Where to get high speed ADC's and DACs

[DecadentLinuxUserNumeroUn]

On Sun, 01 Sep 2013 23:56:31 +0300, upsidedown@downunder.com Gave us:

>I guess that I am responding to a troll, but anyway.

That retarded crack was a troll, dumbfuck. But anyway, jackass...

I am building your continent's communications infrastructure as we
speak. So you can thank me once it gets put into place, and you guys
are no longer paying that one, singular asshole for your DSL hooks.

Currently building state of the art gateways for New South Wales.

Your continent's powers that be buy from us, because we make the best
gear, and write the most secure operating code known to man.

>I have no real understanding, what -127.5 dBm might mean ?

That much is certain. You cannot even figure out that you DO NOT need
to place a space between your fucked in the head remarks, and the
punctuation, so i hardly think you have the aptitude to solve your
current electronic problems.

GPS signals reside (at the earth based receiver) barely above the
thermal noise floor.

Need more help, dork?

<http://en.wikipedia.org/wiki/DBm>

 

[Lasse Langwadt Christense]

On Monday, September 2, 2013 5:36:42 PM UTC+2, Robert Macy wrote:

On Sun, 01 Sep 2013 10:58:17 -0700, Phil Hobbs

pcdhSpamMeSenseless@electrooptical.net> wrote:



...snip...



Well, if we generously assume that an ADC that fast could have a 5V

input range, 1 LSB at 20 bits is 5 uV, and the quantization noise is

1/sqrt(12) times that, or 1.4 uV. In a 100 MHz bandwidth, that's

140 pV/sqrt(Hz). In real life, the input structure would have to be

several times faster than that in order to settle to that accuracy in

the time available, putting the maximum input noise down in the

50 pV/sqrt(Hz) range, not counting the effects of input capacitance.



Good luck with that.



Cheers



Phil Hobbs





Irritated yelling mode...You have just demonstrated once again WHY I

re-derive EVERYTHING! I NEVER trust 'cookbook' equations, especially after

getting severely burnt by an article in EDN showing 'cookbook' values for

a simple 'what's its name? filter [the simple 5-pole low pass type using

two 2N3904's in series]. Whereupon, I was forced to rederive ALL the

values for both Butterworth AND Tschebyshev(sp?) *and* using an HP

calculator with reverse polish input (spit, spit, curse begone!) I

'optimized' a response to obtain values and voila! worked. But that little

effort caught me on a late Friday [deadline Monday morning] to make a

filter that worked! All weekend!



I HATE PACKAGED FORMULAS!!! I have NO idea what these numbers you gave me

should mean to me.



Now back to quiet mode...Thank you for providing 'numbers', although I

have NO idea what they mean, nor how they relate to what I'm doing. I just

checked and found that *if* I use 16 bit 10MHz ADC's in my system, the

system will be next to useless. *IF* I can get 22 bits, it will work

almost as well as the previous system. Probably live with 20 bits, but the

performance is going to suffer.



Digitization noise dominates in my system, NOT the noise density function.

Front end can be lousy at 2nV/rtHz, prefer 1nV/rtHz [50 ohm system], BW is

cutoff at 20MHz, digitize -1V to +1V to 20 bits at 10MS/s the system will

just barely make it. If could digitize to 22 bits, the system will work

acceptibly!



So back to...



Who makes a 20+ bit ADC with 10MS/s capability?

I probaly missing something but what would be the point of
make a 24 bit adc if the 8 lsb will be noise at that
bandwidth in a 50R system?

-Lasse

 

[RobertMacy]

On Sun, 01 Sep 2013 10:58:17 -0700, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...

Well, if we generously assume that an ADC that fast could have a 5V
input range, 1 LSB at 20 bits is 5 uV, and the quantization noise is
1/sqrt(12) times that, or 1.4 uV. In a 100 MHz bandwidth, that's
140 pV/sqrt(Hz). In real life, the input structure would have to be
several times faster than that in order to settle to that accuracy in
the time available, putting the maximum input noise down in the
50 pV/sqrt(Hz) range, not counting the effects of input capacitance.

Good luck with that.

Cheers

Phil Hobbs

Irritated yelling mode...You have just demonstrated once again WHY I
re-derive EVERYTHING! I NEVER trust 'cookbook' equations, especially after
getting severely burnt by an article in EDN showing 'cookbook' values for
a simple 'what's its name? filter [the simple 5-pole low pass type using
two 2N3904's in series]. Whereupon, I was forced to rederive ALL the
values for both Butterworth AND Tschebyshev(sp?) *and* using an HP
calculator with reverse polish input (spit, spit, curse begone!) I
'optimized' a response to obtain values and voila! worked. But that little
effort caught me on a late Friday [deadline Monday morning] to make a
filter that worked! All weekend!

