High Quality White Noise Gen
Jim Slone
generation?
I have been using generic diodes and reversed biased transistors. Then
someone mentioned there are special parts available with better
characteristics.
Can anyone please give me a pointer?
Jim Slone
E
"Jim Slone" <jims...@esterlux.com> kirjoitti
viestiss�:4c03a915...@news.tpg.com.au...
> What are the best options for high quality audio white noise
> generation?
>
You can buy a CD full of white noise.
-ek
Jitt
jims...@esterlux.com says...
noise gen
--- news://freenews.netfront.net/ - complaints: ne...@netfront.net ---
John Larkin
wrote:
You can buy noise diodes from lots of people... just google <noise
diode>
If you want really flat, really gaussian noise, a mathematical random
stream (single-bit) or random word (dac) generator is probably best.
See AoE for details.
For audio, it doesn't matter much. A 10-volt zener biased at a few mA
is fine.
John
Robert Baer
We want Black noise.
John Fields
wrote:
>What are the best options for high quality audio white noise
Michael A. Terrell
Then use some of Obama's speeches.
--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.
John Fields
<mike.t...@earthlink.net> wrote:
>
>Robert Baer wrote:
>>
>> Jim Slone wrote:
>> > What are the best options for high quality audio white noise
>> > generation?
>> >
>> > I have been using generic diodes and reversed biased transistors. Then
>> > someone mentioned there are special parts available with better
>> > characteristics.
>> >
>> > Can anyone please give me a pointer?
>> >
>> > Jim Slone
>> Racist!
>> We want Black noise.
>
>
> Then use some of Obama's speeches.
---
Damn, Michael, Good one! :-)
Dirk Bruere at NeoPax
(c)Rap
--
Dirk
http://www.transcendence.me.uk/ - Transcendence UK
http://www.blogtalkradio.com/onetribe - Occult Talk Show
k...@att.bizzzzzzzzzzzz
<mike.t...@earthlink.net> wrote:
>
>Robert Baer wrote:
>>
>> Jim Slone wrote:
>> > What are the best options for high quality audio white noise
>> > generation?
>> >
>> > I have been using generic diodes and reversed biased transistors. Then
>> > someone mentioned there are special parts available with better
>> > characteristics.
>> >
>> > Can anyone please give me a pointer?
>> >
>> > Jim Slone
>> Racist!
>> We want Black noise.
>
>
> Then use some of Obama's speeches.
Ow!
John Larkin
Homophobe! We want Pink noise!
John
Glenn Gundlach
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Mon, 31 May 2010 15:44:06 -0700, Robert Baer
>
> >Jim Slone wrote:
> >> What are the best options for high quality audio white noise
> >> generation?
>
> >> I have been using generic diodes and reversed biased transistors. Then
> >> someone mentioned there are special parts available with better
> >> characteristics.
>
> >> Can anyone please give me a pointer?
>
> >> Jim Slone
> > Racist!
> > We want Black noise.
>
> Homophobe! We want Pink noise!
>
> John
Adobe Audition generates brown noise in addition to white and pink...
G²
John Larkin
>G�
Save the Earth. Make Green Noise.
John
Don Klipstein
A link previously mentioned in this thread offered what they said
was "pink noise", along with "white noise" and "brown/red noise" IIRC.
I do get a little into a mood to test the "pink noise" to see if it is
"truly pink" as opposed to something towards "purple noise"... I hope
only as a result of some midrange frequency response dip in my computer
loudspeakers ("they are fairly-el-cheapo" that *may* accentuate "lower
treble" and "bass end of midrange").
--
- Don Klipstein (d...@misty.com)
lektr...@gmail.com
> What are the best options for high quality audio white noise
> generation?
>
some white, pink, and other cokored/filtered noises for audio
testing. I used the Audacity free software program. You can generate
colored noise and filter it, then save either the project or the
waveform (various formats) for replay.
George Herold
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Mon, 31 May 2010 12:23:10 GMT, jimsl...@esterlux.com (Jim Slone)
> wrote:
>
> >What are the best options for high quality audio white noise
> >generation?
>
> >I have been using generic diodes and reversed biased transistors. Then
> >someone mentioned there are special parts available with better
> >characteristics.
>
> >Can anyone please give me a pointer?
>
> >Jim Slone
>
> You can buy noise diodes from lots of people... just google <noise
> diode>
Does anyone know what makes a high price "noise diode" any better than
your garden variety Zener?
>
> If you want really flat, really gaussian noise, a mathematical random
> stream (single-bit) or random word (dac) generator is probably best.
> See AoE for details.
>
> For audio, it doesn't matter much. A 10-volt zener biased at a few mA
> is fine.
Yup, and if the voltage asymmetry is a problem you can add the signal
from two diodes, one biased from the positve supply and the other from
the negative. (Though I've never tried this trick.)
>
> John
George H.
John Larkin
<ghe...@teachspin.com> wrote:
>On May 31, 12:56�pm, John Larkin
><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> On Mon, 31 May 2010 12:23:10 GMT, jimsl...@esterlux.com (Jim Slone)
>> wrote:
>>
>> >What are the best options for high quality audio white noise
>> >generation?
>>
>> >I have been using generic diodes and reversed biased transistors. Then
>> >someone mentioned there are special parts available with better
>> >characteristics.
>>
>> >Can anyone please give me a pointer?
>>
>> >Jim Slone
>>
>> You can buy noise diodes from lots of people... just google <noise
>> diode>
>
>Does anyone know what makes a high price "noise diode" any better than
>your garden variety Zener?
Probably a very small junction area (for low capacitance, high current
density) and maybe some doping profile. Not a power device!
Regular zeners get spikey and asymmetric and sort of oscillate at low
current. You can get noise diodes that behave at low currents.
>>
>> If you want really flat, really gaussian noise, a mathematical random
>> stream (single-bit) or random word (dac) generator is probably best.
>> See AoE for details.
>>
>> For audio, it doesn't matter much. A 10-volt zener biased at a few mA
>> is fine.
>
>Yup, and if the voltage asymmetry is a problem you can add the signal
>from two diodes, one biased from the positve supply and the other from
>the negative. (Though I've never tried this trick.)
Or sum the signals from a bunch of them. Central limit theorem.
John
Robert Baer
a 2N3439 (the first metal case transistor pulled out of an old junkbox)
in series with a DN3545 which uses a 33K resistor to make an 80uA
current source.
Using 20-30V, the noise is over 1V peak to peak, sawtooth looking
with some ringing at the start of the positive going ramp.
E-B breakdown was about 15V.
George Herold
Well that is not going to get rid of the voltage asymmetery.
If you need real Gaussian noise you can look at the shot noise from a
photodiode illuminated by an LED. Gives you noise ~100 times bigger
than the johnson noise of the sense resistor. (Assuming a 5 Volt DC
drop across R). But this has one big drawback. It's very sensitve to
vibrations.
George H.
>
> John- Hide quoted text -
>
> - Show quoted text -
George Herold
>
> - Show quoted text -
Oh, I never heard of that. Is it avalanche breakdown of the E-B
junction. Does this give noise out to a higher frequency than a 15V
zener?
George H.
John Larkin
Shot noise is the ultimate asymmetric waveform. It's made of
single-photon unidirectional spikes. If it manages to be Gaussian,
it's because a lot of asymmetric signals are being summed. Central
limit theorem.
http://en.wikipedia.org/wiki/Illustration_of_the_central_limit_theorem
John
George Herold
>
> - Show quoted text -
Hmmm, you are right.... I still don't think that summing the voltage
noise from a bunch of unipolarized zeners is going to get rid of the
voltage assymetry. But I'd be happy to be wrong too. Have you ever
tried this? It would be simple enough to put 5 or 6 together and see
what the output looks like. (As long as you don't mind my summing
with an opamp)... Maybe I can find some 'fun' time on Friday.
