NEW !!! Full differential PI front end.

[moodz]

Thanks vasster ... I am very happy with results so far.

moodz.

[vasster]

First amp results for diff circuit.

pic 1 is the pulse output from a very ordinary diff amp.

gain is wired for 1000.

... looks quiet.



Well would you look at that ... there is absolutely NO way I could look at the output of my first amp at 10mV / div on my single sided PI front end.

I have a wireless mouse and keyboard and a compact flouro desk lamp 20 cm from the prototyping area. With my single sided design I had to turn these off in order to take measurements at much greater volts / div.




looks very promising.

I am too tired to go through target shots but I have found I have to use smaller test targets now because my 'standard' ones rail it !

I have changed the circuit previously published. The opamp config has changed and I dropped the coupling caps.

moodz.

Hi moodz, your work is novel. I'm waiting to see the changes you suggest.
I believe that this is going to be one of the major advances in the art of the PI field.

I mentioned it because it seems people are trying to reduce the capacitance of their TX coils. (I'm behind on this, trying to catch up...)

Regards,

-SB

simon ... fair call ...

My decay time has not improved significantly .... in fact it is pretty bad on the current results. However I feel that the sensitivity and noise is improved. This was my main goal ... I am not interested in developing some sort of new coil configuration for the sake of it.

regards Paul ( moodz )

[simonbaker]

Hi Simon ... you have a point ... I have an RF background and in that space coil geometry is very critical. A center tapped coil and a bifilar coil are different beasts. Your point about the capacitance and connection options are also valid .... The sort of frequencies we are working with here would certainly be seen in an RF engineering environment. ( though we would probably be trying to suppress these type of pulses not amplify them )

The same configuration could physically be produced with two separate single wound coils ... I have not tried it. The twisted pair would be more transmission like in nature due to closer coupling and distributed capacitance.

moodz.

I mentioned it because it seems people are trying to reduce the capacitance of their TX coils. (I'm behind on this, trying to catch up...)

Regards,

-SB

[moodz]

Hi Simon ... you have a point ... I have an RF background and in that space coil geometry is very critical. A center tapped coil and a bifilar coil are different beasts. Your point about the capacitance and connection options are also valid .... The sort of frequencies we are working with here would certainly be seen in an RF engineering environment. ( though we would probably be trying to suppress these type of pulses not amplify them )

The same configuration could physically be produced with two separate single wound coils ... I have not tried it. The twisted pair would be more transmission like in nature due to closer coupling and distributed capacitance.

moodz.

mmm ... I did a spice model and get the same result but no modelling of transmission lines etc ... just the mag coupling ... so maybe two coils connected the same way would work also... dunno have to try it.
moodz

[moodz]

Regarding the "bifilar coil", I'm wondering if there is a subtle issue.

First, for me, center-tap is very appropriate description because with flat coils especially, the physical layout of the wires is normally not a big deal with regard to the basic coil parameters. So if it's scramble-wound, bi-filar wound, or incremental wound, the topology is very similar -- by sorting through the windings you could convert one coil type to the other without rewinding.

Having said that, I'm wondering if there is an important subtle difference, specificially capacitance. With bifilar wound coils (connected like yours), I think the voltage between adjacent windings is about half the voltage across the coil. With incremental wound coils, each adjacent winding is at a much closer voltage, hardly any difference at all. Would that not effectively make the bifilar wound coil have more capacitance than the incrementally wound?

Or does somehow it all work out the same?...

Regards,

-SB

Hi Simon ... you have a point ... I have an RF background and in that space coil geometry is very critical. A center tapped coil and a bifilar coil are different beasts. Your point about the capacitance and connection options are also valid .... The sort of frequencies we are working with here would certainly be seen in an RF engineering environment. ( though we would probably be trying to suppress these type of pulses not amplify them )

The same configuration could physically be produced with two separate single wound coils ... I have not tried it. The twisted pair would be more transmission like in nature due to closer coupling and distributed capacitance.

moodz.

[simonbaker]

Regarding the "bifilar coil", I'm wondering if there is a subtle issue.

First, for me, center-tap is very appropriate description because with flat coils especially, the physical layout of the wires is normally not a big deal with regard to the basic coil parameters. So if it's scramble-wound, bi-filar wound, or incremental wound, the topology is very similar -- by sorting through the windings you could convert one coil type to the other without rewinding.

Having said that, I'm wondering if there is an important subtle difference, specificially capacitance. With bifilar wound coils (connected like yours), I think the voltage between adjacent windings is about half the voltage across the coil. With incremental wound coils, each adjacent winding is at a much closer voltage, hardly any difference at all. Would that not effectively make the bifilar wound coil have more capacitance than the incrementally wound?

