue socket produces an even bigger volume increase than 7 above. I would say the volume achieved like this is more than adequate.
I get the same results whether the set is switched to Long or Medium wave.
I should add that where I live MW is full of noise from modern elctronics, but R%L and a couple of local independents are reasonably OK. R4 long wave is fine-ish but I'm not a cricket fan.
I hope I haven't bored everyone with the length of this post. Any ideas, suggestions, comments very welcome.

[ms660]

The RF might have been tweaked for frame antenna only, I would check the alignment with a sig gen via a standard dummy antenna.

Lawrence.

[Nuvistor]

Explanation of the coloured wires from the frame aerial and the LO, extract from the Trader sheet 917. The aerial if required should only connect to the aerial socket.
Has Lawrence states, check the RF alignment following the service information, if it will not align then fault find.

[G6Tanuki]

I've found that combining frame- or ferrite-rod and longwire antennas is often a bit hit-and-miss: to my mind the issue is that you have no way of controlling the phase-relationship between what the wire picks up and what the rod/frame picks up - there's no guarantee that these two separate signal-components will combine additively!

(Indeed, in receivers designed for direction-finding, a loop and a separate-but-coupled rod antenna are used because it can then provide a null to determine which of the two signal-peaks you'd normally get from just the loop is the one pointing at the transmitter).

You may have to decide which gives you the best results and then optimise that arrangement. Try it with a longer wire (50 feet or so was typically recommended as a MW/LW outdoor antenna in the post-WWII days)

[M0TGX Terry]

Thanks for all the replies. As an experiment, I've just taken the set upstairs in my house, in an attempt to get a better signal, and it has indeed done so! Maybe I should leave it "as is" for the moment and try it out back at its owner's house which is about 25 miles from here, and may have more signal than my part of rural Lincolnshire.
Once again, thanks for all the suggestions. As far as I can see, the RF cores /trimmers haven't been touched since it left the factory, judging by the amount of pretty vintage looking crud all over the adjusters. I'll leave them alone for now. I shall dig out a longer bit of wire and try G6Tanuki's suggestion (I suppose 2 metres of wire doesn't seem much at LW frequencies!)

[Herald1360]

There are quite a few more caps I would replace, or at least megger test, some of which could affect sensitivity such as screen decouplers.

I assume you're using Trader sheet references, in which case they're C8, C11, C12, C13, C20, C26, C27, C31, C32. Some should be X-rated C31 and C32; C1, C2 should be Y-rated.

This sounds a bit OTT but it's an interesting set being ac/dc and push pull output and should give a good account of itself in GWO.

[M0TGX Terry]

Thanks for that, Herald 1360. Mondays are taken up with looking after my 14 month old grandson, so nothing else gets done. I'll have a look at the rest of the re-capping 'to do' list tomorrow.

[cathoderay57]

Hi Terry. One other thing is simple to check but tricky to fix. I have found that frame aerials sometimes have a break in one or more of the turns. It is also all too easy to rupture the coil when removing the frame aerial from the cabinet by catching it on fixings. Check the continuity of each frame coil if you haven't done so already. To repair a break is very tedious as the broken turn(s) can be very hard to spot and require removal of the frame aerial and examination with a magnifying glass (if your eyesight is anything like mine). Finally I had a set that worked but was very deaf and hissy. I found that the wires to the MW and LW sections of the frame aerial had been transposed. Once corrected I found the performance was astonishingly improved such that there was little improvement when connecting a long-wire aerial, unless the set was rotated towards the null direction for the received signal. That set did have an RF amplifier however. Cheers, Jerry

[M0TGX Terry]

Thanks for that Jerry. This set only has one coil on the frame aerial (labelled as L1 on the diagram in my first post). The Trader sheet 917 gives the resistance as 1.4 Ohms. Mine measures 1.3 Ohms so it doesn't seem that is the problem. I've changed all the waxies, but not touched the electrolytics or the silver mica ones. It's better than it was, although still exhibiting the odd behaviour mentioned in post 1. On top of everything else, one of the dial lamps has now burned out. There are two of them in the set, and according to the Trader sheet each is supposed to be rated at 5.2v, 0.15 amps. I have nothing like that, and can't find any on line so I have temporarily stuck in a 2.5 v torch bulb, which to my surprise is actually dimmer than the original.
My inclination now is to put it back into its cabinet and return it to the owner. At least it works now, although not as well as I think it should, and it looks a good deal tidier than when he brought it to me. Sometimes "enough is enough". There's only so many hours in a day, and for a set that might be worth £30 on a good day, I think I've given it my best shot.