I HATE PACKAGED FORMULAS!!! I have NO idea what these numbers you gave me
should mean to me.

Now back to quiet mode...Thank you for providing 'numbers', although I
have NO idea what they mean, nor how they relate to what I'm doing. I just
checked and found that *if* I use 16 bit 10MHz ADC's in my system, the
system will be next to useless. *IF* I can get 22 bits, it will work
almost as well as the previous system. Probably live with 20 bits, but the
performance is going to suffer.

Digitization noise dominates in my system, NOT the noise density function.
Front end can be lousy at 2nV/rtHz, prefer 1nV/rtHz [50 ohm system], BW is
cutoff at 20MHz, digitize -1V to +1V to 20 bits at 10MS/s the system will
just barely make it. If could digitize to 22 bits, the system will work
acceptibly!

So back to...

Who makes a 20+ bit ADC with 10MS/s capability?

Who makes a 22-24 bit DAC that can operate this fast?

 

[Lasse Langwadt Christense]

On Monday, September 2, 2013 8:26:50 PM UTC+2, Tim Wescott wrote:

On Sun, 01 Sep 2013 13:57:16 -0700, John Larkin wrote:



On Sun, 01 Sep 2013 10:45:34 -0700, RobertMacy <robert.a.macy@gmail.com

wrote:



On Sun, 01 Sep 2013 09:44:40 -0700, John Larkin

jjlarkin@highnotlandthistechnologypart.com> wrote:



On Sun, 01 Sep 2013 09:33:15 -0700, RobertMacy

robert.a.macy@gmail.com

wrote:



Where do I get 24 bit high speed ADC and DAC systems out to 10MHz?



Or, turn around how much digitization can I get out to 100MHz today?

20 bits?



This is about right:



http://www.linear.com/designtools/hsadcs.php





We use their 250 MHz, 12-bit LVDS ADC and it's pretty good.



https://dl.dropboxusercontent.com/u/53724080/PCBs/ESM_rev_B.jpg



A 20 or 24-bit ADC, at 100 MHz, probably isn't useful. Wideband noise

would trash a lot of LSBs.









John,



You have no idea how much I respect Linear and their products. but I was

talking about 20+bits not the insignificant 12 bit range.



They have 16 bits at 185 Ms/s.





[I recently did 18 bits at 500MHz - Jim Williams would have been proud,

NEVER AGAIN!!!



From simulations, I need 20+ bits, else quantization noise eats me

alive!



At 500 MHz bandwidth (reasonable s/h bw for a 500 Ms/s ADC) a 50 ohm

resistor makes 20 uV RMS Johnson noise, and I doubt that any actual

front-end amp will be anywhere close to that, probably several times

worse. Seems to me that you'll have many LSBs of noise, which may be OK

if downstream processing is essentially narrowband, like some RF stuff.



John's nailed it.



The determining factor is the noise in the front end of the ADC, and

that's going to be much higher than Johnson noise for a monolithic device.



There may be some ultra-boutique hybrid parts out there that can extend

this, but they'll still be limited by the noise in the comparator and the

need for a high bit-count DAC.



If you really need a 144dB full scale to LSB ratio before the noise

starts interfering with your measurement, then at 10MHz you're probably

screwed. If there is some device that can do this, it's almost certainly

not a chip.



But -- what do you really need? Are you really using the full bandwidth

of this thing, or are you extracting some narrower-bandwidth signal out

of it? Averaging the output of an ADC can do wonders for the precision

of the measurement.

just like a deltasigma adc, but you just trade precision for bandwidth


-Lasse

 

[John Devereux]

RobertMacy <robert.a.macy@gmail.com> writes:

On Sun, 01 Sep 2013 10:58:17 -0700, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...

Well, if we generously assume that an ADC that fast could have a 5V
input range, 1 LSB at 20 bits is 5 uV, and the quantization noise is
1/sqrt(12) times that, or 1.4 uV. In a 100 MHz bandwidth, that's
140 pV/sqrt(Hz). In real life, the input structure would have to be
several times faster than that in order to settle to that accuracy
in the time available, putting the maximum input noise down in the
50 pV/sqrt(Hz) range, not counting the effects of input capacitance.

Good luck with that.