George H.
John Larkin
The math says it must be so. Still, the sum would converge to Gaussian
faster if half of the lopsided signals were inverted.
Zener noise gets more symmetric at higher currents. 10 mA is usually
OK for a small 10-volt zener.
I sometimes generate Gaussian-distributed numbers by summing a bunch
of RAN() calls, which are uniform on [0,1]. Six to ten works well, and
the crest factor is finite and known.
John
whit3rd
> > >> >Yup, and if the voltage asymmetry is a problem you can add the signal
> > >> >from two diodes, one biased from the positve supply and the other from
> > >> >the negative.
> Hmmm, you are right.... I still don't think that summing the voltage
> noise from a bunch of unipolarized zeners is going to get rid of the
> voltage asymmetry.
You can use two zeners on a single power supply, in bridge
configuration; couple the output through a transformer to get
the difference. Symmetry is guaranteed if you balance the bridge
correctly.
I've seen zeners (on a curve tracer) that were so noisy that the
breakover region was a blur.
George Herold
OK I'll just have to try it. I find the math more convincing if I can
see it in some experimental result. (Shot noise is a good example,
but it’s hard to get the current low enough so that you could see a
non-Gaussian distribution.)
>
> Zener noise gets more symmetric at higher currents. 10 mA is usually
> OK for a small 10-volt zener.
Yeah I think that is just the result of the I-V curvature. I'll try
running them down at low currents where the asymmetry is larger. This
is an experiment to show the central limit theorem and not make a good
Gaussian noise source.
George H.
>
> I sometimes generate Gaussian-distributed numbers by summing a bunch
> of RAN() calls, which are uniform on [0,1]. Six to ten works well, and
> the crest factor is finite and known.
>
George Herold
> On Jun 2, 10:37 am, George Herold <gher...@teachspin.com> wrote:
>
> > > >> >Yup, and if the voltage asymmetry is a problem you can add the signal
> > > >> >from two diodes, one biased from the positve supply and the other from
> > > >> >the negative.
> > Hmmm, you are right.... I still don't think that summing the voltage
> > noise from a bunch of unipolarized zeners is going to get rid of the
> > voltage asymmetry.
>
> You can use two zeners on a single power supply, in bridge
> configuration; couple the output through a transformer to get
> the difference. Symmetry is guaranteed if you balance the bridge
> correctly.
Ahh, more than one way to skin that cat.
Say speaking of noise sources, a colleague put together a digital
noise source. A counter steps through a look up table that feeds a
DAC. The lookup table holds a whole comb of sine waves equally spaced
in frequency space up to 32kHz. (I don’t recall the frequency spacing
but I could find out.. a few Hz or so.) The phases of all the sine
waves were chosen randomly. The DAC was 12 bit (an AD7541 I
think). The whole thing was clocked several times lower than Nyquist
limit (~128 kHz). Now the problem we observed, (and could never
cure), was intermodulation distortion above the 32 kHz cutoff. The
signals above the cutoff frequency were down by only 50 dB, and my
colleague was expecting something closer to 70 dB down. (Is that
right for 12 bit resolution on the DAC? ) I worked on all the layout
and analog portions of the circuit but could never make it any
better. There was talk about clock jitter on SED recently and I
wondered if this could be the source of the problem? Or maybe you
have some other idea.
Thanks, And If you think it might be clock jitter then can you tell
me how to measure it also. (OK I can use google too.)
George H.
Paul Keinanen
wrote:
>What are the best options for high quality audio white noise
>generation?
A pseudo random noise (PRN) generator (a shift register and a few XOR
gates in the feedback path) will generate quite good white noise
sequences at frequencies well below the clock frequency.
John Larkin
Here is my elegant, peer-reviewed research paper on the subject:
ftp://jjlarkin.lmi.net/Zener_Noise.pdf
John
John Larkin
The DAC quantization, and any nonlinearity, will add harmonic
distortion. Plus the sines may occasionally peak together and clip the
dac. Plus it's not trivial to get -70 dB distortion at these
frequencies.
We use random number generators and boxcar filters to generate
Gaussian noise to feed into dacs. This little box does this, all in a
Spartan3 FPGA...
http://www.highlandtechnology.com/DSS/T346DS.html
Rob cleverly, somehow, allowed the user to program the noise bandwidth
from mHz to 2 MHz without affecting the RMS amplitude.
John
Vladimir Vassilevsky
Paul Keinanen wrote:
For cheap random number generators, I prefer LCGs to LSFRs. It is very
simple to compute something like:
seed = (seed << 3) - seed + 5;
It cycles through all states, so you don't have to worry about
initialization. It also guarantees maximum period as long as the
multiplier and the additive are simple numbers.
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com
Michael A. Terrell
We might as well. They have no other value, aren't biodegradable and
they need an EPA permit for disposal.
Robert Baer
> non-Gaussian distribution.)
>
>> Zener noise gets more symmetric at higher currents. 10 mA is usually
>> OK for a small 10-volt zener.
>
> Yeah I think that is just the result of the I-V curvature. I'll try
> running them down at low currents where the asymmetry is larger. This
> is an experiment to show the central limit theorem and not make a good
> Gaussian noise source.
the noise going toward zero at around 120uA.
George Herold
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Wed, 2 Jun 2010 11:57:56 -0700 (PDT), George Herold
>
>
>
>
>
Yup, I've got that stored in my Zener noise folder. Someone (Bill
Sloman?) directed me to a previous thread on Zeners and there are some
nice references from Bell labs back in the '50s. One by McKay Phys.
Rev. 94 pg.877 is a nice summary.
George H.
George Herold
Very nice, (Are prices listed on your website?) Y'all make stuff
that's several orders of magnitude above what we're doing.
You know what I'd really like to make (and could probabbly sell too.)
is a random pulse generator. Short little pulses maybe 10nS or less
coming at an average rate of 1us or so. This would be a psuedo shot
noise generator. With a pot on the output one could change the
amplitude of the pulses and see how the noise scaled. ... Hmm it
would be nicer if the average pulse rate could be changed too. So
that one could keep the average 'current' the same, but made with
bigger 'electrons'. Sounds like a digital circuit. (Which I find a
bit boring) an analog 'something' would be more fun.
George H.
John Larkin
You could start with any noise source, like a zener thing, drive a
comparator to nab the peaks, and use a tach-like feedback to adjust
the comparator trip point to servo the rate. Fire a one-shot,
adjustable width if you like, and a pot to set amplitude. All analog,
pretty simple. You could also do the random rate thing digitally.
John
George Herold
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Thu, 3 Jun 2010 06:49:53 -0700 (PDT), George Herold
>
>
>
>
>
>
> - Show quoted text -
Yeah I had something like that in mind. I didn't quite know how to do
the random rate part.. By tach-like I assume you mean a frequency to
voltage converter.
"You could also do the random rate thing digitally."
Oh that sounds easier than the tach-servo idea. Divide by N before
the one shot. I don't need fine resolution on the rate, just the
ability to change it and then adjust the pulse height.
Are one shot's still 'kosher'? I once talked with an FPGA guy who had
never heard of them. Anything 'better' than the 74AHC123 from NXP?
This has a minimum pulse width of 5ns which would be fine. Other one
shots looked to be a lot slower.
Ahh so many circuits to try and so little time.
George H.
Phil Hobbs
How about avalanching a phototransistor? ;)
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
George Herold
wrote:
>
> - Show quoted text -
I never heard of it. Google gets a lot of hits though, I'll have to
explore tomorrow when I have higher bandwidth at work. Do I need a
special phototransistor or will anyone work? I assume I have to
reverse bias the BE junction.... But won't current flow out the
collector?
Thanks for something fun to think about.