Or does somehow it all work out the same?...

Regards,

-SB

[moodz]

First amp results.

First amp results for diff circuit.

pic 1 is the pulse output from a very ordinary diff amp.

gain is wired for 1000.

... looks quiet.



Well would you look at that ... there is absolutely NO way I could look at the output of my first amp at 10mV / div on my single sided PI front end.

I have a wireless mouse and keyboard and a compact flouro desk lamp 20 cm from the prototyping area. With my single sided design I had to turn these off in order to take measurements at much greater volts / div.




looks very promising.

I am too tired to go through target shots but I have found I have to use smaller test targets now because my 'standard' ones rail it !

I have changed the circuit previously published. The opamp config has changed and I dropped the coupling caps.

moodz.

[moodz]

Differential coil pics ...

You can see the twisted pair in the coil binding.

Centre connection is NOT a centre tap.

moodz.

[moodz]

Hi moodz,

What is the purpose of IC2?

It looks like your balanced design is getting better each time.

Hi hobbes ... IC2 is a voltage follower .. you probably figured that .... it isolates the relatively low impedance input on the inverting leg of the diff amp.

We dont want to draw any significant AC current through those caps or there will be balance and freq response problems.

BTW there are much better circuits that could be used here ... the o/p from the diff amp is single ended and pretty much the same as you would get from a traditional design. You could easily bolt the output of the diff amp onto many existing PI designs.

cheers ... moodz.

[hobbes_lives]

Hi moodz,

What is the purpose of IC2?

It looks like your balanced design is getting better each time.

[moodz]

Thanks, Moodz. Sorry about the apparent attitude. I woke up with my head feeling twisted and got off to a bad start. Then I really stepped in it because I meant to go anonymous before making my last post. (Big WHOOPS.)

But, really, it's not only because of your scat, my angst has been building - I get out of sorts when people have multi-winding coils; where it would be sooooooo simple to eliminate ambiguity by simply employing phase dots - instead of assuming the reader knows which way the multiple windings should go. But, NOOOO.

I feel like the Lone Ranger because it's like nobody, but NOBODY knows what a phase dot is, anyway. (Somebody, please help me. Where's a doctor - I think I may need... a LOBOTOMY.)

I've got it. The two windings are continuous * from top to bottom. I know you want to concentrate on less mundane issues than coil phasing... even when phasing is at the heart of the issue... .

(I'll be alright, tomorrow.)

edit:* I had it wrong. My original post stated the windings were anti-phase until I re-read what you wrote and made a change. You don't have to, but please, please find library parts with dots and use them. Maybe it'll catch on, but I won't hold my breath. I wouldn't have made a good RF engineer since I can't even properly phase a bifilar coil just because... .

... fair cop .... Redrawn with dots and hopefully clearer .... cheers from moods

[porkluvr]

For reference the dots would be at the capitalised ends ( ie A & B ).
moodz

Thanks, Moodz. Sorry about the apparent attitude. I woke up with my head feeling twisted and got off to a bad start. Then I really stepped in it because I meant to go anonymous before making my last post. (Big WHOOPS.)

But, really, it's not only because of your scat, my angst has been building - I get out of sorts when people have multi-winding coils; where it would be sooooooo simple to eliminate ambiguity by simply employing phase dots - instead of assuming the reader knows which way the multiple windings should go. But, NOOOO.

I feel like the Lone Ranger because it's like nobody, but NOBODY knows what a phase dot is, anyway. (Somebody, please help me. Where's a doctor - I think I may need... a LOBOTOMY.)

I've got it. The two windings are continuous * from top to bottom. I know you want to concentrate on less mundane issues than coil phasing... even when phasing is at the heart of the issue... .

(I'll be alright, tomorrow.)

edit:* I had it wrong. My original post stated the windings were anti-phase until I re-read what you wrote and made a change. You don't have to, but please, please find library parts with dots and use them. Maybe it'll catch on, but I won't hold my breath. I wouldn't have made a good RF engineer since I can't even properly phase a bifilar coil just because... .

This would be so much more clear if there were dots to indicate proper coil phasing. What is it, the new convention to leave evertbody guessing? Too much trouble, or condescension?

Aww, forget it. Everybody knows what bifilar wound means so phase dots would only be redundancy.

pork ... u r right if it is bifilar the dots are redundant. The dots used to be there I will have a look at my library symbol.

For reference the dots would be at the capitalised ends ( ie A & B ).

moodz

Ok, I was barking up the wrong tree. Would you summarize again what problems are you trying to solve with the differential design?

Regards,

-SB

for me ... noise.

and balancing front amp.

and getting better sensitivity.

and possibly improving decay times.

moodz