Cheers

Phil Hobbs


Irritated yelling mode...You have just demonstrated once again WHY I
re-derive EVERYTHING! I NEVER trust 'cookbook' equations, especially
after getting severely burnt by an article in EDN showing 'cookbook'
values for a simple 'what's its name? filter [the simple 5-pole low
pass type using two 2N3904's in series]. Whereupon, I was forced to
rederive ALL the values for both Butterworth AND Tschebyshev(sp?)
*and* using an HP calculator with reverse polish input (spit, spit,
curse begone!) I 'optimized' a response to obtain values and voila!
worked. But that little effort caught me on a late Friday [deadline
Monday morning] to make a filter that worked! All weekend!

I HATE PACKAGED FORMULAS!!! I have NO idea what these numbers you gave
me should mean to me.

Now back to quiet mode...Thank you for providing 'numbers', although I
have NO idea what they mean, nor how they relate to what I'm doing. I
just checked and found that *if* I use 16 bit 10MHz ADC's in my
system, the system will be next to useless. *IF* I can get 22 bits, it
will work almost as well as the previous system. Probably live with 20
bits, but the performance is going to suffer.

Digitization noise dominates in my system, NOT the noise density
function. Front end can be lousy at 2nV/rtHz, prefer 1nV/rtHz [50 ohm
system], BW is cutoff at 20MHz, digitize -1V to +1V to 20 bits at
10MS/s the system will just barely make it. If could digitize to 22
bits, the system will work acceptibly!

So back to...

Who makes a 20+ bit ADC with 10MS/s capability?

Nobody.

Linear have just released a 20 bit 1MSPS, this is state of the art
AFAIK.

> Who makes a 22-24 bit DAC that can operate this fast?

Don't think they make 22+ bit DACs at all, although it would be possible
in principle.


--

John Devereux

 

[RobertMacy]

On Mon, 02 Sep 2013 08:58:50 -0700, John Devereux <john@devereux.me.uk>
wrote:

RobertMacy <robert.a.macy@gmail.com> writes:
...snip...
So back to...

Who makes a 20+ bit ADC with 10MS/s capability?

Nobody.

Linear have just released a 20 bit 1MSPS, this is state of the art
AFAIK.

Who makes a 22-24 bit DAC that can operate this fast?

Don't think they make 22+ bit DACs at all, although it would be possible
in principle.

Thank you for the info.

 

[Joerg]

RobertMacy wrote:

On Mon, 02 Sep 2013 08:58:50 -0700, John Devereux <john@devereux.me.uk
wrote:

RobertMacy <robert.a.macy@gmail.com> writes:
...snip...
So back to...

Who makes a 20+ bit ADC with 10MS/s capability?

Nobody.

Linear have just released a 20 bit 1MSPS, this is state of the art
AFAIK.

Well, there's the solution

Who makes a 22-24 bit DAC that can operate this fast?

Don't think they make 22+ bit DACs at all, although it would be possible
in principle.



Thank you for the info.

Considering the specs you set out with this seems to be a project of the
budget category "the sky is the limit". So you could take 10 of these
LTC converters, add 10 track&hold circuits if needed, and pipe the
resulting barrage of data to wherever it needs to be crunched.

No kidding, that's how we did high-end ultrasound machines in the days
when 12-bit converters were too slow for the job. If you decide to go
that route and need ways to auto-align them for offset, gain and
aperture jitter let me know because I've BTDT. The ones I've been
involved in had between four and 32 converters ganged. My DSO works the
same way when operating single or dual channel, it has four converters
that are ganged if you don't use all four channels.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[RobertMacy]

On Mon, 02 Sep 2013 09:26:24 -0700, Joerg <invalid@invalid.invalid> wrote:

..snip...

Considering the specs you set out with this seems to be a project of the
budget category "the sky is the limit". So you could take 10 of these
LTC converters, add 10 track&hold circuits if needed, and pipe the
resulting barrage of data to wherever it needs to be crunched.

No kidding, that's how we did high-end ultrasound machines in the days
when 12-bit converters were too slow for the job. If you decide to go
that route and need ways to auto-align them for offset, gain and
aperture jitter let me know because I've BTDT. The ones I've been
involved in had between four and 32 converters ganged. My DSO works the
same way when operating single or dual channel, it has four converters
that are ganged if you don't use all four channels.

sadly not 'sky limit' budget.

I remember those ultrasonic days. Late 70's? I remember the excitement
from getting an 8 bit ADC, much better than the previous 6 bit one, didn't
TRW make it? came in a ?? 64 pin ceramic DIP package, the size of a small
chocolate bar. and ran HOT!


10 S/H's multiplexed into 10 ADC's could work, interesting. although do
need at least two channels.

More detail offline? Or can you share here?