George H.
lang...@fonz.dk
wrote:
point it at an Americium source from a smokedetector :)
real random noise
-Lasse
George Herold
Something optical is nice because you can change light intensity with
a knob.
George H.
>
> -Lasse
Phil Hobbs
I was just janking your chain--I had no idea that anybody had actually
tried that. Phototransistors are so horrible on so many levels that
it's pretty amusing to see someone trying to make fast pulses out of them.
George Herold
>
> - Show quoted text -
Oh...(Silly me) Well so how 'bout a photo-transistor with three
terminals (one where the base has a lead connected to it.) I'll short
the base and collector together and then reverse bias the EB
junction. With the bias voltage set just below where the thing wants
to zener. Then a photon could set the whole thing off?
I thought I had a photo transistor around here someplace, but I
couldn't find it. (The original prototype of optical puming used a
phototransistor, until I convinced every one that a photodiode was,
more sensitive (larger area), less noisy and faster.)
Ahh the other silly thought that this generated was using an avalanche
zener diode as a photo detector. I know the Zener's I use are
sensitive to light, so again if I was to park them just below the knee
voltage and then hit them with light could I get some gain out of
them? Kinda a poor man's APD.
George H.
Phil Hobbs
Shot noise is really, really Gaussian--I've measured it out to 7.1
sigma, where (false alarm rate)/(bandwidth) = 10**-11 (about 1 count per
day in a 1 MHz bandwidth). A decent comparator driving a one-shot can
make nice Poissonian pulses (as someone already suggested).
You can make okay phototransistors by connecting a real photodiode
between the base and collector of a real transistor. Still slow, but
much more sensitive. The next big advance is to replace the transistor
with an op amp. ;)
Bill Sloman
>
> > - Show quoted text -
>
> Oh...(Silly me) Well so how 'bout a photo-transistor with three
> terminals (one where the base has a lead connected to it.) I'll short
> the base and collector together and then reverse bias the EB
> junction. With the bias voltage set just below where the thing wants
> to zener. Then a photon could set the whole thing off?
>
> I thought I had a photo transistor around here someplace, but I
> couldn't find it. (The original prototype of optical puming used a
> phototransistor, until I convinced every one that a photodiode was,
> more sensitive (larger area), less noisy and faster.)
>
> Ahh the other silly thought that this generated was using an avalanche
> zener diode as a photo detector. I know the Zener's I use are
> sensitive to light, so again if I was to park them just below the knee
> voltage and then hit them with light could I get some gain out of
> them? Kinda a poor man's APD.
You might want to search on single photon avalanche photo-detection.
http://www.opticsinfobase.org/oe/abstract.cfm?uri=oe-16-3-2232
I haven't looked at the full paper but the list of references includes
some useful stuff.
--
Bill Sloman, Nijmegen
Phil Hobbs
Geiger-mode APDs are mostly a crock unless you need timing accuracy
better than, say, 10 ns. Their dark count rate is a good six orders of
magnitude worse than a PMT of the same area, and their dead time is 10
times longer.
On the other hand, they don't die if you put them in a helium
atmosphere, and they last longer than 5 years.
John Larkin
They don't do this, either...
http://physicsworld.com/cws/article/news/3446
Hamamatsu makes these tubes. They look like basketballs.
John
Bill Sloman
<pcdhSpamMeSensel...@electrooptical.net> wrote:
> On 6/4/2010 12:01 PM,Bill Slomanwrote:
Horses for courses. They can be a lot more compact and robust than
PMTs - on which I'm rather more expert.
--
Bill Sloman, Nijmegen
John Larkin
Vain fathead.
John
George Herold
>
> - Show quoted text -
PMT's and APD's are too expensive, for a simulation, PD shot noise
should be easy, or a zener. (how much does an APD cost, in hundreds?)
George H.
Bill Sloman
I've worked with PMT's - which is more than I can claim about SPAD's -
and I've persuaded the IEEE that I do know a little about PMTs - see
the IEEE Transactions on Electronic Devices volume 38 pages 679-680,
published in March 1991.
You do need to base your abuse on something more than your right-wing
intuition. Going off half-cocked like this does rather expose your
enthusiasm for believeing what you want to believe despite the absence
of any evidence to support your point of view.
--
Bill Sloman, Nijmegen
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Wed, 2 Jun 2010 11:57:56 -0700 (PDT), George Herold
>
>
>
And the peer was peer of you brain-damaged right-wing friends? Such as
James Arthur?
--
Bill Sloman, Nijmegen
John Larkin
When's the last time you did any real electronics? Or any real work?
You resent me having James as a friend? I sure don't. He can cook,
sing, and think. And ski, albeit sort of slowly. I mean, there's G,
the slope, and the coefficient of friction of p-tex on ice. Why mess
with that?
John
John Larkin
All you do here is claim how "expert" you are, or maybe were, without
ever making actual contributions.
You never *do* anything.
And when I don't have convincing evidence, I experiment and collect
some. A mouse isn't a soldering iron.
John
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sat, 5 Jun 2010 10:47:26 -0700 (PDT),Bill Sloman
>
>
>
> <bill.slo...@ieee.org> wrote:
> >On Jun 5, 1:24 am, John Larkin
> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> >> On Fri, 4 Jun 2010 14:56:44 -0700 (PDT),Bill Sloman
>
> >> <bill.slo...@ieee.org> wrote:
> >> >On Jun 4, 10:40 pm, Phil Hobbs
> >> ><pcdhSpamMeSensel...@electrooptical.net> wrote:
> >> >> On 6/4/2010 12:01 PM,Bill Slomanwrote:
>
> >> >> > On Jun 4, 4:41 pm, George Herold<gher...@teachspin.com> wrote:
> >> >> >> On Jun 4, 10:07 am, Phil Hobbs
>
> >> >> >> <pcdhSpamMeSensel...@electrooptical.net> wrote:
> >> >> >>> On 6/3/2010 11:12 PM, George Herold wrote:
>
> >> >> >>>> langw...@fonz.dk wrote:
> >> >> >>>>> On 3 Jun., 23:11, Phil Hobbs<pcdhSpamMeSensel...@electrooptical.net>
> >> >> >>>>> wrote:
> >> >> >>>>>> On 6/3/2010 9:49 AM, George Herold wrote:
>
> >> >> >>>>>>> On Jun 2, 4:36 pm, John Larkin
> >> >> >>>>>>> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> >> >> >>>>>>>> On Wed, 2 Jun 2010 12:30:19 -0700 (PDT), George Herold
>
> >> >> >>>>>>>> <gher...@teachspin.com> wrote:
> >> >> >>>>>>>>> On Jun 2, 1:59 pm, whit3rd<whit...@gmail.com> wrote:
> >> >> >>>>>>>>>> On Jun 2, 10:37 am, George Herold<gher...@teachspin.com> wrote:
<snip>
> >> >> > You might want to search on single photon avalanche photo-detection.
>
> >> >> >http://www.opticsinfobase.org/oe/abstract.cfm?uri=oe-16-3-2232
>
> >> >> > I haven't looked at the full paper but the list of references includes
> >> >> > some useful stuff.
>
> >> >> Geiger-mode APDs are mostly a crock unless you need timing accuracy
> >> >> better than, say, 10 ns. Their dark count rate is a good six orders of
> >> >> magnitude worse than a PMT of the same area, and their dead time is 10
> >> >> times longer.
>
> >> >> On the other hand, they don't die if you put them in a helium
> >> >> atmosphere, and they last longer than 5 years.
>
> >> >Horses for courses. They can be a lot more compact and robust than
> >> >PMTs - on which I'm rather more expert.
>
> >> Vain fathead.
>
> >I've worked with PMT's - which is more than I can claim about SPAD's -
> >and I've persuaded the IEEE that I do know a little about PMTs - see
> >the IEEE Transactions on Electronic Devices volume 38 pages 679-680,
> >published in March 1991.
