Sansui TU-X1 tuner very poor on 75 ohm input

[stevehertz]

Good point Martin and may well be the case here.

[Synchrodyne]

Originally Posted by stevehertz

For now I have put the isolating trannie to one side and made connections directly to one side of the 300 ohm input and earth (the 300 ohm trannie centre tap). Interestingly, each side of the 300 ohm trannie gives different results. The side that is also connected to F7 (the FM/AM hybrid antenna input socket) gives poor results. This is the side that is also used when the isolating trannie is used - as per normal. The other, 'free' side gives good results, akin to when I use a 300 ohm feeder/balun into the 300 ohm input. That suggests two things..

1) That Sansui 'chose' to connect the 75 ohm input to that (F7) side of the trannie even though intrinsically it was not the best choice - a design error basically.

2) That there is a loading issue with the F7 socket/circuit. Previously I have cleaned and double cleaned and lubricated the plunger switch associated with F7 and it appears to be working ok as it connects and disconnects the signal coming in from the normal FM sockets.

I had wondered about the symmetry – or lack thereof - of the 300 ohm input transformer primary.

The 75 ohm input is permanently connected to the top half of that primary. So a fault (perhaps a ground fault) somewhere in the 75 ohm input assembly/transformer might cause the observed symptom. However, given that the fault appears to have remained after the physical removal of the 75 ohm transformer, then the latter might not be the cause.

The FA-7 input is connected to the top half of the 300 ohm primary only when the FA-7 switch is closed by the fitting of the appropriate plug into the FA-7 socket, at least barring a fault in the switch itself. So a fault in the FA-7 input network and transformer would cause the observed 300 ohm input asymmetry only when the switch is closed. With nothing plugged into the FA-7 socket, the network/transformer cannot load the 300 ohm input.

I think that leaves at least the possibility that the fault might lie in the 300 ohm transformer primary, specifically its upper half. Perhaps the continuity of that half-primary could be checked?


Cheers,

[stevehertz]

Good analysis, thanks I'll look into that. Also, despite my repeated cleaning attempts, it could be that the F7 switch is sticking in some way.

[stevehertz]

Sure enough, there is an open circuit between one of the 300 ohm input terminals and earth; the centre tap of the trannie. Now that does mean that the trannie itself is O/C as there are connections and solder joints along the way and getting at that trannie is going to be difficult to say the least. Anyway, I'm on to it..

[stevehertz]

Measuring from each 300 ohm antenna input to ground exposed that one of them failed give the expected low resistance reading, in fact it measured infinite resistance, suggesting that one half of the 300 ohm input was open circuit. However, there’s quite a few connections between the transformer and the antenna input sockets so it did not necessarily mean that the trannie itself was O/C. The 300 ohm transformer is a PCB mounted device and in fact one of its input wires wasn’t protruding sufficiently through the PCB enough to be soldered, hence the open circuit reading. On examination, of the six PCB pad connections, one of them displayed just a smoothish blob of solder with no wire protruding. Removal of this solder blob exposed a wire that was actually lower than the surface of the PCB tracks, hence it wasn’t soldered, O/C. I managed to get more of the wire to poke through the hole in the PCB pad, and resoldered it. I am still in the process of making a replacement panel for the 75 ohm input socket and 75 ohm matching trannie (the original panel is resistive), but nonetheless, connecting my 75 ohm downlead and antenna now gives good results. I should be able to complete the re-wiring in a day or so and will report back on completion. But anyway, that's the fault, an unsoldered 300 ohm input trannie connection.

[Graham G3ZVT]

That's good news, and the fault was misleading as the 300 ohm terminals appeared to work better than the 75.

If it was me, I would put the board back as it was to preserve originality, even if you can't clear the 600K resistance. That order of leakage will have no detrimental effect on that part of the circuit. Think of it as a static discharge path for your aerial.

That's a beast of a receiver you've got there, and probably quite valuable.

[cathoderay57]

Good example of forensic fault diagnosis - nice work and well done. Cheers, Jerry

[stevehertz]

I have now remade the 75 ohm input socket mounting panel from non-conducting paxolin and put everything back together. Not an easy job as the components involved were situated in a narrow gap between the rear panel and a secondary inner chassis. So it called for a lot of lateral thinking and improvisation to get at nuts, bolts, and soldered joints.

Given the findings of others and also noted in the Tuner Information Centre for vintage tuners regarding this tuner’s input anomalies, I thought it would be worthwhile taking a few signal strength readings under various situations. The signal strength meter is calibrated in dBf and those are the figure I am referring to. Plugged into the 75 ohm input socket and tuned to a decent strength station, the reading was 49. If I then connected the uppermost 300 ohm input terminal ‘01’ (as designated on the schematic) to ground, the signal level increased to 59 – substantial. This is the same result as is mentioned in the TIC write up of the TU-X1. Why is this? Well, with one side of the 300 ohm trannie connected to ground, it provides a better match to the 75 ohm input trannie that I assume is a 1:1 ratio. It’s not quite that straightforward though as this does not tally with the schematic diagram. The schematic shows terminal 01 as being the side of the 300 ohm trannie that is connected directly to the ‘non-ground side’ of the 75 ohm trannie secondary. But it can’t be, because if that point is connected to ground, the signal would be shorted, not boosted?! Here’s the answer. Sansui actually connect the 75 ohm secondary to 02, not 01. So the wiring connections to 01 and 02 are physically swapped compared to the schematic. So, confirming what others have said, when using the 75 ohm input, link the uppermost 300 ohm terminal to ground, but omit or remove this link when using the 300 ohm input itself.

So, not only is the 75 ohm input depicted badly on the schematic, suggesting an open circuit that needs to be linked for the circuit to work, but the 300 ohm input connections are also incorrect. A couple of years ago I worked on a Sansui 9090 receiver and the schematic for that too was very erroneous. So there you go, don’t trust Sansui service data.

[Radio Wrangler]

A successful outcome. A properly working tuner and understanding of what had gone wrong. The only remaining mystery is why Sansui made a bit of a pig's ear of something quite commonly done without problems.

[stevehertz]

Well, a mystery only in so far as 'you wouldn't expect it to be that way'. But experience tells us that Sansui did have a nasty habit of making schematic errors at that time. Strange (a mystery?!) given that their design and build capability was second to none. They obviously needed a tougher boss of the drawing office!