Schneider Melodie 58 - Project No 3

[Barnjet]

Replacement caps on order.

In meantime trying to unstick the variable air capacitor which has a tight spot at about 2/3 of tuning range. It is free either side of this point you can rotate the vanes across the whole range by pressing/pulling on them (gently!). But the cord slips at specifically one point.

I have tried a few drops of 3-in-1 on the bearings and lots and lots of "back and forward twiddling" but the sticking stubbornly remains.

I'm reluctant to undertake a full removal as I may make the whole thing useless. Any tips anyone?

[Barnjet]

Good news! I won't have to dismantle the air tuning cap after all. After a tad more 3-in-1 and further serious "twiddling", it is now free across the range.

The cap across the dual-ganged potentiometers is definitely 40nF (see phot)

There are 2 things I want to confirm with absolute certainty because they are safety related:

1. The circuit diagram shows (to me anyway) the Live is switched and the Neutral is fused (see phot).
2. I've searched everywhere for a fuse and I can't find anything that looks like a glass cartridge. Therefore... is this a 1.5A fuse? (see phot)

Grateful for confirmation...... or otherwise

[ms660]

Can't comment on the fuse except to say that the fuse in the plug should be 3A max.

If you want that capacitor to be in circuit make sure it's connected to mains Neutral.

If that was my receiver I would want to keep it in circuit unless it was causing a lot of interference or tripping an RCD, other's will have different views.

Lawrence.

[Gabe001]

Picture 3 is the voltage selector

If you can't find a fuse, use a 2a fuse in the plug as Lawrence has advised and stop worrying about it.

French 2 pin plugs (if I'm not mistaken) could be plugged in/wired either way, so live and neutral were theoretically interchangeable. I could be wrong though.

If you're wiring live to the on-off switch and neutral to the voltage selector and installing c24, you will need to move this cap to the neutral line as Lawrence says. C24 isn't essential for the working of the set, so if you aren't 100% confident just omit it; it's meant to help reduce mains borne RF interference but chances are you won't notice any difference.

Gabriel

[Herald1360]

That voltage selector is the fuse- IF it has a suitably fine piece of wire as its continuity element. It's a common arrangement on French sets of the time.

ISTR that a single strand of 0.2mm dia copper wire is about 5A fuse rating so pretty thin!

Just looked: https://electricguru.in/page_view.php?id=29

So 40SWG for 1.5A.

[Barnjet]

Originally Posted by Herald1360

That voltage selector is the fuse- IF it has a suitably fine piece of wire as its continuity element. It's a common arrangement on French sets of the time.

Thanks Chris, I can confirm 100% it's a fuse and a voltage selector. There is a very thin wire (<5 thou) running through the fibre strip between the 2 eyelets. (see phot). I will still use a 3-amp fuse in the plug, as I do routinely for all my radios. Good back-up and worth the reminders.


My current plan for wiring the mains 3-core and the Y-class filter cap C24 is (see phot)

Mains N to Term A which has 0 ohms continuity with the fused voltage selector
Mains L to Term C which has on-off switch in series with Term C and B.
Mains Earth to chassis

Cap C24 across Term A (Mains N) and Term D (continuity to chassis)

If you see any problems with this arrangement, then please shout.

[Barnjet]

When I say "If you see any problems with this arrangement, then please shout." I mean "anyone", not just Herald1360!!

Apologies, I don't seem to have an Edit button.

[Herald1360]

You only get 30 minutes for editing.

[Barnjet]

1st switch-on through the limiter and RCDI with mixed results

Good news - both dial lamps ok, no smoke, smells, bangs or pops. HT smoothing showing 227V and 168V. Anode voltages all good.

Not so good - no speaker crackles, just faint hum. Control grid voltages on output 6AQ5 started slightly -ve (-20mV) then slowly became slightly +ve (5mv). Similar story for control grids on ECH81 and EBF80.

Not sure where to go next, as I have previously confirmed output transformer and L/S as ok, and I believe I have replaced majority of coupling/de-coupling caps.

Some advice much appreciated

Max

[ms660]

Were those voltage measurements with or without the lamp limiter?

Lawrence.

[Barnjet]

They were measured with the lamp limiter on. I will measure them again tomorrow with the limiter off.

[Gabe001]

I am sure that the more experienced people will be able to advise you better than myself, but here are a few pointers. I'm assuming your only diagnostic tool is your multimeter - I've been there.
I don't have a circuit diagram to refer to so the advice below is generic.