>
> >You do need to base your abuse on something more than your right-wing
> >intuition. Going off half-cocked like this does rather expose your
> >enthusiasm for believeing what you want to believe despite the absence
> >of any evidence to support your point of view.
>
> All you do here is claim how "expert" you are, or maybe were, without
> ever making actual contributions.
Nothing that you'd be willing to acknowledge, particularly since most
of my contributions are references to the publshed literature, a
source that you seem ill-equipped to exploit
> You never *do* anything.
Not at the moment, and I find it frustrating.
> And when I don't have convincing evidence, I experiment and collect
> some. A mouse isn't a soldering iron.
You burble about 140dB of ripple rejection and then post the results
of your experiment that shows 67dB. You've got a lab full of expensive
equipment - if we are to take your boasting seriously - and boast like
a lion about your "insanely good" electronic designs, but then post a
result that might qualify you as a mouse in an undemanding
environment.
Somebody with your experience ought to be aware that if you want 140dB
of attenuation you are going to need more than a single stage of
filtering - stray impedances usually make it difficult to get more
than 60dB in a single stage.
--
Bill Sloman, Nijmegen
John Larkin
Burble? I questioned whether the Spice models of the c-multiplier were
accurate at mid-frequencies. None came close to 140, or even 80, dB at
frequencies where Early slope matters. So I tried some experiments.
and then post the results
>of your experiment that shows 67dB. You've got a lab full of expensive
>equipment - if we are to take your boasting seriously - and boast like
>a lion about your "insanely good" electronic designs, but then post a
>result that might qualify you as a mouse in an undemanding
>environment.
This is the sort of thing you could do. It doesn't take expensive
equipment to measure, say, 80 or 100 dB of ripple rejection. Just a
little thought and patience.
So try it - for real - and we'll compare numbers.
If not for money, just to keep your mind from rotting.
>
>Somebody with your experience ought to be aware that if you want 140dB
>of attenuation you are going to need more than a single stage of
>filtering - stray impedances usually make it difficult to get more
>than 60dB in a single stage.
My experience was entirely inadequate to make such a pronouncement.
I did some tests. Now I know more than I did before. You don't
approve?
John
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sat, 5 Jun 2010 15:00:09 -0700 (PDT),Bill Sloman
Not to mention the transistors, the capacitors and the resistors - I
don't keep a stock of components, and while I could get them, there is
a mininum order charge which menas that I'd better order all of what
I'd need in one hit
> So try it - for real - and we'll compare numbers.
I'm still trying to use the gEDA circuit design software to prepare a
schematic - and the concommitant parts list - for my oscillator
project, which is more interesting, if still not interesting enough to
motivate me to spend enough time rooting around in the gEDA
hierachical design system.
> If not for money, just to keep your mind from rotting.
My mind is fine, but my motivation isn't what it could be.
> >Somebody with your experience ought to be aware that if you want 140dB
> >of attenuation you are going to need more than a single stage of
> >filtering - stray impedances usually make it difficult to get more
> >than 60dB in a single stage.
>
> My experience was entirely inadequate to make such a pronouncement.
It probably isn't, but you clearly didn't think hard about the earlier
experience you were getting while you were getting it.
> I did some tests. Now I know more than I did before. You don't
> approve?
It is certainly a step in the right direction - but one that you might
have taken a few decades earlier. In your case, it obviously needs the
right sort of customer asking the right sort of questions to get you
interested. People with academic curiousity tend to get more out of
the experience they have had because they put more time into making
sense of their results. They often find that some academic had written
up the whole issue a few decades earlier. It's hard to avoid re-
inventing the wheel if you did't know that the wheel existed in the
first place.
--
Bill Sloman, Nijmegen
Bill Sloman
>
> > - Show quoted text -
>
> PMT's and APD's are too expensive, for a simulation, PD shot noise
> should be easy, or a zener. (how much does an APD cost, in hundreds?)
I've got a pair in my drawer that cost about $100 each, some ten or
twelve years ago. I don't know if they work ...
--
Bill Sloman, Nijmegen
George Herold
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sat, 5 Jun 2010 15:00:09 -0700 (PDT), Bill Sloman
>
>
>
>
>
> I questioned whether the Spice models of the c-multiplier were
> accurate at mid-frequencies. None came close to 140, or even 80, dB at
> frequencies where Early slope matters. So I tried some experiments.
The Early effect goes away at at high frequencies?
George H.
>
> and then post the results
>
> >of your experiment that shows 67dB. You've got a lab full of expensive
> >equipment - if we are to take your boasting seriously - and boast like
> >a lion about your "insanely good" electronic designs, but then post a
> >result that might qualify you as a mouse in an undemanding
> >environment.
>
> This is the sort of thing you could do. It doesn't take expensive
> equipment to measure, say, 80 or 100 dB of ripple rejection. Just a
> little thought and patience.
>
> So try it - for real - and we'll compare numbers.
>
> If not for money, just to keep your mind from rotting.
>
>
>
> >Somebody with your experience ought to be aware that if you want 140dB
> >of attenuation you are going to need more than a single stage of
> >filtering - stray impedances usually make it difficult to get more
> >than 60dB in a single stage.
>
> My experience was entirely inadequate to make such a pronouncement.
>
> I did some tests. Now I know more than I did before. You don't
> approve?
>
k...@att.bizzzzzzzzzzzz
wrote:
>On Jun 5, 6:52�pm, John Larkin
At high frequencies other parasitics become more important? Early gets
swamped by Cce?
George Herold
wrote:
> On Sun, 6 Jun 2010 13:36:09 -0700 (PDT), George Herold <gher...@teachspin.com>
>
> - Show quoted text -
Hmm OK... LTspice didn't seem to show that, (90dB of attenuation at
100kHz), but maybe I have to do some real measurements. I'm afraid I
don't really understand the Early effect/ voltage.
George H.
Bill Sloman
> On Jun 5, 6:52 pm, John Larkin
>
> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
And they can offer single photon detection of longer wavelength
photons than any photomultiplier tube can pick up. For some
apllication this is vital.
> > >> >> >Horses for courses. They can be a lot more compact and robust than
> > >> >> >PMTs - on which I'm rather more expert.
>
> > >> >> Vain fathead.
>
> > >> >I've worked with PMT's - which is more than I can claim about SPAD's -
> > >> >and I've persuaded the IEEE that I do know a little about PMTs - see
> > >> >the IEEE Transactions on Electronic Devices volume 38 pages 679-680,
> > >> >published in March 1991.
>
> > >> >You do need to base your abuse on something more than your right-wing
> > >> >intuition. Going off half-cocked like this does rather expose your
> > >> >enthusiasm for believeing what you want to believe despite the absence
> > >> >of any evidence to support your point of view.
>
> > >> All you do here is claim how "expert" you are, or maybe were, without
> > >> ever making actual contributions.
>
> > >Nothing that you'd be willing to acknowledge, particularly since most
> > >of my contributions are references to the publshed literature, a
> > >source that you seem ill-equipped to exploit
>
> > >> You never *do* anything.
>
> > >Not at the moment, and I find it frustrating.
>
> > >> And when I don't have convincing evidence, I experiment and collect
> > >> some. A mouse isn't a soldering iron.
>
> > >You burble about 140dB of ripple rejection
>
> > Burble?
>
> Maybe he say you coming through the turgey(sp) wood.
Charles Lutwidge Dodgson worked as an academic mathematician in the
areas of geometry, matrix algebra and mathematical logic, none of
which would appeal to John Larkin.
> > I questioned whether the Spice models of the c-multiplier were
> > accurate at mid-frequencies. None came close to 140, or even 80, dB at
> > frequencies where Early slope matters. So I tried some experiments.
