John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
I'd honestly be more worried about them catching fire and potentially compromising the launch. Once in space/etc the mission is done and it's all PR.

As far as filling the cargo with meaningful scientific pursuit, I thought there's regulations on test flights against doing just so. With the caveat that I expect I'm very wrong about this. 🙂

Here's one heretical measurement, albeit not with a distortion magnifier: Nichicon Muse ES bipolar caps measured: <-120dB THD, <-140dB IMD
 
20 uF, 10k Mylar coupling C with Pease DA model extra RC branches, scaled

trace is Pease' DA modeled Coupling Cap vs Spice C

cursor inset for delta dB, phase, group delay

Now listen to it! Flip the bypass switch in and out until you hear a difference. Then tell me your ears match the simulation!!! Then we can discuss what a simple sine wave sweep does not show.

Dr Derfy,

Linear function (calculus - Wikipedia)
 
Last edited:
Nah. The rather embarassing* Ariane V first launch failure had the Cluster mission in the fairing, along with about 5 tonnes of concrete. The concrete made a significant hole when in landed again...

*Porting code that is proven to work is good. Not checking that the system it's being plugged into handles lockout the same way isn't. It basically registered a gyro failure and locked out the redundant gyro rather than the failed one. Ooops 🙂
 
Now listen to it! Flip the bypass switch in and out until you hear a difference. Then tell me you ears match the simulation!!! Then we can discuss what a simple sine wave sweep does not show.

Dr Derfy,

Linear function (calculus - Wikipedia)


And this is much more germane to the conversation of convolutions and Fourier transforms when referring to capacitor "linearity". 😉

Linear system - Wikipedia

And Bill, I figured I was wrong. 🙂
 
And this is much more germane to the conversation of convolutions and Fourier transforms when referring to capacitor "linearity". 😉

Linear system - Wikipedia
)

And thus the definition gap. I write linearly, Scott changes it to linearity and confusion enters. Or as perhaps mentioned before Scott and I just can't communicate.

The whole bit comes from looking at how an RC network charges at DC which meets the definition of linearity but not linear. The phase shift introduced does change the perceived sound of complex waveforms. But without communication arguments wander aimlessly.
 
20 uF, 10k Mylar coupling C with Pease DA model extra RC branches, scaled

trace is Pease' DA modeled Coupling Cap vs Spice C

cursor inset for delta dB, phase, group delay

I looked a bit more carefully at what you did and you are showing the effect. If you can try a square wave as the source. Then compare the input and output Fourier transforms. They will be different but not reàlly enough to be perceptable. Then compare the peak to peak voltage that should nicely show changes that are perceptable.
 
Then compare the input and output Fourier transforms. They will be different but not reàlly enough to be perceptable. Then compare the peak to peak voltage that should nicely show changes that are perceptable.

They contain the same information via the inverse transformations. Your statement above implies that they somehow indicate different things. BTW the p-p time domain aberrations are equally trivial, we measured them with the bridge setup. It sometimes took 1000x magnification to be visible.

I'm sure jcx can oblige since he has the circuits already entered.
 
Getting back to the 'real world', '-) This is what we published more than 30 years ago, and I stand by the measurements, since I made them myself.
Note Photo 6, testing a 10uf electrolytic with a 50K load. Not nothing, even though the loading is very small. Now let us put the same cap in a feedback loop with 500 ohms or so as a feedback resistor load. Look at Photo 1.
 

Attachments

Thank you Scott, for giving me a reason to explain why we chose to eliminate caps, once we measured them and found that best quality caps were both too big and too expensive, making servoing cheaper, with less design real estate, (one of my fav IC servo amps was the AD711 at the time by the way), and with lower noise and frequency response.
For example the REF cap that I used, cost about $100 each at the time. I have a few, want to buy some from me? '-) The measurement caps that I used in these measurements were almost as good, as shown in Photo 7, but they were not cheap either. Mylar (Photo 8.) was just so, so. And we found that we could hear Mylar caps, much to my consternation, because I used them up to the time where they were found to be audible.
However, in general, it is cheaper to make a servo, than to use good coupling caps.
 
I always wondered why you guys didn't concentrate on finding cheap well distributed film caps that had no problems rather than showing the worst and leaving the door open to the snake oil and voodu vendors.

I did show the worst and good and better and best. Up to others to try them/find them. There are many cheap PP caps that will get the job done. Panasonic comes to mind.

PS type are slowly disappearing --- only 1-2 mfr left on the planet now. Order some now or they may go the way of some of our favorite transistors.


-RM
 
Last edited:
I did show the worst and good and better and best. Up to others to try them/find them. There are many cheap PP caps that will get the job done. Panasonic comes to mind.

PS type are slowly disappearing --- only 1-2 mfr left on the planet now. Order some now or they may go the way of some of our favorite transistors.


-RM

Is it really a problem? C0G ceramic measure almost the same and are a lot more modern assembly process friendly. PS doesn't like the heat, that's for sure.
 
I do keep coming back to the thought of the stereo in the car, floating in space, belting out Bowie's "Space Oddity," speakers pumping away, making not a sound... Such strange imagery.

You would have to adjust the xover anyway since there is no air load on the speakers. And the enclosure. Etc ;-)

The race is on... 🙂
 

Attachments

  • space car.PNG
    space car.PNG
    120.8 KB · Views: 225
Last edited:
Status
Not open for further replies.