Tape Echo Bias Oscillator questions

[snk]

Hello,
I recently got an obscure tape echo (from an unknown brand, without any schematic... I suspect it from being built in very small batch in a boutique a couple of decades ago) which needs servicing/repair.
I cleaned it (and demagnetized the signal path), the motor runs fine, and I can hear the dry signal coming through. Nice 
I made a tape loop, and I can hear some sound from it, but it's very quiet, and I think that what I hear is the sound which was already on the tape, not the echo. It could be because there is not enough contact/pressure between the recording head and the tape (the whole signal patch mechanism looks a bit wonky), but I also suspect the bias oscillator circuit not doing its duty 

Today I managed to access what I think should be the bias oscillator circuit board, and noticed a few weird stuff :
-first, the coil seems unwound and the plastic enclosure broken : could anyone tell me from the pictures if it is indeed the coil, and if the loose wires are as they should? Is there anything wrong, and is there any way to test?
- Secondly, there is a strange, big, green component which could be a capacitor (?), but i'm not sure : could anyone tell me if i'm right assuming that it is a capacitor, and if the soldering on the top part is ok (it looks a bit odd to me)?







I can post other pictures of the tape echo later, but at the moment it's a bit dissassembled on my desk...
I don't know how far I want to go with it (it's not a Space Echo or an Echoplex, I don't have any schematic, the overall construction seems a bit amateurish... but on the other hand it's a curiosity with its charms and some nice features like sound on sound). The motor is running, the preamp part seems ok, as well as the playback head, so it might be worth a try... unless trying to fix the bias oscillator is too difficult?

Thank you in advance for any kind of help!

[Rob Strand]

The wires hanging out of the transformer are normal.

The crack is not normal.   It's a ferrite pot core and the ferrite has cracked.   That's going to screw up the bias oscillator.  It could even pull a heap of current and fry the oscillator transistors.

Unfortunately the behaviour those type of oscillators is quite critically dependent on the specifics of the ferrite core, especially the oscillation frequency.

You could glue it together with an ultra thin layer of super glue then push is together but it's almost certain not to work correctly because the tiny air gap at the repair point will affect how the core/oscillator behaves.  Trying to find a replacement core is also going to be difficult.  A look-alike replacement core probably won't oscillate at the right frequency.

Not sure what the green thing is.   Simplest possibility is it's a resistor (or inductor) on input power rail to stop noise getting out on the power rail.   It could also be part of the bias filter.

Here's some examples of those oscillators:

[snk]

Thank you, Rob, that's very helpful.

So it seems there aren't many chances to make it work if a thin layer of superglue doesn't do the job?
I will try, because if there is a chance to make it work, it's worth a try.

Not sure what the green thing is.   Simplest possibility is it's a resistor (or inductor) on input power rail to stop noise getting out on the power rail.   It could also be part of the bias filter.

Ok, I had never seen such an unusual resistor !
My interrogation is that a wire (going to the ground) was disonnected, and i am not sure where it's supposed to go. First, I thought it was going to the metal socket of the DIN plug (because it had solder on it and the wire had the right lenght), but the solder on that green thing makes me wonder if it isn't supposed to be wired to something on that side 

[Rob Strand]

So it seems there aren't many chances to make it work if a thin layer of superglue doesn't do the job?
I will try, because if there is a chance to make it work, it's worth a try.

Yes it's worth a try.   The superglue needs to be *ultra* thin and you need to push the halves together to make the glue exits the gap to reduce any a significant air-gap at the join.  The thing working on your side is some of the magnetic paths haven't broken.

The core is made from two halves which touch at the mid-line.   You don't want *any* glue seeping out and getting on to those mating surfaces.   Even go to the point where you don't put glue along that edge.   If you have any glue which stops the cores mating perfectly it can be worse than the air-gap at the repair join.   (When new these things look like they mate perfectly but there is actually 10's of micrometers of roughness and even that affects the core - it's just that it's known to be there at the start and accounted for when you use the core.)



You need to remove the core from the coil to do the repair.  Let the glue cure *completely* maybe a day or two.   When you mount the core go easy on tightening the screw.  You don't want over tighten it.

So if you are going to succeed at all you need to take a lot of care.

Ok, I had never seen such an unusual resistor !

Before about 1970 there were heaps of weird looking parts.   Companies made whatever.   Then things standardized and all looked the same.

[snk]

Thank you, Rob.
Unfortunately, I didn't manage to make it work : from what I hear, it seems that there are other stuff going on (weak sound, etc), despite other functions working as expected (bass, treble, output volume, switches, etc).
I struggle to have the input gain producing any echo, as well as having the erase head work...

[Rob Strand]

Thank you, Rob.
Unfortunately, I didn't manage to make it work : from what I hear, it seems that there are other stuff going on (weak sound, etc), despite other functions working as expected (bass, treble, output volume, switches, etc).
I struggle to have the input gain producing any echo, as well as having the erase head work...

You should be able to measure "something" in the order of 75kHz on the output of the transformer; see comment below.  You need to use the frequency range on your DMM.  At least that will tell you the bias oscillator is working.  If you don't get anything maybe check the transistors on the bias board are OK, especially no shorts between C and E; in-circuit measurements will be affected by low resistance TX windings.  AC volts pretty useless on a DMM at 75kHz unless you have a more expensive wide bandwidth DMM.  An oscilloscope would be better.

Check the DC resistance of *all* the heads.   Sometimes heads do go open circuit.

Another possible issue is head alignment.   You really don't won't to play around with that unless you can be sure everything else is working, then you need to read-up on how to set it up.

Maybe get some set-up tips (for all aspects) from service manuals like this,
https://www.synthxl.com/roland-re-201/
Something came to mind.  If the bias oscillator is off then you might have to adjust the bias traps.     If the bias oscillator frequency is way off then you might have to increase the bias oscillator caps to lower the frequency and make it close to what it was.    If it were mine I wouldn't mess with the bias traps at first.   What I'd do is try to measure the what frequency the bias traps are set to (the frequency they block) then see how far the bias oscillator is off and take it from there.

The 75kHz I mentioned earlier is just a ball-park number it could be anything from 40kHz to 150kHz; for old equipment the top end is probably less that 100kHz.

I don't mind telling you it's while since I've played with tapes and cassettes and it seems a lot of details have falling off the list.

[snk]

Wow, thank you once again Rob 
(It seems I don't get forum notifications anymore for some reasons, so I didn't see your reply until now)
I will proceed to the tests you suggest this weekend and see what it gives.