>
> The Early effect goes away at at high frequencies?
No, but it tends to get swamped by Miller capacitance.
--
Bill Sloman, Nijmegen
John Larkin
No, but the optput capacitor impedance continues to decline.
There are a few frequency zones:
Dc to where the base lowpass filter kicks in: 0 dB ripple attenuation.
A region where the Early thing works, roughly -50 dB.
A slope downward, beginning at the corner frequency set by Re and the
output filter capacitance Cf.
Some high frequency where the attenuation is Cf/Cce
More or less.
I was interested in that second one, where simple calculations or
Spice aren't necessarily predictive.
John
John Larkin
Just imagine a resistor from collector to emitter. The question is,
what's the value?
John
John Larkin
Tulgey.
>
>Charles Lutwidge Dodgson worked as an academic mathematician in the
>areas of geometry, matrix algebra and mathematical logic, none of
>which would appeal to John Larkin.
I can recite "Jabberwocky" by heart, and do sometimes if the beer or
wine are of sufficient quality.
This Bandit chardonnay here is actually pretty good.
John
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sun, 6 Jun 2010 15:37:19 -0700 (PDT),Bill Sloman
My wife recently found an excuse to open some of our 2004 Gosset
Polish Hill Riesling - last year she thought that the 2003 was better,
but this year the 2004 does seem to come into its own.
http://www.grosset.com.au/wines_polishhillriesling.htm
Happily, it doesn't inspire either of us to recite poetry. If it did,
this might be more appropriate than "Jabberwocky".
http://www.poetry-online.org/chesterton_wine_and_water.htm
--
Bill Sloman, Nijmegen
John Larkin
How can a wine critic know that the wine tastes of "glacial gravel,
slate and shale"? Does he munch on rocks for reference?
I've seen a lot of "rock" references in wine criticism lately. "Hints
of cherry and apricot mold" are passe. Wine snobs are like
audiophools. Double-blind testing shows them as the delusional
fatheads that they are.
There is a trend in California to buy tasty cheap wine. Makes sense to
me.
The other trend is bars and restaurants that serve "draft" wine, right
out of the barrel.
John
Jim Thompson
Early effect changes the base WIDTH, thus the current gain.
Stick to sales, John :-)
...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
The only thing bipartisan in this country is hypocrisy
George Herold
Ahh excellent. It's this one that makes it not work so well when I
load the output too much. (Or do I just need a bigger cap?)
>
> Some high frequency where the attenuation is Cf/Cce
>
> More or less.
>
> I was interested in that second one, where simple calculations or
> Spice aren't necessarily predictive.
>
> John- Hide quoted text -
Thanks so much for spelling it all out.
George H.
George Herold
My daughter (age 10) recited it at the last company/ holiday poetry
reading.
She may make a mistake or two, but she’ll kill ya with charm.
>
> This Bandit chardonnay here is actually pretty good.
>
George Herold
Current gain is modulated by the collector-base voltage? That seems
to explain it.
George H.
>
> Stick to sales, John :-)
>
> ...Jim Thompson
> --
> | James E.Thompson, CTO | mens |
> | Analog Innovations, Inc. | et |
> | Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
> | Phoenix, Arizona 85048 Skype: Contacts Only | |
> | Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
> | E-mail Icon athttp://www.analog-innovations.com| 1962 |
>
> The only thing bipartisan in this country is hypocrisy- Hide quoted text -
John Larkin
For the purposes of measurement and modeling, an equivalent c-e
resistor value is perfectly appropriate. That's how it behaves here.
Without real numbers, all models are worthless.
John
Jim Thompson
In your dreams, dorkfish :-)
...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
John Larkin
At modest currents, emitter dynamic resistance Re is inverse on
current, actually about 25 ohms divided by emitter current in mA. So
the corner frequency of Re * Cl changes with load. I don't know how
the Early feedthrough changes with load current.
Capacitor ESR also forms a voltage divider with Re, so yet more ripple
blasts through as Re goes down at higher currents.
Not a simple circuit!
John
Phil Hobbs
>
> PMT's and APD's are too expensive, for a simulation, PD shot noise
> should be easy, or a zener. (how much does an APD cost, in hundreds?)
>
> George H.
The nice thing about PIN photodiodes is that there's a first-principles
relationship between the DC and noise currents. That's a great
calibration principle for instruments.
I have about 100 InGaAs APD/preamp modules that I got for about 75 cents
each--probably 0.5 cents on the dollar.
Michael A. Terrell
George Herold wrote:
>
> My daughter (age 10) recited it at the last company/ holiday poetry
> reading.
> She may make a mistake or two, but she’ll kill ya with charm.
Oh, that's just great! More Juvenile crime. ;-)
--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Sun, 6 Jun 2010 19:19:03 -0700 (PDT),Bill Sloman
Max Coltheart is a professor of cognitive psychology who has a long-
running experiemnt in which he gives a talk about wines and wine
tastng to academic audiences - including a variety of this kind of
wine-tasting terminology - tests his audience's performance as wine
tasters and records the results. He hasn't published it yet - and I
took part in one such experiment nearly thirty years ago. Informally,
I've heard from him that some phrases are more useful than others, but
the communication is not all that analytical
http://homepages.wmich.edu/~dilworth/Imaginative_versus_Analytical_Experiences_of_Wines.pdf
Wine-tasters are anything but fat-headed, but their language is
imprecise, if evocative.
> There is a trend in California to buy tasty cheap wine. Makes sense to
> me.
It would. You aren't great of acquiring background knowledge.
> http://www.enthusiasticspirits.com/r/products/three-thieves-bandit-pi...
>
> The other trend is bars and restaurants that serve "draft" wine, right
> out of the barrel.
Modern wines are often made to be instantly drinkable. This doesn't
mean that they can't get better with age, but many don't. If the wine-
maker has a gas-chromatograph - ideally with a tandem mass-
spectrometer as a detector - they can have a very clear idea of what
their wine tastes and smells like, and the good ones know how to
create an attractive mixture of sensations, and have a pretty good
idea of how that mixture will change and develop with time,
It an art, and there are some great artist out there. Wolfgang Blass
in Australia was an early pioneer of this approach, and while he isn't
a great artist, I've drunk some very pleasant wines sold under his
name.
http://en.wikipedia.org/wiki/Wolf_Blass
--
Bill Sloman, Nijmegen
George Herold
Yup, It's all making sense. A few years ago a made a bunch of RC
lowpasses and measured ESR's. I came to the conclusion that you
couldn't expect anything much lower than 0.1 ohms. But I didn't try
any really big Al electro's (>100's uF)
George H.
>
> Not a simple circuit!
>
> John- Hide quoted text -
>
George Herold
Yeah, Lafe Spietz does shot noise thermometry with tunnel junction
diodes. I've got an idea for a similar circuit trick that will,
balance shot noise and johnson noise to measure temperature... The
shot noise gives you the circuit gain and bandwidth, which then drop
out in a ratio.... looks good on paper but I still have to try it.
(You run the same DC current from the shot noise through the resistor
that gives you the johnson noise and it's DC voltage is then the
thermal voltage. (or maybe two times it.)
George H.
> I have about 100 InGaAs APD/preamp modules that I got for about 75 cents
> each--probably 0.5 cents on the dollar.
>
> Cheers
>
> Phil Hobbs
>
> --
> Dr Philip C D Hobbs
> Principal
> ElectroOptical Innovations
> 55 Orchard Rd
> Briarcliff Manor NY 10510
> 845-480-2058
> hobbs at electrooptical dot nethttp://electrooptical.net- Hide quoted text -
John Larkin
<bill....@ieee.org> wrote:
>Modern wines are often made to be instantly drinkable. This doesn't
>mean that they can't get better with age, but many don't. If the wine-
>maker has a gas-chromatograph - ideally with a tandem mass-
>spectrometer as a detector - they can have a very clear idea of what
>their wine tastes and smells like, and the good ones know how to
>create an attractive mixture of sensations, and have a pretty good
>idea of how that mixture will change and develop with time,
Can a GC tell that the vines grew in water that had been filtered
through glacial gravel, slate and shale?
