[sdiy] Easy transformer question

[Ken Stone]

Of course, you can take the "UBER" approach, and use four unselected
germanium diodes, and whatever signal transformers you have to hand, as long
as they "more or less" match. I did this back as my first DIY synth module,
and found that the missmatches actually colored the signal somewhat, though
depending on how you think, this can be a good thing. It is a very easy
circuit to get going. 

Ken

>I've toyed arround with such circuits a while ago:
>
>At first, the real ring modulator circuit comes from
>radio circuits, the useage of coils and transformers
>is very common there, and due to the narrow bandwith
>and high frequency it is easy to wind your own.
>
>The schematic you mentioned has only two transformers,
>the ones for the "audio" path.
>
>The transformer ring modulator (real ringmodulator) circuit
>lives from diode bridge biasing. That's why you need at least
>two transformers with a center tap. Almost always these
>taps are built using two windings in series, that's why
>you see four wires.
>Real ring modulators ofen have a third transformer in order
>to drive the bridge. But this can be circumvented
>by electronic means. However, the third transformer will enable
>you to have symmetric wiring and it will prevent DC problems.
>
>Sometimes there is no indication of winding polarity.
>So you have a 50% chance of not functional circuit.
>Better measure: drive the transformer primary with some (safe!)
>AC voltage (I assume it is a 1:2 or 1:4 or so type).
>Now connect the two secondary coils in series.
>You should be able to measure a substantial AC voltage
>on the secondary side, just as transformer ratio says.
>If the voltage is very very small, the two secondary
>coils are out of phase, try the other connection possibility.
>
>Diodes: Ge diodes are expensive and hard to get.
>It makes some difference in sound.
>Matched quadruples of diodes are near to impossible to get.
>The bridge offset depends on the diodes.
>I.e. they should be all the same and thermally coupled.
>Perhaps get a lot (a belt) of diodes and try to measure
>diode voltage at two different currents (forward of course).
>You can also add some resistance to each diode.
>
>If the third transformer is replaced by electronic means,
>there is a possibility to add DC voltage in order to cancel
>feedthrough. However, feedthrough will vary with temperature
>and aging, so a panel knob would be a good idea for that.
>
>
>Of course you need good audio transformers. 
>These are not cheap. But this  is audio application,
>not RF.
>The problem is often in the low end. Farnell seems to carry
>such items, with Flow ~ 40Hz or so. I got three of them.
>Below that you'll get a lot of distortion and signal loss.
>
>If the three transformer circuit is used, the modulator can of course
>not go to DC.
>
>m.c.
>
>
>
>
>-----Original Message-----
>From: Stephen Begin [mailto:trypannon at hotmail.com]
>Sent: Samstag, 8. März 2003 00:11
>To: synth
>Subject: [sdiy] Easy transformer question
>
>
>I finally got around to ordering the parts for a passive ring modulator, and
>I remember somebody told me that the transformer they'd seen used was mouser
># 42tl028.  http://www.mouser.com//catalog/specsheets/164249.pdf
>
>The circuit (http://www.synthfool.com/ringmod.html) shows two connections on
>one side of the transformer and three on the other, but this transformer has
>three connections on one side and four on the other, and I've been
>unsuccessful in my search for some sort of transformer 101 type pages.
>Basically what it boils down to is...do i only use two pins on the four pin
>side, or do I pair some of the pins together, or what?
>
>Thanks,
>Steve
>
>
_______________________________________________________________________
Ken Stone   sasami at hotkey.net.au  
Modular Synth PCBs for sale <http://www.blaze.net.au/~sasami/synth/>
Australian Miniature Horses & Ponies <http://www.blaze.net.au/~sasami/>