Transformer output for zen I/V

[dave slagle]

I just realized that Nelson published his Zen I/V converter and here is a slight modification to replace the two 10uf caps with a transformer.



LTspice file

Jfet Models

dave

[tubesguy]

Nice. Ideas/suggestions for the x-former? - Pat

[Zen Mod]

completely logical ;

it's easier for you to make xformers than to make resistors



in any case - to keep things clean and safe - it's wise to decrease PS to +/-12V or - even better - to increase it slightly , and cascode Jfets

[dave slagle]

Since I make resistors all the time, don't you mean it is easier for me to make transformers than 10uf capacitors :-)

In any event the low voltage drop of the transformer primary allows 15V supplies to be used which I guess could be spun as a green solution nearly halving the energy consumtion of the circuit.

I do want to say that outside of spice, this hasn't been tested yet so I'm just throwing it out there for discussion.

dave

[Ken Newton]

Originally Posted by tubesguy

Nice. Ideas/suggestions for the x-former? - Pat

A line-level transformer intended for push-pull dive and sufficiently gapped to handle the JFET bias currents. Perhaps, one of the amorphous core Lundahls would serve quite nicely, if not inexpensively, I should think.

[Zen Mod]

sorry - I just didn't saw these 1K resistors still in drains (I'm sorta in hurry - some 8 hours from your time )



what about not having drain resistors at all ?

even if Jfet paralleling is necessary to keep L and Rdc of primary as low as possible .....

[Ken Newton]

Originally Posted by Zen Mod

sorry - I just didn't saw these 1K resistors still in drains (I'm sorta in hurry - some 8 hours from your time )



what about not having drain resistors at all ?

even if Jfet paralleling is necessary to keep L and Rdc of primary as low as possible .....

You have to have the drain resistors so that you are not trying to drive the transformer only from the JFET drains, Their impedance is way too high and would cause both distortion and low-frequency roll-off with the transformer. Those resistors essentially present a 1k ohm source impedance for each primary winding.

Another huge advantage of transformer coupling this circuit is the dramatically increased PSRR. The cap coupled version has essentially a PSRR of zero, with respect to imbalanced power supply noise. This is due to the current-source like JFET drain nodes working against us. The transformer coupled version should have excellent PSRR due, again, to the current-source like drain nodes, but this time working in our favor.

[Zen Mod]

Originally Posted by Ken Newton

You have to have the drain resistors so that you are not trying to drive the transformer only from the JFET drains,.......

agree with everything ( even if I didn't quote entire post) - but I meant on custom made xformer ,optimized for that .

anyway - just a thought .......

[Ken Newton]

Originally Posted by Zen Mod

agree with everything ( even if I didn't quote entire post) - but I meant on custom made xformer ,optimized for that .

anyway - just a thought .......

Yes, I see what you would hope to do. The problem would be that I don't believe any custom audio transformer could be made with a high enough primary inductance. Such a configuration as you're thinking is possible with tubes because they have an internal plate resistance appearing in parallel with their transconductance element.

A grounded source FET has no such internally parallel drain resistance and so act only like transconductance elements (high impedance current-source like nodes) as seen from their drains. BTW, this is also true of bipolar transistors and tube pentodes.

[Ken Newton]

Originally Posted by dave slagle

Since I make resistors all the time, don't you mean it is easier for me to make transformers than 10uf capacitors :-)

dave

Dave, I just realized who you are. Whoops, sorry about the Lundahl suggestion.

It sounds as if you may be thinking of creating a transformer specifically for the Zen I/V circuit.