John
John Larkin
The 120 uF polymer aluminums I'm using are 25 milliohms typ. I use one
to LC filter the +15 from the wart, another at the c-mult output. Each
takes about a quarter of a square inch of PCB surface, not too bad.
John
Martin Brown
GC can't but ICP-MS can. I worked on techniques used to prove provenance
of very expensive French wines a long time ago based on the pattern of
various alkali metals and rare earths present at ultra trace levels. We
required about 10ml of wine of which only a couple of ml were analysed.
That left the rest of the bottle to be disposed of.
They were concerned about rather high quality forgeries. A more recent
paper (not by our group) but using similar kit is online at:
SIRA can also spot adulteration by adding cane sugar to grape juice.
Regards,
Martin Brown
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Mon, 7 Jun 2010 02:16:57 -0700 (PDT),Bill Sloman
>
Beats me. Why not ask a wine-maker? In theory, the mineral content of
the water could influence the volatile organic compounds that give
wine its odour and flavour, but I'd be surprsied if the comment was
much more than decorative verbiage.
The GC can detect what you can smell, with somewhat more
discrimination, allowing a more rational approach to wine-making than
was possible in previous centuries, not that the more subjective
approach couldn't produce great wines.
--
Bill Sloman, Nijmegen
John Larkin
<bill....@ieee.org> wrote:
>On Jun 7, 3:55 pm, John Larkin
><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> On Mon, 7 Jun 2010 02:16:57 -0700 (PDT),Bill Sloman
>>
>> <bill.slo...@ieee.org> wrote:
>> >Modern wines are often made to be instantly drinkable. This doesn't
>> >mean that they can't get better with age, but many don't. If the wine-
>> >maker has a gas-chromatograph - ideally with a tandem mass-
>> >spectrometer as a detector - they can have a very clear idea of what
>> >their wine tastes and smells like, and the good ones know how to
>> >create an attractive mixture of sensations, and have a pretty good
>> >idea of how that mixture will change and develop with time,
>>
>> Can a GC tell that the vines grew in water that had been filtered
>> through glacial gravel, slate and shale?
>
>Beats me. Why not ask a wine-maker? In theory, the mineral content of
>the water could influence the volatile organic compounds that give
>wine its odour and flavour, but I'd be surprsied if the comment was
>much more than decorative verbiage.
In other words, delusional fat-headedness. There's a lot of that going
around. He should be careful about chewing on rocks... it's bad for
your teeth.
>
>The GC can detect what you can smell, with somewhat more
>discrimination, allowing a more rational approach to wine-making than
>was possible in previous centuries, not that the more subjective
>approach couldn't produce great wines.
So the GC knows more about what wines I like than I do?
No, thanks. The Bandit tastes great.
John
Phil Hobbs
Fun.
John Larkin
<bill....@ieee.org> wrote:
>On Jun 6, 12:52 am, John Larkin
><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> On Sat, 5 Jun 2010 15:00:09 -0700 (PDT),Bill Sloman
>>
>>
>>
>> <bill.slo...@ieee.org> wrote:
>> >On Jun 5, 8:18 pm, John Larkin
>> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> >> On Sat, 5 Jun 2010 10:47:26 -0700 (PDT),Bill Sloman
>>
>> >> <bill.slo...@ieee.org> wrote:
>> >> >On Jun 5, 1:24 am, John Larkin
>> >> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> >> >> On Fri, 4 Jun 2010 14:56:44 -0700 (PDT),Bill Sloman
>>
>> >> >> <bill.slo...@ieee.org> wrote:
>> >> >> >On Jun 4, 10:40 pm, Phil Hobbs
>> >> >> ><pcdhSpamMeSensel...@electrooptical.net> wrote:
>> >> >> >> On 6/4/2010 12:01 PM,Bill Slomanwrote:
>>
>> >> >> >> > On Jun 4, 4:41 pm, George Herold<gher...@teachspin.com> wrote:
>> >> >> >> >> On Jun 4, 10:07 am, Phil Hobbs
>>
>> >> >> >> >> <pcdhSpamMeSensel...@electrooptical.net> wrote:
>> >> >> >> >>> On 6/3/2010 11:12 PM, George Herold wrote:
>>
>> >> >> >> >>>> langw...@fonz.dk wrote:
>> >> >> >> >>>>> On 3 Jun., 23:11, Phil Hobbs<pcdhSpamMeSensel...@electrooptical.net>
>> >> >> >> >>>>> wrote:
>> >> >> >> >>>>>> On 6/3/2010 9:49 AM, George Herold wrote:
>>
>> >> >> >> >>>>>>> On Jun 2, 4:36 pm, John Larkin
>> >> >> >> >>>>>>> <jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> >> >> >> >>>>>>>> On Wed, 2 Jun 2010 12:30:19 -0700 (PDT), George Herold
>>
>> >> >> >> >>>>>>>> <gher...@teachspin.com> wrote:
>> >> >> >> >>>>>>>>> On Jun 2, 1:59 pm, whit3rd<whit...@gmail.com> wrote:
>> >> >> >> >>>>>>>>>> On Jun 2, 10:37 am, George Herold<gher...@teachspin.com> wrote:
>>
>> ><snip>
>>
>> >> >> >> > You might want to search on single photon avalanche photo-detection.
>>
>> >> >> >> >http://www.opticsinfobase.org/oe/abstract.cfm?uri=oe-16-3-2232
>>
>> >> >> >> > I haven't looked at the full paper but the list of references includes
>> >> >> >> > some useful stuff.
>>
>> >> >> >> Geiger-mode APDs are mostly a crock unless you need timing accuracy
>> >> >> >> better than, say, 10 ns. Their dark count rate is a good six orders of
>> >> >> >> magnitude worse than a PMT of the same area, and their dead time is 10
>> >> >> >> times longer.
>>
>> >> >> >> On the other hand, they don't die if you put them in a helium
>> >> >> >> atmosphere, and they last longer than 5 years.
>>
>> >> >> >Horses for courses. They can be a lot more compact and robust than
>> >> >> >PMTs - on which I'm rather more expert.
>>
>> >> >> Vain fathead.
>>
>> >> >I've worked with PMT's - which is more than I can claim about SPAD's -
>> >> >and I've persuaded the IEEE that I do know a little about PMTs - see
>> >> >the IEEE Transactions on Electronic Devices volume 38 pages 679-680,
>> >> >published in March 1991.
>>
>> >> >You do need to base your abuse on something more than your right-wing
>> >> >intuition. Going off half-cocked like this does rather expose your
>> >> >enthusiasm for believeing what you want to believe despite the absence
>> >> >of any evidence to support your point of view.
>>
>> >> All you do here is claim how "expert" you are, or maybe were, without
>> >> ever making actual contributions.
>>
>> >Nothing that you'd be willing to acknowledge, particularly since most
>> >of my contributions are references to the publshed literature, a
>> >source that you seem ill-equipped to exploit
>>
>> >> You never *do* anything.
>>
>> >Not at the moment, and I find it frustrating.
>>
>> >> And when I don't have convincing evidence, I experiment and collect
>> >> some. A mouse isn't a soldering iron.
>>
>> >You burble about 140dB of ripple rejection
>>
>> Burble? I questioned whether the Spice models of the c-multiplier were
>> accurate at mid-frequencies. None came close to 140, or even 80, dB at
>> frequencies where Early slope matters. So I tried some experiments.