 

[John Devereux]

Joerg <invalid@invalid.invalid> writes:

RobertMacy wrote:
On Mon, 02 Sep 2013 08:58:50 -0700, John Devereux <john@devereux.me.uk
wrote:

RobertMacy <robert.a.macy@gmail.com> writes:
...snip...
So back to...

Who makes a 20+ bit ADC with 10MS/s capability?

Nobody.

Linear have just released a 20 bit 1MSPS, this is state of the art
AFAIK.


Well, there's the solution


Who makes a 22-24 bit DAC that can operate this fast?

Don't think they make 22+ bit DACs at all, although it would be possible
in principle.



Thank you for the info.


Considering the specs you set out with this seems to be a project of the
budget category "the sky is the limit". So you could take 10 of these
LTC converters, add 10 track&hold circuits if needed, and pipe the
resulting barrage of data to wherever it needs to be crunched.

Yes, Vlads Solution, downscaled into the realms of possibility.

No kidding, that's how we did high-end ultrasound machines in the days
when 12-bit converters were too slow for the job. If you decide to go
that route and need ways to auto-align them for offset, gain and
aperture jitter let me know because I've BTDT. The ones I've been
involved in had between four and 32 converters ganged. My DSO works the
same way when operating single or dual channel, it has four converters
that are ganged if you don't use all four channels.

--

John Devereux

 

[Tim Wescott]

On Sun, 01 Sep 2013 14:41:17 -0700, DecadentLinuxUserNumeroUno wrote:

On Sun, 1 Sep 2013 13:43:25 -0700 (PDT), Lasse Langwadt Christensen
langwadt@fonz.dk> Gave us:

On Sunday, September 1, 2013 9:39:41 PM UTC+2,
DecadentLinuxUserNumeroUno wrote:
On Sun, 01 Sep 2013 22:21:15 +0300, upsidedown@downunder.com Gave us:


Look into what GPS receivers use.



They are the lowest level signals we discriminate currently. They

reside just above the noise floor. At -127.5 dBm.



But that is a receiver signal value, not an ADC SNL function.

Still, one might find some pointers by examining what those folks do.

GPS only use a few bits of adc if not only one bit ...

One *still* must pull that bit out of the noise... correctly,
continually.

You have it backwards. The one-bit ADCs used in GPS receivers are
followed by a system that has lots and lots of coding gain. The _signal_
has to be pulled out of the noise correctly and continually, but the ADC
value includes lots of noise.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

 

[Tim Wescott]

On Sun, 01 Sep 2013 13:57:16 -0700, John Larkin wrote:

On Sun, 01 Sep 2013 10:45:34 -0700, RobertMacy <robert.a.macy@gmail.com
wrote:

On Sun, 01 Sep 2013 09:44:40 -0700, John Larkin
jjlarkin@highnotlandthistechnologypart.com> wrote:

On Sun, 01 Sep 2013 09:33:15 -0700, RobertMacy
robert.a.macy@gmail.com
wrote:

Where do I get 24 bit high speed ADC and DAC systems out to 10MHz?

Or, turn around how much digitization can I get out to 100MHz today?
20 bits?

This is about right:

http://www.linear.com/designtools/hsadcs.php


We use their 250 MHz, 12-bit LVDS ADC and it's pretty good.

https://dl.dropboxusercontent.com/u/53724080/PCBs/ESM_rev_B.jpg

A 20 or 24-bit ADC, at 100 MHz, probably isn't useful. Wideband noise
would trash a lot of LSBs.




John,

You have no idea how much I respect Linear and their products. but I was
talking about 20+bits not the insignificant 12 bit range.

They have 16 bits at 185 Ms/s.


[I recently did 18 bits at 500MHz - Jim Williams would have been proud,
NEVER AGAIN!!!

From simulations, I need 20+ bits, else quantization noise eats me
alive!

At 500 MHz bandwidth (reasonable s/h bw for a 500 Ms/s ADC) a 50 ohm
resistor makes 20 uV RMS Johnson noise, and I doubt that any actual
front-end amp will be anywhere close to that, probably several times
worse. Seems to me that you'll have many LSBs of noise, which may be OK
if downstream processing is essentially narrowband, like some RF stuff.

John's nailed it.

The determining factor is the noise in the front end of the ADC, and
that's going to be much higher than Johnson noise for a monolithic device.

There may be some ultra-boutique hybrid parts out there that can extend
this, but they'll still be limited by the noise in the comparator and the
need for a high bit-count DAC.

If you really need a 144dB full scale to LSB ratio before the noise
starts interfering with your measurement, then at 10MHz you're probably
screwed. If there is some device that can do this, it's almost certainly
not a chip.

