Few question , plz help

[larryB]

I am confused as to how exactly the waveforms function when using a phase splitter transformer like in the zeus amp , this tramsformer is driven class a open ended type , what do the voltage waveforms look like , reference to center tap . My drawing is simply similar to the zeus amp with input transformer missing and waveform shown .

Plz enlighten me from the image I made , wich of these is the corrent interpretarion .

Also , can this be done with an op amp or any source that is driven from a seperate yet lower voltage power supply .

thx you

[Miles Prower]

Originally Posted by larryB

...what do the voltage waveforms look like , reference to center tap...

Ask and ye shall receive (see attached)

Since this is a Class A amp, each final has a duty cycle of 360deg. The phase of the output depends on the phase of the OPT, and how the finals are connected. If this used gNFB off the secondary (which it doesn't) and you had positive feedback, you could change that by reversing the connections at the primary.

The phase splitting IST simply gives two signals 180deg out of phase, and (hopefully) of equal magnitudes.

Also , can this be done with an op amp or any source that is driven from a seperate yet lower voltage power supply .

See Here, Fig 24. This is how to implement a paraphase splitter with op-amps.

[larryB]

Miles thx you for your response .

One simple question , those 2 out of phase pics with the mosfets .

What are they referenced too . Imagine this as it were dc .

Are they floating at around + 2 volts ? Or are each waves from the transformer to center tap real AC , real alternating electrons like like wall current . This seems to be so from the ac books i read .

here is the zeus amp

http://www.tubecad.com/2004/image097.gif

[Miles Prower]

Originally Posted by larryB

What are they referenced too . Imagine this as it were dc .

Are they floating at around + 2 volts ?

The input AC rides on whatever the DC gate bias happens to be to set the no-signal bias current through the finals. That might be around 2.0Vdc, depending on what MOSFETs it uses.

[larryB]

When you say the ac signal rides on top of the dc , would that look more like the lieft most or the right most of the input pics .

thx you

What if both devices are not entirely matched , how can I bias them individualy ?

[Miles Prower]

Originally Posted by larryB

When you say the ac signal rides on top of the dc , would that look more like the lieft most or the right most of the input pics .

The left most one.

What if both devices are not entirely matched , how can I bias them individualy ?

You'd have to capacitor couple the inputs so each MOSFET would connect to its own bias network. For current matching, you would include a small current sense resistor in each source lead, usually 1R will do nicely for that. You could also use the Rdc of the OPT as a current sense resistor, but beware that the resistance of each side of the center tap won't be equal.

[larryB]

Ok my next question is , Altho I believe I know the answer , the voltage waveform from sourre of mosfet buz901 referenced to center tap will look like the right most ? correct ?

My next question is this , what if i were to get nearly matched mosfets and use this technique , see pic , is this good , i I read everywhere that differential is the way to go for phase splitting but I dont understand well ,

balanced/unbalanced preamplifier

my anwer to these people , show me the schematics !!!!

I modified a circuit , see pic attached , is this good ? Could this work ? modified from this .

http://www.analog.com/static/importe...40_00_1024.gif

I read somewhere that lateral mosfets really thrive with high voltage 30 ish voltage , problem is theres no good components reaching those , I wanted to use 28 volts I guess 13 volts will do . Either get better signal driving with less mosfet thriving or more happy mosfets and worse signal ... ??

Also , plz answer this question for me , I am dying to know , what is , is there , a difference between these 2 configurations when using an center tap output transformer push pull type .

http://sub.allaboutcircuits.com/images/03474.png

http://www.tubecad.com/2004/image097.gif

Thx you

[Miles Prower]

Originally Posted by larryB

Ok my next question is , Altho I believe I know the answer , the voltage waveform from sourre of mosfet buz901 referenced to center tap will look like the right most ? correct ?

Not if it's Class A, then it will look like the ones on the left. What you've got for the right-hand outputs is Class B, and that's really hideous for audio, and should be used only where you need lotsawatts and efficiency, not sonic excellence.

My next question is this , what if i were to get nearly matched mosfets and use this technique , see pic , is this good , i I read everywhere that differential is the way to go for phase splitting but I dont understand well , my anwer to these people , show me the schematics !!!!

For phase splitting, your only real choices are either a differential splitter or a cathodyne. Paraphase splitters in all their various iterations are more in the "quick 'n' dirty" category. So far, I've implemented these with vacuum tubes, and not solid state yet. The hollow state differential features excellent phase-to-phase balance, and sounds just great.

is this good ? Could this work ? modified from this .

Depends. I don't have any experience with that particular IC. May be someone here knows better about that dealie.

I read somewhere that lateral mosfets really thrive with high voltage 30 ish voltage , problem is theres no good components reaching those , I wanted to use 28 volts I guess 13 volts will do . Either get better signal driving with less mosfet thriving or more happy mosfets and worse signal ... ??

It's better to give the MOSFET finals a higher voltage to keep 'em away from the rails. Internal MOSFET capacitances do some squirrelly things when Vds drops to a very low value. See MOSFET spec sheets for this.

Also , plz answer this question for me , I am dying to know , what is , is there , a difference between these 2 configurations when using an center tap output transformer push pull type .

http://sub.allaboutcircuits.com/images/03474.png

http://www.tubecad.com/2004/image097.gif

Both of those are equally horrible, Class B types that will sound like due to the cross-over distortion these will produce. Best to reserve these types of finals to RF, where you can rely on LC tuners and/or bandpass filters to remove the harmonic distortion. You definitely don't want that in an audio final.

[larryB]

Hi , thx for all the response ,

when I showed these pic , I was just asking myself why one has grounded center tap and one has b+ center tap , sorry for not being clear enough .

http://sub.allaboutcircuits.com/images/03474.png

http://www.tubecad.com/2004/image097.gif

So the zeus amp with the input transformer has class b output ? , is what I was trying to say , the output waveform will be that of the right most ? , since mosfets dont work when there is reverse voltage ? It has continuous bias of 500mV distortion wont be so bad right ?

Yes mmmmmm....tasty.... phase splitters , any schematics ?

I will be going class a myself , I want the purest of ac sinewaves , with pure sinus current as well .

I thought mosfets liked going close to the fail and clipped smoothly ?

Not going to the rail = more power wasted , I am gonna ( try ) to use all i got .

[CBS240]

Hi

One thing to keep in mind is that Av of the inverting op-amp is -(Rfb/Rin) and for the non-inverting op-amp Av is 1+(Rfb/Rin). For a proper phase splitter -Av should equal +Av.