Final 03 Electrostatic Hybrid Speakers

You should first check the transformers.

I suspect, however, that this is the picture of the defective side.
FWICS the mains transformer (the small black one top right corner) looks burnt.
Take out the PCB and measure Ohms on the primary and secondary side of the small transformer.

Check if the other one looks 'normal'.

Finals are excellent speakers.

Just NEVER work on them with audio playing and wait for some time after the mains is disconnected.
Preferably short the blue and yellow wire before working on the electrostat (don't forget to remove the short when testing/using !!)

The transformers used are designed to have 230V on the primary side and must deliver 12V under a 1.5W load.
They are not designed to be used in the other 'direction' even though this is possible.

As you measured the good one has 330V out instead of 230V out which means 1.4x the nominal value.
Therefore it is more than likely that the magnetic core material is saturated which results in overheating.




As you can see they cost next to nothing so replacing them with this one is easy to do.
However, if it were my speakers I would either use these trafo's (and replace both) + the power supply trafo I would replace with one that delivers around 12V without a load.

OR (and this is perhaps easier and cheaper) replace the little trafo's with 15V versions.
These won't burn out and you can use the current mains transformer.

Chances are both speakers will play again.


Now... there is a downside as well in this case.
This is the efficiency of the stats will be lower than with the 12V trafo's.

The AC out will probably be around 240V from the 15V trafos where now the voltage is 330V.
This means the output voltage of the voltage multiplier will thus also be lower in the same factor.
I figure the DC bias voltage will be around 2500V with the 12V trafo's and around 2000V with the 15V trafos.
The efficiency of the stats will thus drop by about 2dB.

This wouldn't be much of a problem when you have an active X-over for the woofer, but most likely the systym is powered by one amp.
This means the woofer may be 2dB too loud now compared to when 12V trafo's are used.

IF this is indeed the case the simplest (and best) solution would be to ADD another 'section' to the voltage multiplier when using the 15V trafo.
Just 2 diodes + 2 capacitors are needed.


Then there is the question of how good the stats still work.
Some of them loose conductivity in the membrane lowering the efficiency, some loose tension which can cause distortion, arcing or sticking.
That's for later and I assume if the other one plays well the second one will also be good.

If the good one sounds a bit 'bass/midbass/lower mids' heavy chances are the stats need some TLC as well.
I hope not so IF the 'good' one sounds good the second one may only need it's trafo replaced.

I can add nothing constructive to this thread... just a comment or two.

Firstly I am amazed at the breadth and depth of knowlege in here. Someone comes in and says "help me fix my 'statics' " and BAM! Totally pleased to be a member here.

When I have heard electrostatics at hi-fi shows I'm usually impressed. Although, having said that, I haven't heard a good demo of any of the recent Quads. However, the few times I've been privileged to hear Martin Logans... wow.

Also I quite like the Magneplanar approach; in case anyone doesn't know, these look like 'statics (in that they are a panel loudspeaker) but they are actually a clever variation on conventional speaker drivers which use magnets and wires.

I bought an ex-demo pair of Magnepan's little SMGa. I say 'little', they are 4' tall. Paid £350 quid back in 1990. One panel has suffered a little damage due to a Musical Fidelity B200 going crazy. I reckon it put too many volts through the bass conductor wire which heated it up and caused it to part company with the bass membrane. I've managed to do a repair with some cling film. Still sounds ok. Incidentally, the crossover in the SMGa is a series circuit, not the usual parallel component arrangment.

The top end of the SMGa is a marvel. Initially it sounds a little soft but it actually sounds very natural when you adjust to the fact that it shares none of the characteristic resonances that normal tweeters have. The SMGa doesn't even have the 'quasi-ribbon' tweeter that current models have, just a few folds of wire glued to a plastic sheet. It sounds so sweet and detailed.

Bass extension is naturally limited in this, the smallest Maggie. But there's just enough output at 40Hz which means bass guitar is not bad, not bad at all. If you listen critically you can tell that the MDF speaker frames tend to resonate a bit and this helps the impression of fullness and warmth. Some people re-build the frames to reduce this effect but then I've heard that if you do this then they don't sound as good.

My Maggies live behind my home office door. Don't bring them out to play often as I can't have them far enough out into the room to work at their best. Besides, I'm still stunned by my Spendor A5s.

Taken a while back...



Derek

Should I think about doing this while I am messing around with the electronics? if so is it a complicated job.

I'm not in any hurry to sort these speakers out, my wife bought me them as a Christmas present last year and it has only taken me 7 months to pluck up the courage and attempt repairs

I think Slow but Sure - take it nice and steady one step at a time and they will soon be singing sweetly to me

Sorry, didn't intend to drag this so far off-topic.

Yes, agreed it's usually him :-)

My B200 actually failed... I then had it repaired at Hampshire Audio where I bought it. It fried both the left channel Maggie and the input stage of my then subwoofer's active crossover. True, the B200 didn't have quite enough current to drive the 4 Ohm SMGa, but the sound of the amp running out of juice was so obvious I could easily back things up a bit. That very rarely happened as one gets used to what is a safe position on the volume knob. However, one piece that often caught me out was a section in Faure's Requiem where the orchestra, organ and large choir are going at it like crazy. That tended to stretch things a bit. The B200 had plenty of Watts but not enough Amps for such a low impedance load? My present amp, the A3.2cr has plenty of current and when I do play with the Maggies all sounds in control. Very much so.  That old B200 failed yet again a couple of years later. This time it burnt out an Audax Kevlar driver that was part of an IPL DIY kit speaker. Totall toasted the voice coil.

In that picture from about '99 we had less furniture than now! Where the left speaker was there's now a sofa so it's not on 'fraid to say.



Current set-up, near as damn-it...

When the stats are a goner first attempt to open them up and see if the can be fixed.

The tensioning and making them conductive is the difficult part.

The stats I made do NOT have tensioned membranes at all, unique AFAIK.
I used a handy trick to overcome sticking. For that reason the stats sound better (not like a drum).

Won't disclose how to make them in public but can tell how to make them via PM.
Is a lot of work though, not difficult...

You need to get the bias voltage going first.
The membrane will only stick to one side when the bias voltage is present.

When you have the 15V trafos in you need to measure the AC voltage and could measure the DC voltages over the individual sections.

When that's all O.K. you can follow this procedure:

When powered off for a while:
1: connect all 3 stat wires together and leave them connected this way for an hour or so.
2: Remove the short between the stat wires after that hour(or a few hours).
3: Disconnect the woofer (if possible)
4: connect an amplifier and play music. (it should play very soft and distorted if anything does come out)
5: connect the mains.
The speakers should be starting to make sound rather quickly and become louder about equally fast for L and R
It could sound distorted but should become good within a very short while.
When it doesn't go any louder disconnect mains again and see if the speakers keep playing.
After quite some time they should start playing softer again and eventually distort.

This test tells us something about the conductivity of the membrane and state of the stats themselves.

Let's see how this works out.