[kipman725]

Quote:

Originally Posted by amb /img/forum/go_quote.gif I guess you haven't seen Christer's noise measurements of LEDs and zeners as voltage references, as published on diyaudio.com. His finding was that higher voltage zeners, such as the BZX55C12 has very low noise, as good as the best (red) LEDs. It's the 5-6V zeners that perform worst. Kind of blows conventional wisdom out of the water.

thats just about the opposite of what my text books claim...

I may have to build a precision amplifier and do some testing myself as all the theory points to 5-6v zenners been the lowest noise (without using things like subsurface zenners etc.) as does the manufacturers data sheets.

big problem with using LEDS as a voltage refrance though for me is the % variance in voltage drop on the data sheets which as I design with worst case senarios in mind makes them unsuitible.

 

[Roam]

Quote:

Originally Posted by dsavitsk /img/forum/go_quote.gif I agree that I'd rather listen to a good tube amp than a good sand amp, but I don't think that either of those projects are one's I'd include on the list of Reigning King.

Try this one. Beware of 40-50MHz full power oscillations if the grounding and wiring layout isn't perfect. Finding a nicely matched pair of input tubes can also be a frustrating experience. Not shown is the power supply and the hookup for the hum-nuller circuit on the interstage transformer. Dropping the bias to 300V & 80-90mA will allow the use of 45's which may be easier to find than 50's since they're in current production.

 

[Tomo]

Wow, Master Gilmore!

You call that a challenging!? ... I don't think I can count the number of components on your design. See? 1, 2, ... , 10, many.

As for the Zener thingie, ... what do you consider "noise"? There are three types of noises that we should be concerned.

1. Johnson Noise ... (Thermal Noise or White Noise(?) )
2. 1 / f Noise
3. Shot Noise

There are more but they are insignificant. You all know the first one. You get that when you heat up carbon resistors. (Which is why people like metal film resistors.)

The second one is fairly obsecure as you only encounter discussions on it for semiconductor (internal) designs. You can see them in opamp spec sheets. They become important when you try to get high speed opamps, because they tend to have high 1 / f noise as we operate our chips at relative low frequencies.

The third one is of quantum mechanical nature (well call it Einstein-Boodoo). I think Zener diode or LED noise is from this. I was a BUM and I never studied its mechanics. But it has to do with current fluctuation when the device is on.

But I must say Shot Noise is REALLY REALLY small, so I could be wrong about the Zener noise is from Shot Noise ...

Tomo

 

[kipman725]

as far as I know the zenner exibits all of those noise characteristics but they are insignificant compared to the zenner noise (which I will just go and look up what its actauly caused by..)

The resistor noise that your refering to with heating up I think is Jhonson noise? which is unavoidable unless you can cool the circuit down or use smaller value resistors. It's much smaller than the noise caused by the film type though so I woulden't worry about it.

1/f noise I don't rearly understand yet but It effects all components I think?

 

[Pars]

Quote:

Originally Posted by Tomo /img/forum/go_quote.gif Wow, Master Gilmore! You call that a challenging!? ... I don't think I can count the number of components on your design. See? 1, 2, ... , 10, many. <snipped>

Actually, if you are referring to the hardwareanalysis site, that is Bruno Putzeys' and Sander Sassen's design.

More info on the hardwareanalysis site (run by Sander). I think this was also discussed on diyhifi.org and probably on diyaudio as well. PCBs will be (or are) available as well.

 

[rsaavedra]

Quote:

Originally Posted by kevin gilmore /img/forum/go_quote.gif If you want a real challenge you can try and build this thing. http://www.hardwareanalysis.com/content/image/11997/ distortion supposed to be below all of the best test equipment out there.

Hey sounds like what I've been looking for. A class A 100-watt in 8 ohms amp!

Here's the link to the info associated to those schematics:
http://www.hardwareanalysis.com/cont...diy_amplifier/

 

[lcp12345]

It is a pity that there is nobody to make the PCB of Mr Gilmore dynagbffet. I have built Balanced Dynalo, balanced M3(90%) and Beta22 (PCB in hand). If there is Dynagbfet, I will build one as well.

 

[NRD]

Mr. Gilmore,

Just curious why you don't use an active ground channel in your designs? Thanks for enlightening me as I am still learning.

All the Best