opamp sine wave oscillator

[Joerg]

How does booze "[go] bad"?

No idea but...

Actually I think Rich was asking how the booze doesn't get
consumed immediately it's in your possession, instead being
kept for long enough for *anything* else to happen to it

Well, visitors mostly drank it. Before leaving Europe for good we did
throw one heck of a good-bye party for neighbors and friends but they
all wanted beer and wine.

 

[Jim Thompson]

What sort of harmonic distortion does a speaker have? I've seen
numbers like 6% called "good."

John

For harmonic distortion, 0.1% is probably below what most people can
hear.

Personally I think TIM (Transient Inter-Modulation) from crossover
dead-band is more noticeable... particularly when a French Horn and a
Clarinet, in a Mozart woodwind ensemble, mix :-( At least 35 years
ago I demonstrated that a nearly ideal class-AB output stage could be
obtained, with much reduced TIM. I couldn't get anyone interested in
funding a chip. Maybe in my retirement I'll run a batch thru MOSIS
and then run it up the flag-pole ;-)

...Jim Thompson

 

[Vladimir Vassilevsky]

A very familiar picture

- Our device is more accurate then HP!
- Wow! How did you verify that?
- We verified that with HP!

What sort of harmonic distortion does a speaker have? I've seen
numbers like 6% called "good."

It largely depends on the excursion of the cone, which is higher at low
frequencies. Can be anywhere from ~10% at low to 0.1% or so at the
middle to high frequencies. For the speakers, the IMD is much more
representative parameter then the THD.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

 

[John Devereux]

What sort of harmonic distortion does a speaker have? I've seen
numbers like 6% called "good."

And what about the human ear?

 

[Rich Grise]

How does booze "[go] bad"?

No idea but...

Actually I think Rich was asking how the booze doesn't get
consumed immediately it's in your possession, instead being
kept for long enough for *anything* else to happen to it

Heh. Actually, I wasn't thinking that at all - if the seals
leaked, the alcohol could evaporate, and I can see wine turning
bad, but I've never considered wine (or beer) to be "booze."

And, of course, there's some truth there - it's why I don't try to get
into microbrewing or bootlegging - I don't know if I could make the stuff
as fast as I could drink it! ;-D

Cheers!
Rich

 

[Rich Grise]

Rich Grise wrote:
How does booze "[go] bad"?
No idea but...

Actually I think Rich was asking how the booze doesn't get
consumed immediately it's in your possession, instead being
kept for long enough for *anything* else to happen to it

Well, visitors mostly drank it. Before leaving Europe for good we did
throw one heck of a good-bye party for neighbors and friends but they
all wanted beer and wine.

Vodka, brown sugar, and seltzer tastes just like root beer. ;-)

Cheers!
Rich

 

[David Harmon]

On Tue, 05 Feb 2008 11:55:13 -0500 in sci.electronics.design, Phil

That's usually cad sulphide (CdS), cad selenide (CdSe), or an alloy of
the two. They're very sensitive, adequately slow for lots of things,

Maybe for the ROHS version you could use a light bulb.

 

[Clifford Heath]

if the seals
leaked, the alcohol could evaporate, and I can see wine turning
bad,

Actually the alcohol oxidizes rather than evaporates. My son just
had to chuck out a 25L batch of home-brew that he left in the vat
for too long without bottling it, and the seal had failed. It
tasted pretty bad! So the same happens with beer. It's why wine
makers have turned to screw-caps...

 

[Winfield]

Win, How was that measured?

My old hp distortion analyzer (can't remember the model
number) only worked down to about 0.001% (IIRC).

IIRC, the hp generators have worse specs than that,
although some may be able to measure better than
their spec, e.g., to 10ppm. I have three Krohn-Hite
distortion analyzers, purchased on eBay, that can do
quite a bit better than that, 3 - 4ppm, IIRC, but our
oscillator's 2ppm distortion was measured on Paul's
xxx analyzer, which he got at the Flea at MIT. I'll
get the detailed info and post it. I don't know how
well the better Audio Precision distortion analyzers
can do, but I think it's very well indeed.

 

[MooseFET]

Interesting. We didn't choose to do that,
and instead placed the JFET directly in the
DC signal pathway. Hmm, would you say we
could have further reduced our distortion
below the 2ppm = 0.0002% we observed?

I think you may be down at the limit of the opamps and resistors etc.
For that matter this sounds like it also down at the limits of the
measurement system.

 

[MooseFET]

0.0002% sounds like the -112 dB THD+N, at least for audio frequencies, of
really expensive products (but really cool, I'm working with one at a
clients site) like this one:http://ap.com/products/2700.htm

I'd try this guy since I can get my hands on it.

http://www.thinksrs.com/products/SR785.htm

 

[MooseFET]

MooseFET wrote: [...]

!
=== C1
!
!------+
Control-->! !
!-- === C2
! !
GND GND

It divides down the amplitude the JFET sees and C2 also tends to short
out the harmonics.

Cute--that's like tapping a varactor down on a tank circuit. With an op
amp it isn't hard to know the total range of gains you can have, so you
can use a vernier control and get better nonlinearity.