But -- what do you really need? Are you really using the full bandwidth
of this thing, or are you extracting some narrower-bandwidth signal out
of it? Averaging the output of an ADC can do wonders for the precision
of the measurement.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

 

[Jim Thompson]

On Mon, 02 Sep 2013 10:17:48 -0700, RobertMacy
<robert.a.macy@gmail.com> wrote:

On Mon, 02 Sep 2013 09:26:24 -0700, Joerg <invalid@invalid.invalid> wrote:

..snip...

Considering the specs you set out with this seems to be a project of the
budget category "the sky is the limit". So you could take 10 of these
LTC converters, add 10 track&hold circuits if needed, and pipe the
resulting barrage of data to wherever it needs to be crunched.

No kidding, that's how we did high-end ultrasound machines in the days
when 12-bit converters were too slow for the job. If you decide to go
that route and need ways to auto-align them for offset, gain and
aperture jitter let me know because I've BTDT. The ones I've been
involved in had between four and 32 converters ganged. My DSO works the
same way when operating single or dual channel, it has four converters
that are ganged if you don't use all four channels.


sadly not 'sky limit' budget.

I remember those ultrasonic days. Late 70's? I remember the excitement
from getting an 8 bit ADC, much better than the previous 6 bit one, didn't
TRW make it? came in a ?? 64 pin ceramic DIP package, the size of a small
chocolate bar. and ran HOT!


10 S/H's multiplexed into 10 ADC's could work, interesting. although do
need at least two channels.

More detail offline? Or can you share here?

I was making 8-bit DAC's in the mid-60's, and no trimming required.

...Jim Thompson
--
| James E.Thompson | mens |
| Analog Innovations | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| San Tan Valley, AZ 85142 Skype: Contacts Only | |
| Voice480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.

 

[Joerg]

RobertMacy wrote:

On Mon, 02 Sep 2013 09:26:24 -0700, Joerg <invalid@invalid.invalid> wrote:

..snip...

Considering the specs you set out with this seems to be a project of the
budget category "the sky is the limit". So you could take 10 of these
LTC converters, add 10 track&hold circuits if needed, and pipe the
resulting barrage of data to wherever it needs to be crunched.

No kidding, that's how we did high-end ultrasound machines in the days
when 12-bit converters were too slow for the job. If you decide to go
that route and need ways to auto-align them for offset, gain and
aperture jitter let me know because I've BTDT. The ones I've been
involved in had between four and 32 converters ganged. My DSO works the
same way when operating single or dual channel, it has four converters
that are ganged if you don't use all four channels.


sadly not 'sky limit' budget.

That would mean the proverbial rock and hard spot. "We want creme brulee
but we can only pay a buck fifty".

I remember those ultrasonic days. Late 70's? I remember the excitement
from getting an 8 bit ADC, much better than the previous 6 bit one,
didn't TRW make it? came in a ?? 64 pin ceramic DIP package, the size of
a small chocolate bar. and ran HOT!

In the early 80's ITT made the best ones, and cheap. I used them in my
master's project.

10 S/H's multiplexed into 10 ADC's could work, interesting. although do
need at least two channels.

More detail offline? Or can you share here?

You've got mail.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[Phil Hobbs]

On 9/2/2013 11:36 AM, RobertMacy wrote:

On Sun, 01 Sep 2013 10:58:17 -0700, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...

Well, if we generously assume that an ADC that fast could have a 5V
input range, 1 LSB at 20 bits is 5 uV, and the quantization noise is
1/sqrt(12) times that, or 1.4 uV. In a 100 MHz bandwidth, that's
140 pV/sqrt(Hz). In real life, the input structure would have to be
several times faster than that in order to settle to that accuracy in
the time available, putting the maximum input noise down in the
50 pV/sqrt(Hz) range, not counting the effects of input capacitance.

Good luck with that.

Cheers

Phil Hobbs


Irritated yelling mode...You have just demonstrated once again WHY I
re-derive EVERYTHING! I NEVER trust 'cookbook' equations, especially
after getting severely burnt by an article in EDN showing 'cookbook'
values for a simple 'what's its name? filter [the simple 5-pole low pass
type using two 2N3904's in series]. Whereupon, I was forced to rederive
ALL the values for both Butterworth AND Tschebyshev(sp?) *and* using an
HP calculator with reverse polish input (spit, spit, curse begone!) I
'optimized' a response to obtain values and voila! worked. But that
little effort caught me on a late Friday [deadline Monday morning] to
make a filter that worked! All weekend!

I HATE PACKAGED FORMULAS!!! I have NO idea what these numbers you gave
me should mean to me.

