John Curl's Blowtorch preamplifier part II

[john curl]

We have found from experience that precision metal film resistors, notably bulk metal and some others, are as good and even subjectively better sonically than wirewound resistors, in general. For serious work, we sometimes use wirewound for power applications. For preamps, notably the CTC Blowtorch, we found that the original Roderstein 1/2W metal film resistors worked very, very, well, and they cost us $0.1 dollars each. Tube equipment might more usefully use fixed wirewound resistors, but not solid state, in general.

 

[john curl]

I just read the DALE spec. sheet off the internet, and it gives the Excess Noise rating.

 

[john curl]

The Dale rating is .1uV/V over a decade of frequency. Once you get that, I will translate it for you.

 

[iko]

Thanks John, I got that. And by the way, thanks for being patient with a beginner. If you guys feel that I'm wasting your time and no longer want to do a little teaching, just say so.

Vishay application note an0003 is being cited in the "Sound of silence" by Burkhard Vogel.

http://www.rironic.com/docs/49414/vse-an00.pdf

They list the following figures:

Bulk metal foil -40 dB
Wirewound -38 dB
Metal film -32 dB to -16 dB
Thick film -18 dB to -10 dB
Carbon composition -12 dB to +6 dB

They refer to this as the "noise index" and is calculated from the current noise according to

dB = 20 log (noise voltage / DC voltage)

where noise voltage is the current noise and is in uV/V.

The Sound of silence book defines the noise index as

NI = 1uV/1V/1decade

and

NI_e = 20 log NI

It can be a bit confusing that others call "current noise" the ratio of excess noise intensity to thermal noise intensity.

 

[iko]

OK John, so your Dale example of 0.1uV/V/decade would be
NI = -20 dB = 20 log (0.1)

I know it's trivial, but I finding it helps many people to spell things out sometimes.

 

[john curl]

You got it right! I could not have done it, without the book.

 

[simon7000]

We have found from experience that precision metal film resistors, notably bulk metal and some others, are as good and even subjectively better sonically than wirewound resistors, in general. For serious work, we sometimes use wirewound for power applications. For preamps, notably the CTC Blowtorch, we found that the original Roderstein 1/2W metal film resistors worked very, very, well, and they cost us $0.1 dollars each. Tube equipment might more usefully use fixed wirewound resistors, but not solid state, in general.

John,

Everyone knows that there is no way a cheap Resita resistor can be as good as an audio grade wirewound. That is just the way it is.

Of course the measurements may be a bit of an eye opener.

I should be able to post them Tuesday, consider it a St Pats day present.

 

[1audio]

Noise

Attached is the additional notes on noise from the Quan-Tech 315B resistor noise test set manual.

 

[john curl]

There can actually be overlap between the BEST metal film resistors and the WORST wirewound resistors. Crimped connections can be noisy. However, the BEST wirewound resistors will be generally quieter than most metal film resistors, if this is important.

 

[simon7000]

John,

Attached are the results from a Mills 5 watt wire-wound resistor, a Holco H4 and a vintage Resista. As the Mills is a 5 watt I would have expected it to have almost 20 db less distortion that the 1/2 watt models.

The Resista results were a bit surprising!

As usual anyone who wish to re-post this image to this thread as a PDF has my permission.

 

[iko]

I know you're talking to John, but may ask something? Can you please say what the dB reference level isdBr is?

From these images the best I can infer is that the Holco fundamental is at -85 and H3 is 80 dB lower at -165 dB, the Resista fundamental appears at -94 dB and H3 also about 80 dB lower at 0-174 dB, while the Mills wire-wound fundamental at -120 dB and it's H3 only 33 dB lower, at 153 dB.

So it's no problem to compare with the fundamental at such different values?

 

[dodo]

John,

Attached are the results from a Mills 5 watt wire-wound resistor, a Holco H4 and a vintage Resista. As the Mills is a 5 watt I would have expected it to have almost 20 db less distortion that the 1/2 watt models.

The Resista results were a bit surprising!

As usual anyone who wish to re-post this image to this thread as a PDF has my permission.

Is there a chance to see the test results for other resistors
for example Caddock MK132 ; Vishay102 , or PPR ?

 

[simon7000]

The fundamental is at 0db which is 15.81 volts. The spike at 1k is the residual after trimming for the lowest value with a 10 turn 20 ohm equivalent trim pot at the base of the 4 1k resistors in the bridge. Trim balance has some effect on the even order distortions but almost none on the odd orders. The residual changes during the hour or so it takes to get the readings.



if you didn't see this before has my baseline tests are shown a bit before here,

http://www.diyaudio.com/forums/atta...er-part-ii-reference-20resistor-20test-1-.pdf

As to the other resistors, when it all is done it will probably be published.

 

[john curl]

Hi Simon, can you get me a PDF so that I can read it too? Thanks in advance.

 

[leapcat]

pdf attached

 

[john curl]

Thanks Leapcat, I am computer challenged. '-)

 

[simon7000]

pdf attached

Wow PDF is screwed up showing Holco and Mills but Resista only shows label not graph!

Resista graph is almost mind blowing!

 

[iko]

Here.

 

[simon7000]

Here.

Thanks,

Now John has his St. Pat's day present. The Resista low cost resistor measures better than anything I have tried so far, including the Caddocks and Vishays!

Of course they are no longer made, when they can sell much more expensive stuff.

 

[iko]

Sure, nice to see for historical reasons, but would be nice to find something like this that's obtainable.