>>
>> and then post the results
>>
>> >of your experiment that shows 67dB. You've got a lab full of expensive
>> >equipment - if we are to take your boasting seriously - and boast like
>> >a lion about your "insanely good" electronic designs, but then post a
>> >result that might qualify you as a mouse in an undemanding
>> >environment.
>>
>> This is the sort of thing you could do. It doesn't take expensive
>> equipment to measure, say, 80 or 100 dB of ripple rejection. Just a
>> little thought and patience.
>
>Not to mention the transistors, the capacitors and the resistors - I
>don't keep a stock of components, and while I could get them, there is
>a mininum order charge which menas that I'd better order all of what
>I'd need in one hit
>
>> So try it - for real - and we'll compare numbers.
>
>I'm still trying to use the gEDA circuit design software to prepare a
>schematic - and the concommitant parts list - for my oscillator
>project, which is more interesting, if still not interesting enough to
>motivate me to spend enough time rooting around in the gEDA
>hierachical design system.
>
>> If not for money, just to keep your mind from rotting.
>
>My mind is fine, but my motivation isn't what it could be.
>
>> >Somebody with your experience ought to be aware that if you want 140dB
>> >of attenuation you are going to need more than a single stage of
>> >filtering - stray impedances usually make it difficult to get more
>> >than 60dB in a single stage.
>>
>> My experience was entirely inadequate to make such a pronouncement.
>
>It probably isn't, but you clearly didn't think hard about the earlier
>experience you were getting while you were getting it.
I've always ignored Hre up to now, because it's generally swamped by
other parameter variations. I certainly never had a reason to map Hre
vs Ie and Vce.
The Fairchild 2N3904 datasheet actually has a curve, sadly only at 10
volts.
>
>> I did some tests. Now I know more than I did before. You don't
>> approve?
>
>It is certainly a step in the right direction - but one that you might
>have taken a few decades earlier. In your case, it obviously needs the
>right sort of customer asking the right sort of questions to get you
>interested. People with academic curiousity tend to get more out of
>the experience they have had because they put more time into making
>sense of their results. They often find that some academic had written
>up the whole issue a few decades earlier. It's hard to avoid re-
>inventing the wheel if you did't know that the wheel existed in the
>first place.
I don't have a lot of "academic curiousity"... I'm motivated by
solving real problems. I've never before explored nanovolt-noise-level
LDO regulators because I'd never needed one. Given an essentially
infinite list of things to explore, I'd just as soon pick the ones
that are potentially useful. Exceptions occasionally made for
super-cool stuff, of course.
John
John Larkin
Children are cute as a defense against their parents strangling them.
Then they become teenagers, and are strong enough that you can't
easily strangle them.
John
Spehro Pefhany
"Cute" is a programmed emotional response, triggered in order to cause
you protect, or at least not to harm relatively harmless creatures you
share DNA with. Baby snakes are perceived as less cute than puppies,
which are in turn perceived as less cute than human babies. Dogs may
have a different perspective on the matter.
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Mon, 7 Jun 2010 08:41:20 -0700 (PDT),Bill Sloman
>
>
>
> <bill.slo...@ieee.org> wrote:
> >On Jun 7, 3:55 pm, John Larkin
> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> >> On Mon, 7 Jun 2010 02:16:57 -0700 (PDT),Bill Sloman
>
> >> <bill.slo...@ieee.org> wrote:
> >> >Modern wines are often made to be instantly drinkable. This doesn't
> >> >mean that they can't get better with age, but many don't. If the wine-
> >> >maker has a gas-chromatograph - ideally with a tandem mass-
> >> >spectrometer as a detector - they can have a very clear idea of what
> >> >their wine tastes and smells like, and the good ones know how to
> >> >create an attractive mixture of sensations, and have a pretty good
> >> >idea of how that mixture will change and develop with time,
>
> >> Can a GC tell that the vines grew in water that had been filtered
> >> through glacial gravel, slate and shale?
>
> >Beats me. Why not ask a wine-maker? In theory, the mineral content of
> >the water could influence the volatile organic compounds that give
> >wine its odour and flavour, but I'd be surprsied if the comment was
> >much more than decorative verbiage.
>
> In other words, delusional fat-headedness. There's a lot of that going
> around. He should be careful about chewing on rocks... it's bad for
> your teeth.
It's not all delusional fat-headedness, though publicists do latch
onto to some of the more impressionistic terms as a basis for the
inevitable chunks of decorative verbiage.
There are wine-tasting terms that most people understand - "slatey"
and "grassy" comes to mind - even though the connection between the
word and the flavour is somewhat arbitrary. My wife describes
particular wines as "pink" by which she seems to mean having a high
concentration of a particular floral ester. If we had a GC we could
probably tell you which floral ester.
People - like you - who don't pay much attention to what they are
drinking, and don't develop a vocabulary of terms that allow them to
talk about the differences between particular wines, defensively
devalue the comments of those who have tasted a wide variety of wines
and can articulate the differences between them.
> >The GC can detect what you can smell, with somewhat more
> >discrimination, allowing a more rational approach to wine-making than
> >was possible in previous centuries, not that the more subjective
> >approach couldn't produce great wines.
>
> So the GC knows more about what wines I like than I do?
The GC doesn't know anything. The wine-makers who look at the results
of gas chromatic analysis of their wines (and those of others) know a
great deal more about wines than you do, and - given a short list of
wines that you do like - and could probably predict what other wines
you would like a great deal more reliably than you could.
> No, thanks. The Bandit tastes great.
If enough people share your opinion, it will get to be expensive and
hard to find, and you would want to find something that tastes similar
and is less widely known. If you could describe what it tastes like in
terms that a wine critic or a a good wine merchant could understand,
the process of finding an alternative would be easier.
--
Bill Sloman, Nijmegen
John Larkin
Right. Fatheads. Hilarious failures in double-blind tests.
>
>> >The GC can detect what you can smell, with somewhat more
>> >discrimination, allowing a more rational approach to wine-making than
>> >was possible in previous centuries, not that the more subjective
>> >approach couldn't produce great wines.
>>
>> So the GC knows more about what wines I like than I do?
>
>The GC doesn't know anything. The wine-makers who look at the results
>of gas chromatic analysis of their wines (and those of others) know a
>great deal more about wines than you do, and - given a short list of
>wines that you do like - and could probably predict what other wines
>you would like a great deal more reliably than you could.
>
>> No, thanks. The Bandit tastes great.
>
>If enough people share your opinion, it will get to be expensive and
>hard to find, and you would want to find something that tastes similar
>and is less widely known. If you could describe what it tastes like in
>terms that a wine critic or a a good wine merchant could understand,
>the process of finding an alternative would be easier.
No, There are lots of inexpensive, great-tasting wines. Cynthia, the
wine buyer down at Canyon Market, puts out stacks of good, cheap
stuff.
John
Bill Sloman
<jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
> On Tue, 8 Jun 2010 04:47:14 -0700 (PDT),Bill Sloman
Fatheads do fail ludicrously on double-blind tests, but blind tasting
is popular amongst wine buffs, and some people do remarkably well.
Some of their success has nothing to do with the taste and smell of
the wine - one of my wife's colleagues paid particular attention to
the shapes of the bottles, which provide useful extra information and
someone who knows what the wine merchants have been pushing in the
week or so before a wine tasting can have a pretty fair idea of what
might being offered at a blind tasting - but people who know wine can
get a lot from the aroma and taste.
> >> >The GC can detect what you can smell, with somewhat more
> >> >discrimination, allowing a more rational approach to wine-making than
> >> >was possible in previous centuries, not that the more subjective
> >> >approach couldn't produce great wines.
>
> >> So the GC knows more about what wines I like than I do?