Sorry? Surely anybody designing high performance A/D subsystems knows
that 2**10 ~= 1000, that the RMS quantization noise of an ideal
digitizer is 1/sqrt(12) of the LSB, and that the noise is more or less
white? You can derive it yourself in about three lines.

All I'm saying is:

1. 1 LSB = FSR/2**N

2. 2**20 ~= 10**6.

Therefore, 1 LSB ~= 5 uV.

3. RMS quantization noise is 1/sqrt(12)* 1 LSB ~= 5 uV/3.46 ~= 1.4 uV,
spread out evenly over the Nyquist interval.

4. The Nyquist bandwidth is 50 MHz (not 100), so given that the noise is
white, the quantization noise PSD is 1.4 uV/sqrt(50 MHz) ~=200 pV/sqrt(Hz).

5. To get that many bits to stay reasonably still, your RMS input noise
has to be well below that. Even slow delta-sigmas are hard pressed to
reach a genuine 20 bits, and most of them actually crap out around 18 or
19, AFAICT.

6. IOW, good luck.

That's just engineering rules of thumb.

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510 USA
+1 845 480 2058

hobbs at electrooptical dot net
http://electrooptical.net

 

[Gerhard Hoffmann]

Am 02.09.2013 19:17, schrieb RobertMacy:

I remember those ultrasonic days. Late 70's? I remember the excitement
from getting an 8 bit ADC, much better than the previous 6 bit one,
didn't TRW make it? came in a ?? 64 pin ceramic DIP package, the size of
a small chocolate bar. and ran HOT!

Yes, we used it also:
< http://www.hoffmann-hochfrequenz.de/project_gallery/project_gallery.html >

Third picture from the top, as blurred background for its successor.
Lower right, the big white letters said "TRW". We also got a nekkid one
in a cube of Plexiglas. One could see the reference ladder without
microscope.

That was for testing the inner enclosure of nuclear reactors with
ultrasonics.

10 S/H's multiplexed into 10 ADC's could work, interesting. although do
need at least two channels.

You cannot buy interesting S/Hs nowadays, unless you accept an
ADC on the same chip. So you probably end up at the LTC2209 or its
competitors from AD and TI.

When TI had no 100MHz 16Bit ADC to offer, they published an app note
on averaging some 14 bit ADCs to get better dynamic range. Quite
interesting. I'd really like to try that. It is probably easier
than a time staggered setup since you can shift the sampling window
to a moment when the digital part is quiet.
Also there is much less danger of messing up the sampling clock.
(phase noise, spurs..)

Averaging over a large number of parts works nicely as in
< http://www.hoffmann-hochfrequenz.de/downloads/lono.pdf >

Yes, brute force! Nothing to tune or even select.
And crossing the A to D boundary makes no systematic difference.

BTW something like that could be useful to check your supplies
and reference voltages.

regards, Gerhard

 

[Phil Hobbs]

On 9/2/2013 10:34 PM, RobertMacy wrote:

On Mon, 02 Sep 2013 18:55:03 -0700, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...
On 9/2/2013 9:40 PM, RobertMacy wrote:
On Mon, 02 Sep 2013 13:02:10 -0700, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...
On Sun, 01 Sep 2013 10:58:17 -0700, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...

Well, if we generously assume that an ADC that fast could have a 5V
input range, 1 LSB at 20 bits is 5 uV, and the quantization noise is
1/sqrt(12) times that, or 1.4 uV. In a 100 MHz bandwidth, that's
140 pV/sqrt(Hz). In real life, the input structure would have to be
several times faster than that in order to settle to that accuracy in
the time available, putting the maximum input noise down in the
50 pV/sqrt(Hz) range, not counting the effects of input capacitance.

Good luck with that.

Cheers

Phil Hobbs


Irritated yelling mode...You have just demonstrated once again WHY I
re-derive EVERYTHING! I NEVER trust 'cookbook' equations, especially
after getting severely burnt by an article in EDN showing 'cookbook'
values for a simple 'what's its name? filter [the simple 5-pole low pass
type using two 2N3904's in series]. Whereupon, I was forced to rederive
ALL the values for both Butterworth AND Tschebyshev(sp?) *and* using an
HP calculator with reverse polish input (spit, spit, curse begone!) I
'optimized' a response to obtain values and voila! worked. But that
little effort caught me on a late Friday [deadline Monday morning] to
make a filter that worked! All weekend!

I HATE PACKAGED FORMULAS!!! I have NO idea what these numbers you gave
me should mean to me.

(Context restored)

Sorry? Surely anybody designing high performance A/D subsystems knows
that 2**10 ~= 1000, that the RMS quantization noise of an ideal
digitizer is 1/sqrt(12) of the LSB, and that the noise is more or less
white? You can derive it yourself in about three lines.