The down side is that the frequency of operation is more sensitve to
component values. BTW

 

[Phil Hobbs]

MooseFET wrote: [...]

!
=== C1
!
!------+
Control-->! !
!-- === C2
! !
GND GND
helps.
It divides down the amplitude the JFET sees and C2 also tends to short
out the harmonics.

Cute--that's like tapping a varactor down on a tank circuit. With an op
amp it isn't hard to know the total range of gains you can have, so you
can use a vernier control and get better nonlinearity.

The down side is that the frequency of operation is more sensitve to
component values. BTW

That's because the phase shift depends on the control voltage. If you
used two resistors instead, you could get the same effect, minus the
couple of dB worth of harmonic rolloff, and no significant phase shift.

Cheers,

Phil Hobbs

 

[John Larkin]

I think you may be down at the limit of the opamps and resistors etc.
For that matter this sounds like it also down at the limits of the
measurement system.

I'd imagine that most capacitors add distortion, too.

John

 

[Winfield Hill]

I'd imagine that most capacitors add distortion, too.

Our 2ppm may well be coming from the caps.

 

[Satoru Uzawa]

I think you may be down at the limit of the opamps and resistors etc.
For that matter this sounds like it also down at the limits of the
measurement system.

Touru Kuroda in Japan published a low distortion oscillator with
JFET and opamp. He achieved 0.000064% distortion with a 2SK30A (a
jelly bean JFET in Japan, very cheap there) withtwo NE5532 and a
TL071. So 5 opamps not 2 like Win's design. Kuroda used 1:1 resistor
divider instead of capacitors like this. I wonder this idea could
reduce the distortion in Kuroda's circuit, too. Hmmm, interesting.
Oh, the outputfrequency of the actual assembly was 995Hz and the
output was 7.04Vrms with 0.000064% distortion.

By the way, the oscillator was published in a book titled "First
hands on Transistor Circuit Design", a book intended for a beginner.
A sine wave oscillator with -124dBc distortion (0.000064%) for a
beginner?? Well, you judge.
The book contained a full PCB pattern so a lot of people in Japan
made this oscillator. I don't know how many of us could measure
the actual distortion of the oscillator, though.

Satoru

 

[Robert Latest]

Interesting. We didn't choose to do that,
and instead placed the JFET directly in the
DC signal pathway. Hmm, would you say we
could have further reduced our distortion
below the 2ppm = 0.0002% we observed?

speaking of distortion measurements, how does that work, anyway? Feed a
"zero-distortion" signal into the DUT, subtract the output from the
input, and what's left is the distortion? I'd image gain and phase
adjustment finicky, to say the least. Or some synchronous demodulation
technique -- the same thing in the frequency domain, essentially an
ultra-narrow bandpass. What's not inside the delta transfer function has
got to be distortion.

So how is it done?

robert

 

[Fred Bartoli]

Winfield Hill a écrit :

Our 2ppm may well be coming from the caps.

Don't know but I've stressed some philips leaded 500V-100pF NP0 at 1kV
while measuring the cap deviation which stayed under the bridge
resolution, ie under 0.001% (not discernible from the normal last digit
flip).

Using these at normal IC voltages would certainly ensure sub ppm cap
induced distortion.

At this level, lots of weird things can happen from class B operation
and 'hidden' components (PCB, parasitics) interaction.

 

[Arie de Muynck]

It largely depends on the excursion of the cone, which is higher at low
frequencies. Can be anywhere from ~10% at low to 0.1% or so at the middle
to high frequencies. For the speakers, the IMD is much more
representative parameter then the THD.

I still wonder if sending e.g. 100 Hz and 5 kHz at the same time to a
speaker wouldn't give a 100Hz FM modulation of the 5 kHz tone (doppler
effect caused by the cone movement).
But maybe the microphone has an inverse effect?
But then, what about a pressure sensor as microphone (assuming no diafragm
movement)?

Arie de Muijnck

 

[Vladimir Vassilevsky]

Robert Latest wrote:

speaking of distortion measurements, how does that work, anyway? Feed a
"zero-distortion" signal into the DUT, subtract the output from the
input, and what's left is the distortion? I'd image gain and phase
adjustment finicky, to say the least. Or some synchronous demodulation
technique -- the same thing in the frequency domain, essentially an
ultra-narrow bandpass. What's not inside the delta transfer function has
got to be distortion.

First, the fundamental should be attenuated by a notch or lowpass filter
so the distortion of the instrument itself would not affect the result.
After that we can measure the harmonics by the synchronous detector or
some other technique. The bandwidth of the detector determines the noise
floor, and it is determined by the short term frequency stability of the
oscillator, which is not so good for the Wein bridge.

For the claimed distortion levels below -120dB, the notch should
attenuate the fundamental by at least 20dB or so, and the nonlinear
artifacts of the filter itself should stay below -130dB. I don't think
that is attainable with the active filter. With the passives, the huge
L and C should be used to keep the distortion at minimum.

So how is it done?

Nobody yet provided an attestation of their claims. When it comes to the
THD levels below 0.001%, I have some doubts.


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com