>
> >The GC doesn't know anything. The wine-makers who look at the results
> >of gas chromatic analysis of their wines (and those of others) know a
> >great deal more about wines than you do, and - given a short list of
> >wines that you do like - and could probably predict what other wines
> >you would like a great deal more reliably than you could.
>
> >> No, thanks. The Bandit tastes great.
>
> >If enough people share your opinion, it will get to be expensive and
> >hard to find, and you would want to find something that tastes similar
> >and is less widely known. If you could describe what it tastes like in
> >terms that a wine critic or a a good wine merchant could understand,
> >the process of finding an alternative would be easier.
>
> No, There are lots of inexpensive, great-tasting wines. Cynthia, the
> wine buyer down at Canyon Market, puts out stacks of good, cheap
> stuff.
There are lots of inexpensive good-tasting wines. "Great" implies
something more, and great wines don't stay cheap and available for
very long. The invisible hand of the market guarantees that. Cynthia
will tell you all about it if you ask her nicely.
--
Bill Sloman, Nijmegen
Richard Henry
k...@att.bizzzzzzzzzzzz
wrote:
The three 'C's. Clarity, Cost, and Condesention.
Bill Sloman
<k...@att.bizzzzzzzzzzzz> wrote:
> On Tue, 8 Jun 2010 14:26:10 -0700 (PDT), Richard Henry <pomer...@hotmail.com>
Condescension. In fact, blind tasting of wines is aimed at identifying
the wine being tasted. Which grape it was made from, which country it
was grown in, which area in that country, which year?
Some wines are easier than others. Bordeaux from the Margaux area has
a characteristic scent of violets which even quite unsophisticated
palates can detect.
http://www.thebordeauxcellar.co.uk/page8.htm
--
Bill Sloman, Nijmegen
JosephKK
<ghe...@teachspin.com> wrote:
>On Jun 2, 10:55 am, John Larkin
><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> On Wed, 2 Jun 2010 07:32:55 -0700 (PDT), George Herold
>>
>>
>>
>>
>>
>> <gher...@teachspin.com> wrote:
>> >On Jun 1, 5:51 pm, John Larkin
>> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> >> On Tue, 1 Jun 2010 11:35:59 -0700 (PDT), George Herold
>>
>> >> <gher...@teachspin.com> wrote:
>> >> >On May 31, 12:56 pm, John Larkin
>> >> ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote:
>> >> >> On Mon, 31 May 2010 12:23:10 GMT, jimsl...@esterlux.com (Jim Slone)
>> >> >> wrote:
>>
>> >> >> >What are the best options for high quality audio white noise
>> >> >> >generation?
>>
>> >> >> >I have been using generic diodes and reversed biased transistors. Then
>> >> >> >someone mentioned there are special parts available with better
>> >> >> >characteristics.
>>
>> >> >> >Can anyone please give me a pointer?
>>
>> >> >> >Jim Slone
>>
>> >> >> You can buy noise diodes from lots of people... just google <noise
>> >> >> diode>
>>
>> >> >Does anyone know what makes a high price "noise diode" any better than
>> >> >your garden variety Zener?
>>
>> >> Probably a very small junction area (for low capacitance, high current
>> >> density) and maybe some doping profile. Not a power device!
>>
>> >> Regular zeners get spikey and asymmetric and sort of oscillate at low
>> >> current. You can get noise diodes that behave at low currents.
>>
>> >> >> If you want really flat, really gaussian noise, a mathematical random
>> >> >> stream (single-bit) or random word (dac) generator is probably best.
>> >> >> See AoE for details.
>>
>> >> >> For audio, it doesn't matter much. A 10-volt zener biased at a few mA
>> >> >> is fine.
>>
>> >> >Yup, and if the voltage asymmetry is a problem you can add the signal
>> >> >from two diodes, one biased from the positve supply and the other from
>> >> >the negative. (Though I've never tried this trick.)
>>
>> >> Or sum the signals from a bunch of them. Central limit theorem.
>>
>> >Well that is not going to get rid of the voltage asymmetery.
>>
>> >If you need real Gaussian noise you can look at the shot noise from a
>> >photodiode illuminated by an LED. Gives you noise ~100 times bigger
>> >than the johnson noise of the sense resistor. (Assuming a 5 Volt DC
>> >drop across R). But this has one big drawback. It's very sensitve to
>> >vibrations.
>>
>> Shot noise is the ultimate asymmetric waveform. It's made of
>> single-photon unidirectional spikes. If it manages to be Gaussian,
>> it's because a lot of asymmetric signals are being summed. Central
>> limit theorem.
>>
>> http://en.wikipedia.org/wiki/Illustration_of_the_central_limit_theorem
>>
>> John- Hide quoted text -
>>
>> - Show quoted text -
>
>Hmmm, you are right.... I still don't think that summing the voltage
>noise from a bunch of unipolarized zeners is going to get rid of the
>voltage assymetry. But I'd be happy to be wrong too. Have you ever
>tried this? It would be simple enough to put 5 or 6 together and see
>what the output looks like. (As long as you don't mind my summing
>with an opamp)... Maybe I can find some 'fun' time on Friday.
>
>George H.
Actually differencing them in twos, then summing seems more likely to
reduce the asymmetry.
JosephKK
>The math says it must be so. Still, the sum would converge to Gaussian
>faster if half of the lopsided signals were inverted.
And that is where you tripped yourself up. Sums of lopsided signals are
still lopsided. Differences may work better.
>
>Zener noise gets more symmetric at higher currents. 10 mA is usually
>OK for a small 10-volt zener.
Direct from the nature of the I-V curve.
>
>I sometimes generate Gaussian-distributed numbers by summing a bunch
>of RAN() calls, which are uniform on [0,1]. Six to ten works well, and
>the crest factor is finite and known.
>
>John
>
John Larkin
Do I now have to explain to you what the word "inverted" means?
>>
>>Zener noise gets more symmetric at higher currents. 10 mA is usually
>>OK for a small 10-volt zener.
>
>Direct from the nature of the I-V curve.
I don't see that. Please explain.
John
John Larkin
When did addition stop being associative?
John
MooseFET
Perhaps when you do it with an op-amp.
(10+10)-(10+10) clips
(10-10)+(10-10) doesn't
That would be a whole lot of noise spiking :)
>
> John
George Herold
> On Wed, 02 Jun 2010 10:46:33 -0700, John Larkin
>
>
>
>
>
Differences will work better, but simple adding works too! The sum
may still be lopsided, but less lopsided than the input. It's the
Gaussian nature of shot noise (very asymmetric if you could look fast
enough.) that made me reassess my previous misconception. (As a
recent ‘convert’ to the central limit theorem I’m perhaps more ardent
than those with long held ‘belief’.)
George H.
>
>
>
> >Zener noise gets more symmetric at higher currents. 10 mA is usually
> >OK for a small 10-volt zener.
>
> Direct from the nature of the I-V curve.
>
>
>
>
>
> >I sometimes generate Gaussian-distributed numbers by summing a bunch
> >of RAN() calls, which are uniform on [0,1]. Six to ten works well, and
> >the crest factor is finite and known.
>
> >John- Hide quoted text -
>
> - Show quoted text -- Hide quoted text -
John Larkin
wrote:
The signals (zener noise) are mildly asymmetric noise, not DC values.
When you add noise, it doesn't matter what order you add or subtract
them in.
And besides, a true Gaussian distribution has probability tails to
infinity; the opamps have to clip once in a while. A rational designer
would just makes sure it doesn't happen often. [1]
JKK is so eager to be right, and for others to be wrong, he makes a
lot of stupid statements. He's not the only one.
Allow me to pass on what Miss Denton taught me in 6th grade:
Check your work.
John
[1] summing computer-generated random numbers results in a
distribution with finite peaks, basically Gaussians with their tails
clipped. That can be handy.