All I'm saying is:

1. 1 LSB = FSR/2**N

2. 2**20 ~= 10**6.

Therefore, 1 LSB ~= 5 uV.

3. RMS quantization noise is 1/sqrt(12)* 1 LSB ~= 5 uV/3.46 ~= 1.4
uV, spread out evenly over the Nyquist interval.

4. The Nyquist bandwidth is 50 MHz (not 100), so given that the noise
is white, the quantization noise PSD is 1.4 uV/sqrt(50 MHz) ~=200
pV/sqrt(Hz).

5. To get that many bits to stay reasonably still, your RMS input
noise has to be well below that. Even slow delta-sigmas are hard
pressed to reach a genuine 20 bits, and most of them actually crap out
around 18 or 19, AFAICT.

6. IOW, good luck.

That's just engineering rules of thumb.

Cheers

Phil Hobbs




Sorry? LOL! I actually read that out loud for effect.

(Context restored again)

So apart from posturing, what do you actually disagree with, and why?

Cheers

Phil Hobbs


Don't take umbrage. I don't think I disagree. I'm sure everything you
said was correct, and as you said, they are rules of thumb for
engineering. It's just that nothing there was useful for me. Well, at
first glance seemed like almost nothing. I couldn't see how to apply all
you said to what I'm doing.

In retrospect, not nothing, because you made me THINK. Your comment,
'good luck with that' challenged my thinking. Perhaps, I was wrong in
what I was doing. So I went back to my basic system and had to recreate
all my equations from ground zero back up to try operating at 10-100MS/s
rates. Why ground zero? I don't even trust my own equations. I question
EVERYTHING. So four hours of intensive effort later, I verified a lot
and feel better, thanks. Plus, you triggered a topological approach
using 'available' parts that had NEVER occurred to me until just now.
And, all those recreated equations allowed me to verify [not rigorously,
but at least a bit empirically] that the approach would be possible.
Even better, for my requirements, I might even be able to get more than
24 bits at 100MS/s.

Gone brain dead here, have I sent any images of results using the
present system to you? If not, send me an email address, and I'll send a
couple.

The address in my sig works. You haven't given any details, but very
simple and compelling arguments of the above sort tell me that you
aren't getting anywhere near 22 bits at 100 MHz, unless you're using all
low-TC SQUID-based circuitry.

If I'm wrong about that, and you really have found a way to do it for
real, you're going to be extremely famous in very short order.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510 USA
+1 845 480 2058

hobbs at electrooptical dot net
http://electrooptical.net

 

[RobertMacy]

On Mon, 02 Sep 2013 19:35:34 -0700, John Larkin
<jjlarkin@highnotlandthistechnologypart.com> wrote:

On Mon, 02 Sep 2013 18:40:18 -0700, RobertMacy <robert.a.macy@gmail.com
wrote:
...snip...
In the present 24 bit Data Acquisition system I get something like a
measured 22.5 bits, meaning not quite 23, but better than 22. And, I
couldn't believe I actually ran up to 91% of the Nyquist rate! but I
backed it off down to 89% so didn't see ANY limiting effects.

How are you measuring that 22.5 bits?

As with most of the measurements, made them, continually have their values
verified indirectly while using the system, so don't revisit.

 

[Phil Hobbs]

On 9/2/2013 9:40 PM, RobertMacy wrote:

On Mon, 02 Sep 2013 13:02:10 -0700, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...
On Sun, 01 Sep 2013 10:58:17 -0700, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...

Well, if we generously assume that an ADC that fast could have a 5V
input range, 1 LSB at 20 bits is 5 uV, and the quantization noise is
1/sqrt(12) times that, or 1.4 uV. In a 100 MHz bandwidth, that's
140 pV/sqrt(Hz). In real life, the input structure would have to be
several times faster than that in order to settle to that accuracy in
the time available, putting the maximum input noise down in the
50 pV/sqrt(Hz) range, not counting the effects of input capacitance.

Good luck with that.

Cheers

Phil Hobbs


Irritated yelling mode...You have just demonstrated once again WHY I
re-derive EVERYTHING! I NEVER trust 'cookbook' equations, especially
after getting severely burnt by an article in EDN showing 'cookbook'
values for a simple 'what's its name? filter [the simple 5-pole low pass
type using two 2N3904's in series]. Whereupon, I was forced to rederive
ALL the values for both Butterworth AND Tschebyshev(sp?) *and* using an
HP calculator with reverse polish input (spit, spit, curse begone!) I
'optimized' a response to obtain values and voila! worked. But that
little effort caught me on a late Friday [deadline Monday morning] to
make a filter that worked! All weekend!