1. Check if there is a switch that silences the speaker when an external LS is connected. Check that it's not "engaged" - bypass it if necessary.

2. Using an old mobile or iPod etc (something potentially expendible) and a plugged in headphone cable, which you will need to cut, you can try "injecting" some music at the volume control, or pick-up, or both. If the radio plays the music without issue your output stage if probably fine, and focus your fault finding efforts elsewhere

3. Have a good look at the waveband switch. Check that the contacts look ok. A spray with servisol and a gentle clean with a toothbrush are a good idea. I had a kbmr10 which didn't make a peep and the problem was a poor contact on the waveband switch.

4. Measure all the relevant voltages at the valve bases. If you don't have a service sheet, look up the working voltages of the valves online and compare your readings to those. Make a table including expected voltages, valve and pin number and post the results. You can sometimes work out the current consumption by measuring the voltage drop across the anode resistor.

5. Check the resistor values. It may not be possible to get accurate readings in circuit for all of them, but you should be able to identify if any one is open circuit for example. I "revived' 2 dead Philips sets by simply replacing a resistor in each.

6. Check that all valves are lighting up/getting warm

Finally, and I'm sure you know this, but grid bias is usually measured with reference to cathode and not to chassis (although there are exceptions)

Fault finding is the fun part. Best of luck. The most important tool in your arsenal is the ability to look at things in a logical way, and you can get quite far with a DMM alone.

Gabriel

[ms660]

Here's the schematic that's being used:

http://ekladata.com/2lF8ervfQQQAu85F6oO07yiHDgU.png

Lawrence.

[Gabe001]

Thx Lawrence. I note the caps rated at an unnecessary 3000v - same as the ones in my Phonola

[Barnjet]

Thanks all, some very helpful pointers. All measurements from now on are with the limiter off. Reference values are shown in brackets:

HT smoothing cap 1 = 269V (255V)
HT smoothing cap 2 = 196V (210V)

The current draw for the set at switch was was 147mA rising to 201mA at steady state, equating to about 48W which is consistent with the set's specification of 55W.

I have recorded all the pin voltages on the pin diagrams for each valve (except the TI, about which I am not concerned at the moment) Measured values are in red and reference values are in black, in brackets. Control grid voltages were measured with respect to the cathode i.e the red probe to the control grid pin and the black probe to the cathode pin. See phot 1

There could be an issue with EZ80 where the valve heater pins are at the same potential as the cathode. I did check the continuity of all the valve heaters (with the valves removed) and they were in the region of 2-4 ohms

Other than possibly EZ80, these measurements look reasonable.

[ms660]

Nothing too far out voltage wise but it's worth noting that the g2 voltages are low and the voltage drop across the HT filter resistor (2.2k) is a bit high, the reasons for those might be worth investigating, having said that there's nothing about those voltages that points to the receiver not working.

The HT voltage on the rectifier heater is normal as the heater is connected externally to the cathode.

The oscillator seems to be working (-10v on its grid)

So far as I can make out the voltages on the schematic were measured using a valve voltmeter (VTVM) which would normally have a very high input resistance and cause very little loading at the measuring point.

Lawrence.

[Gabe001]

Next step is try injecting an audio signal via pick-up (make sure the PU option is selected on the piano keys), and also at the volume control.

You can also double check whether the oscillator is running using another set, but it seems that it is.
Gabriel

[Barnjet]

Originally Posted by ms660

the voltage drop across the HT filter resistor (2.2k) is a bit high,

This resistor measures 2.4k in situ, so some of the drop can be attributed to that.

Originally Posted by ms660

So far as I can make out the voltages on the schematic were measured using a valve voltmeter (VTVM) which would normally have a very high input resistance and cause very little loading at the measuring point.

Aha, hence the reason that the method of measuring voltages was specified.

Originally Posted by Gabe001

Next step is try injecting an audio signal via pick-up (make sure the PU option is selected on the piano keys), and also at the volume control.

Will try this next.

Originally Posted by Gabe001

You can also double check whether the oscillator is running using another set, but it seems that it is.
Gabriel

I'll have to ask you how to do this, but after I've completed the audio tests.

[Barnjet]

I used a very old mobile phone as the audio source and separated one of the headphone leads (very fine wires - see photo) and brushed these across the PU terminals.

No audio signal heard, but some crackling and a buzzing sound. Hopefully, this is a good sign?

[Gabe001]

Radio on Pick-up setting
Croc clip those leads to the pick-up, signal to signal,ground to ground
Play a music file on your mobile, it should play through the radio speaker if your output phase is ok

Gabriel