Zero Feedback Impedance Amplifiers

[PMA]

Originally posted by planet10


Earl Geddes has recently done research that shows that standard THD figures have almost no correlation to how things sound.

http://www.gedlee.com/distortion_perception.htm

dave

I have analyzed the wav files they are using for demonstration. The reason why they sound horrible is that the signals are corrupted at low level (parts of the signals are simply missing). This has not much to do with THD/IMD and IMHO this kind of demonstration is missleading.

On the other hand, I agree that THD and IMD as a simple number does not say much, that's why I am speaking about the necessity of spectral analysis.

[Susan-Parker]

Hi Everyone,

This weekend has been busy for me so only just back at my computer. Will reply to questions asap.

I have set up and measured the noise figures for the amplifier with the STW34NB20 PowerMESH mosfets.

Measured across the 8 ohm load. Test setup on bench. No special grounding or other tweeks so some of the noise may still be pickup in the leads etc. as it is very difficult to measure down at these levels.

Measurment bandwidth is 10 Hz to 100 kHz.

Full power is 20 Volts ac RMS - 50 watts - just into clipping.

Noise is 13 uV (micro-volts).

Best wishes,
Susan.

P.S.

Have to make supper now.

[Susan-Parker]

Hi hostmaster ,

Originally posted by hostmaster
Hi Susan,

Why that ?

What about 16VA toroid with 2x 9V with 2x 115V ?
Where did the end ?

Because I want to buy some one and i'try it then.

all the best

- uwe

From the results I have from the 30VA and 15VA toroids.

http://www.susan-parker.co.uk/zeus-toroid-amp-1.htm

Input Transformer - 15VA
Bandwidth: 10Hz to 25.5 kHz (-3 dB), 31.5 kHz (-6dB)

With modern winding and construction techniques I am hazarding that the transformers will be similar enough to give those sort of figures.

At the end of the day one has to try one out. These are made for 50/60 Hz mains and the wideband performance is much better that I would have thought.

At least the cost is small relative to custom transformers.

Best wishes,
Susan.

[Susan-Parker]

Hi extremy,

Originally posted by extremy
I had to ask for a visual on the windings. I'm asking about setting up the individual wire. Included is a graphic I hope is sent correctly.

Are you using a group of windings all aligned at a starting point, aligned on common center, or threaded in on a ratio? The last option I think was covered in a post about keeping the windings equal and therefore all phase correct. Although I have wondered if layering and then using a separate layer-wound concept would help.

I needed to ask about the windings. Otherwise, I can collect together most of the needed parts list from early posts.

If you are making a full transformer rather than a center tapped inductor then the best way is to wind all the wires together.

With 1.0 mm wire you could just use three windings although four would be preferable.

If you can wind all the turns in one go then that is best.

Separating the windings into layers - whilst functional - decreases the coupling and reduces the bandwidth. However this may still be more than adequate.

I haven't tried winding one (yet) so I can't judge the performance.

I'm also waiting to hear of compatible preamps. This will be my only hurdle, because I'd like to use an AV unit for the preamp ( a few video inputs, and audio inputs, and control of ), and run these amps for individual channels or on active crossover setups on all channels. If cost begins to be an issue, I will probably plan for a left/right/sub configuration, instead of 5.1 or 6.1.

A line driver unit using a DRV134 and AD815 per channel from your AV unit line outputs should be fine to drive the amplifier input transformer. Crossovers can be done here too.

Have to use the decoded outputs as Dolby won't license their decoder unless one has a big factory churning out boom boxes.

I, for one, wish there were a more powerful word than congratulations to Susan for the amp design.

Many thanks all,
GH

Thank you.

Best wishes,
Susan.

[Susan-Parker]

Dear All,

Just checked the output impedance of the amplifier with the STW34NB20 PowerMESH mosfets and the 75 watt transformer.

It's 0.9 ohms

That's configured 2:1 with two paralleled 1.0 mm output windings.

Best wishes,
Susan.

[Susan-Parker]

Dear All,

Some "real" test results now at:

http://www.susan-parker.co.uk/zeus-t...9-fft-1khz.htm

Best wishes,
Susan.

[Susan-Parker]

Dear All,

From Sowter:

=============

The price for the would bobbin only is UKP 72.25 each.

Complete transformer is UKP 128.95 each.

We will assign the type number 9840. Anyone can use this to order.

S/T/O suffixes as normal for the Fully Shrouded / Drop Through / Open Frame options with dimensions and weights for the N type.

SOWTER AUDIO TRANSFORMERS (E A Sowter Ltd)

Web site: http://www.sowter.co.uk

SALES: sales@sowter.co.uk

SUPPORT: support@sowter.co.uk

=============

Best wishes,
Susan.

[darkmoebius]

Originally posted by Susan-Parker
Just checked the output impedance of the amplifier with the STW34NB20 PowerMESH mosfets and the 75 watt transformer.

So, how does "Zeus" sound on a diet of STW34NB20's?

Is there a big noticable difference compared to the IRFP's? (given that there's no chance for A/B testing)

[Susan-Parker]

Hi Darkmoebius,

Originally posted by darkmoebius


So, how does "Zeus" sound on a diet of STW34NB20's?

Is there a big noticable difference compared to the IRFP's? (given that there's no chance for A/B testing)

There is a diference, but it is quite subtle.

The bass seems tighter with the bigger transformer and lower output impedance, but that could also be due in part to the TI headphone amp drive to the input transformer.

I need to make up a pair of amps that I can A/B test against my original monoblocks.

BTW there are now FFT tests on my original amp at:

http://www.susan-parker.co.uk/zeus-o...s-fft-1khz.htm

Which has pics of how it is built, plus a HP 8903B distortion v power level for both 2:1 and 4:1 output transformer configurations, plus 1 watt and 2 watt FFTs.



I have used the HP generator as I have to be very careful about not blowing up the soundcard input and I am still sorting out my methodologies to ensure that I don't.

I am confident that the results will improve with the FFT software's internal generator (the stuff above 8 khz in particular).

Best wishes,
Susan.

[Susan-Parker]

Dear All,

Added Sowter information to this page.

http://www.susan-parker.co.uk/zeus-out-tx-75w.htm

And started on a page for construction information, notes and hints at:

http://www.susan-parker.co.uk/zeus-construction.htm

Best wishes,
Susan.