I HATE PACKAGED FORMULAS!!! I have NO idea what these numbers you gave
me should mean to me.

(Context restored)

Sorry? Surely anybody designing high performance A/D subsystems knows
that 2**10 ~= 1000, that the RMS quantization noise of an ideal
digitizer is 1/sqrt(12) of the LSB, and that the noise is more or less
white? You can derive it yourself in about three lines.

All I'm saying is:

1. 1 LSB = FSR/2**N

2. 2**20 ~= 10**6.

Therefore, 1 LSB ~= 5 uV.

3. RMS quantization noise is 1/sqrt(12)* 1 LSB ~= 5 uV/3.46 ~= 1.4
uV, spread out evenly over the Nyquist interval.

4. The Nyquist bandwidth is 50 MHz (not 100), so given that the noise
is white, the quantization noise PSD is 1.4 uV/sqrt(50 MHz) ~=200
pV/sqrt(Hz).

5. To get that many bits to stay reasonably still, your RMS input
noise has to be well below that. Even slow delta-sigmas are hard
pressed to reach a genuine 20 bits, and most of them actually crap out
around 18 or 19, AFAICT.

6. IOW, good luck.

That's just engineering rules of thumb.

Cheers

Phil Hobbs




Sorry? LOL! I actually read that out loud for effect.

So apart from posturing, what do you actually disagree with, and why?

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510 USA
+1 845 480 2058

hobbs at electrooptical dot net
http://electrooptical.net

 

[RobertMacy]

On Mon, 02 Sep 2013 18:55:03 -0700, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...
So apart from posturing, what do you actually disagree with, and why?

Cheers

Phil Hobbs

Don't take umbrage. I don't think I disagree. I'm sure everything you said
was correct, and as you said, they are rules of thumb for engineering.
It's just that nothing there was useful for me. Well, at first glance
seemed like almost nothing. I couldn't see how to apply all you said to
what I'm doing.

In retrospect, not nothing, because you made me THINK. Your comment, 'good
luck with that' challenged my thinking. Perhaps, I was wrong in what I was
doing. So I went back to my basic system and had to recreate all my
equations from ground zero back up to try operating at 10-100MS/s rates.
Why ground zero? I don't even trust my own equations. I question
EVERYTHING. So four hours of intensive effort later, I verified a lot and
feel better, thanks. Plus, you triggered a topological approach using
'available' parts that had NEVER occurred to me until just now. And, all
those recreated equations allowed me to verify [not rigorously, but at
least a bit empirically] that the approach would be possible. Even better,
for my requirements, I might even be able to get more than 24 bits at
100MS/s.

Gone brain dead here, have I sent any images of results using the present
system to you? If not, send me an email address, and I'll send a couple.

 

[RobertMacy]

On Mon, 02 Sep 2013 13:02:10 -0700, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

...snip...
Sorry? Surely anybody designing high performance A/D subsystems knows
that 2**10 ~= 1000, that the RMS quantization noise of an ideal
digitizer is 1/sqrt(12) of the LSB, and that the noise is more or less
white? You can derive it yourself in about three lines.

All I'm saying is:

1. 1 LSB = FSR/2**N

2. 2**20 ~= 10**6.

Therefore, 1 LSB ~= 5 uV.

3. RMS quantization noise is 1/sqrt(12)* 1 LSB ~= 5 uV/3.46 ~= 1.4 uV,
spread out evenly over the Nyquist interval.

4. The Nyquist bandwidth is 50 MHz (not 100), so given that the noise is
white, the quantization noise PSD is 1.4 uV/sqrt(50 MHz) ~=200
pV/sqrt(Hz).

5. To get that many bits to stay reasonably still, your RMS input noise
has to be well below that. Even slow delta-sigmas are hard pressed to
reach a genuine 20 bits, and most of them actually crap out around 18 or
19, AFAICT.

6. IOW, good luck.

That's just engineering rules of thumb.

Cheers

Phil Hobbs

Sorry? LOL! I actually read that out loud for effect.

In the present 24 bit Data Acquisition system I get something like a
measured 22.5 bits, meaning not quite 23, but better than 22. And, I
couldn't believe I actually ran up to 91% of the Nyquist rate! but I
backed it off down to 89% so didn't see ANY limiting effects.

I tell youu I'm REALLY impressed with this board! buried inside a PC
WITHOUT shielding! I have more trouble from picking up noise from the SMPS
buried inside the LCD scope sitting next to the breadboard. And don't get
me started on having an old CRT monitor turned on in the lab, ...across
the room.