[Mike. Watterson]

Series. So 6.3V 150mA is better than 2.5V 300mA, for instance.

[M0TGX Terry]

Thanks Mike. I'll see if I can find a couple of suitable lamps.
While I'm here, can anyone advise me as to what voltage rating the 20microfarad electrolytic caps on the cathodes of the output valves should be. The Trader sheet just gives the capacitance and the existing caps don't appear to be marked.One of them is getting noticeably warm.
As always, any advice will be very welcome.

[Mike. Watterson]

In general on Cathode decoupling caps:
A 50V rating would be plenty. It's basically the grid bias, so depends on O/P valve type. Often 25V is fine.

[Mr 1936]

Hi Terry. You mention Aerials, but do you have a decent Earth connection as well ? If not, you are relying on chance earthing via capacitance , or via the mains connection. As others have said, when these sets were new an outdoor aerial wire of tens of feet was the norm. It was typically a vertical section, plus a horizontal section which adds capacitance and increases signal pickup. If you try this, keep it all as far as possible from the noisy house mains wiring. As the RF level is steadily increased, to begin with the audio level will increase. However, once the AGC begins to operate, further increases in RF level will not cause much increase in audio. Ideally your aerial should deliver enough signal from your favourite station to bring the AGC into play. A couple of metres is unlikely to be enough, and its signal will be "fighting" with the pick up from the frame aerial.

[M0TGX Terry]

This set is a live chassis AC/DC one with a 2 core mains cable and no provision for an earth connection. Today's plan includes rigging up a much longer temporary wire antenna to see if that helps. It had never occurred to me that a short wire could actually make things worse. Just shows how very little I really know.
Thanks Mike for the pointer on voltages for the electrolytics. Now to try to find some small enough in diameter to fit the under chassis clips. The originals approx 10mm dia by 36 mm long, axial types with a paper sleeve which insulates the body from the chassis clip. There's no room for anything bigger. Hey ho.... onward and (hopefully) upward. Perhaps I won't throw in the towel quite yet.

[Herald1360]

If those clips are for the cathode decoupling electrolytics I'd think your problem would be finding modern parts large enough! Try going way up on voltage rating or restuff the old tubes. You might even find a non electrolytic axial cap that would fit.

[M0TGX Terry]

This afternoon I strung up about 65 feet of wire as a temporary aerial, and it dramatically increased the number and strength of receivable stations - can't wait to see how it is after dark!
Of course, the amount of interference also increased - a fact of MW listening life around here.
Oddly, all the weird results re the point to which the aerial is connected still apply, even with the 65 foot wire.
The push pull output valve both have a 20 microfarad cathode capacitor, and I noticed that one of them gets noticeably warm in use. Once I find a suitable replacement I'll swap it and see if that makes a difference. In the meantime, many thanks to all those who have taken the trouble to reply to this thread. It's good to know that I'm not alone.

[Mike. Watterson]

Cut them both off. Risk of over-running a valve otherwise. It will only reduce volume a bit and increase quality due to negative feedback.

[M0TGX Terry]

I'll cut 'em out in the morning. As ever, I am astonished at the knowledge amongst this forum's members, and by their generosity in sharing it with others.

[M0TGX Terry]

Well ...... following Mike's suggestion I've just snipped the two cathode electrolytics out of circuit. The sound quality has improved markedly, with no discernable loss of volume. Will it be Ok to leave them out permanently? I am currently listening to Absolute Radio on MW at a louder than I want it volume using my 2 mtr long bit of wire. Without the wire it's usably loud even in my very poor reception location. The 65 foot temporary wire I threw up yesterday doesn't make a significant difference to the signal but certainly works efficiently as an interference collector.
As always, many many thanks to everyone who has helped with advice on this set.

[Mike. Watterson]

It will have more gain with them replaced. That's why they are usually fitted as the makers / designers regarded gain more important than local negative feedback to improve linearity.

If you are not worried about originality, it's always safe to leave off output stage cathode decoupling caps. If you look at various circuits at RF & IF, you'll see they are small values to only increase RF or IF gain. They should not be left off. They are never electrolytic.