good grief!

[John Larkin]

http://www.analog.com/UploadedFiles/Data_Sheets/272777363AD5933_prb.pdf



John

 

[John Fields]

http://www.analog.com/UploadedFiles/Data_Sheets/272777363AD5933_prb.pdf

---
Yow!!!

I wonder where the gotchas are?

Last time I used them I wound up with some "interesting" power supply
sequencing problems.

 

[John Larkin]

---
Yow!!!

I wonder where the gotchas are?

Last time I used them I wound up with some "interesting" power supply
sequencing problems.

Yeah, the ADI stuff does tend to have latchup quirks. We use one of
their octal dacs and it wound up bristling with schottky diodes to
keep it from suiciding.

But this is cool. Somebody could pop it in a box with some sort of pic
or a pc interface and have a killer R-L-C analyzer.

I wonder what it costs.

John

 

[Bob]

http://www.analog.com/UploadedFiles/Data_Sheets/272777363AD5933_prb.pdf

But this is cool. Somebody could pop it in a box with some sort of pic
or a pc interface and have a killer R-L-C analyzer.

I wonder what it costs.

John

Yeah, or a nifty audio spectrum analyzer (with tracking oscillator).
However, it seems too niche for it to be widely used. I wonder how many
they've sold?

Bob

 

[Nico Coesel]

Yeah, the ADI stuff does tend to have latchup quirks. We use one of
their octal dacs and it wound up bristling with schottky diodes to
keep it from suiciding.

But this is cool. Somebody could pop it in a box with some sort of pic
or a pc interface and have a killer R-L-C analyzer.

That's not really necessary. The device goes up to 100kHz. A proper
soundcard should be able to handle 90kHz / 24bit. You could make a
similar device using a standard soundcard or USB soundcard chipset if
the input circuitry needs to be a bit more fancy than a standard
soundcard offers.

 

[John Woodgate]

I read in sci.electronics.design that Nico Coesel <[email protected]>

That's not really necessary. The device goes up to 100kHz. A proper
soundcard should be able to handle 90kHz / 24bit. You could make a
similar device using a standard soundcard or USB soundcard chipset if
the input circuitry needs to be a bit more fancy than a standard
soundcard offers.

I don't quite see how you could easily turn a sound card into a network
analyser.

 

[Don Pearce]

I read in sci.electronics.design that Nico Coesel <[email protected]>


I don't quite see how you could easily turn a sound card into a network
analyser.

I think it was a spectrum analyser he was talking about.

You can turn it into a network analyser, though and I have done it.
But it isn't easy making the directional bridge couplers. Probably
somebody makes them, but I wanted to do it for nothing.

After that the software for 12 term normalization and calibration
takes some writing.

d

Pearce Consulting
http://www.pearce.uk.com

 

[Nico Coesel]

I read in sci.electronics.design that Nico Coesel <[email protected]>


I don't quite see how you could easily turn a sound card into a network
analyser.

It has an input and an output. Generate a sweep on the output, sample
the signal accross the network, FFT the signal and you'll have a
frequency and phase plot of the network.

 

[John Woodgate]

I read in sci.electronics.design that Nico Coesel <[email protected]>

It has an input and an output. Generate a sweep on the output, sample
the signal accross the network, FFT the signal and you'll have a
frequency and phase plot of the network.

I may be being dense, but it seems that the sound card is just producing
the sweep. To do the f-Z-phi analysis, you need a dual-channel FFT as
well.

 

[Nico Coesel]

I read in sci.electronics.design that Nico Coesel <[email protected]>


I may be being dense, but it seems that the sound card is just producing
the sweep. To do the f-Z-phi analysis, you need a dual-channel FFT as
well.

No, the soundcard should also be used as a sampler at the same time.
The PC does the signal processing and other math.

 

[Joerg]

Hello John,

But this is cool. Somebody could pop it in a box with some sort of pic
or a pc interface and have a killer R-L-C analyzer.

It is cool. But most impedance analysis takes place well above the
100kHz limit. Mine are mostly 3-25MHz, or higher. When measuring
material properties it goes to a GHz easily, which is why that freaking
heavy HP impedance analyzer in the lab still has its place.

This AD chip seems to be geared towards the audio market. I wonder if
that is enough of a market in terms of size.

I wonder what it costs.

Me, too.

Regards, Joerg

 

[Terry Given]

Hello John,



It is cool. But most impedance analysis takes place well above the
100kHz limit. Mine are mostly 3-25MHz, or higher. When measuring
material properties it goes to a GHz easily, which is why that freaking
heavy HP impedance analyzer in the lab still has its place.

This AD chip seems to be geared towards the audio market. I wonder if
that is enough of a market in terms of size.



Me, too.

Regards, Joerg

It might be pretty good for a control system loop analyser. As long as
Fsw < 1MHz or so, 100kHz is fine. The real problem there of course is
getting the signals in and out without killing anyone

Cheers
Terry

 

[Jonathan Kirwan]

Me, too.

An announcement page I saw put it this way:

http://www.eeproductcenter.com/printableArticle.jhtml?printable=true&articleID=60406316

"All devices are currently sampling with production quantities of the
CDCs (AD7745 and AD7746) and the IDCs (AD5933: 12-bit, 1
Megasamples/s; and AD5934: 12-bit, 250 Kilosamples/s) slated for May.
The AD7747 CDC is currently sampling with production quantities
planned for August. The devices are available in small 16-lead TSSOP
and 16-lead SSOP with prices ranging from $4.60 to $4.95 (CDC) and
$4.35 to $6.65 (IDC) per unit in 1,000-piece quantities."

I'd guess this means the AD5933 is on the high side of the IDC's, so
$6.65 in 1000's.

Jon

 

[Joerg]

Hello Jonathan,

I'd guess this means the AD5933 is on the high side of the IDC's, so
$6.65 in 1000's.

That would be a good price. But since the chips are for audio range apps
I am not so sure that there is a large market.

Non-destructive testing could be an additional market for these.

Regards, Joerg