High Resolution AD Converter Noise

A customer is using a ready off the shelf AD Converter board using 8 differential channels with a resolution of 16 bit @ 100 khz. The circuit contains 8 parallel switched capacitor SAR (no analog mux) from Analog Devices AD976. It is using the internal reference and the input circuit is a INA103 from Burr Brown. With gain=1 the input range is +-10Volt.

Now customer complains about unknown noisy results (of about 20 LSB) using that board. To examine this, I use the board at my table in clean environment. After connecting all differential ends to ground, I can easily see 10 diffrent numbers at a simple 100mS refreshed display what is equal to an error of 10 LSB. With a gain of 1, the INA103 can hardly cause this noise.

My last AD design was long ago with only 12 bit resolution at lower speed but I was able to see with a 10gear poti any number at the screen without its neighbour and it was even possible to adjust the poti to the edges where the result toggled to the upper or to the lower neighbour.

I can hardly believe the AD976 is poor becouse the noise is much above the INL what is specified with +-2LSB max. I have no idea how the given "transition noise" of 1 LSB is defined in the data sheet. The gausian histogramm shows only 7 codes for DC code uncertainty. 4 of them are extremly rare and only 2 of them make about 2/3 of the results.

Assume all, I think the board layout or grounds are poor and we meassure some crosstalk of the bus or the close DC/DC converters. Therefore it would be nice to hear your experience with AD convertes at similar resolution and maybe someone already examined a 16 bit or above reference board with its own design?

Reply to
Jürgen Veith
Loading thread data ...

t

is

ing

sily

al

is

ed

ut

the

e

ould

and

ts

Switched capacitor ADCs require a quite strong drive to overcome these input capacitances.

Rene

--=20 Ing.Buero R.Tschaggelar -

formatting link
& commercial newsgroups -
formatting link

Reply to
Rene Tschaggelar

"Jürgen Veith" a écrit dans le message de news:42add359$0$27794$ snipped-for-privacy@newsread2.arcor-online.net... ...

would

and

the power supply for the INA103 & the AD976 had to be really clean for a fast 16bits. you talk of DC/DC convertors; are they powering the op-amp & converter..? there should be a linear regulator between them and good filtering and analog ground. and the analog and digital parts on the board should be on separate location on the board..

Reply to
Fred*

Check for power supply noise. Power supply noise rejection figures given for op-amps often apply at low frequencies and noise from (e.g.) DC-DC converters appears on the output. Are you limiting the bandwidth of the signal out of the op-amp? That might be worth a try.

Look on the bright side - 10DN of peak-to-peak noise is only the bottom four bits and is 3.6 DN rms. You've got 12 good bits of data.

Cheers TW

Reply to
Ted

No, it is not. Having a cap over the feedback doesn't limit the noise bandwidth. A bandwidth limit has to be external, eg as RC following the OPAmp.

Rene

--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net
Reply to
Rene Tschaggelar

If the board is off-the-shelf, please specify which board it is and whether the schematic diagram is available on line. In the best case, there may be someone here who has experience with that particular board.

The DC-DC converters can be a problem---especially if they are running their oscillators at a few hundred KHz and you are collecting data at the same rate.

If you don't need 100KHz sampling rates, you could oversample and use the average to reduce noise. This will work if you are seeing noise with a Gaussian distribution. If the noise is nearly synchronous due to the DC-DC converters, oversampling may not work as well.

I've used the AD974 (AD976 with 4x front end mux) and gotten about

2 LSBs noise (standard deviation of 1000 samples) with a nice Gaussian distribution of the samples.

Mark Borgerson

Reply to
Mark Borgerson

You can always use a properly designed common mode choke filter on the supply lines to make sure you dont get noise around frequencies of interest.

Reply to
Correlious

How long are the wires connecting all these differential inputs from the onboard circuitry, via the the normal connector on an off the shelf board to your connection scheme?

Leaving aside noise inside a PC, what is the possibility of noise in the wiring externally.

I would actually consider rerunning the the experiment with a special connector to the off the shelf board with connector having each input shorted to itself (i.e. +ve to -ve) or via a 10R resistor, then shielding the connector with a connection to frame ground at least.

With these low values on a 16bit convertor without knowing the exact lengths connections and intermediate connections and other influences small levels of noise can be an issue.

What gain settings are the conversions done at? Because quite a lot of these cards have front end gain amplification sometimes as high as x1000.

My other issue is connecting all the differential inputs together leaves possiblities for ground loops defeating the purpose of a differential input. You have made a differential input effectively into a single ended input with 0V input.

Does the card have an EMI screened A/D converter section on the board?

This is from my working over 10 years ago with PC add on cards for A/D converters and video cards for a well known company in that field at the time.

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Click to see the full signature
Reply to
Paul Carpenter

See

formatting link
which is a white paper on the subject of dealing with noise issues in measurement systems. Also worth looking at is "Low Level measurements Handbook - 6th Edition". Also, if you can get on one of their free seminars on the topic you will find them quite worthwhile.

--
********************************************************************
Paul E. Bennett ....................
 Click to see the full signature
Reply to
Paul E. Bennett

The original poster may well do. My comments are based on how many add on cards have "screw terminal panels" that are PCBs on at least 0.5m of often unshielded cable to the off the shelf card. Then people ground all the "screw terminals" and with at least a 1m ground loop, wonder why they get so much noise on low level readings with 16bit A/D. Once I remember a customer complaining of the noise levels being very high, when he was doing something similar at a gain level of x100 (effective 16bit range of +/-0.1V).

--
Paul Carpenter          | paul@pcserviceselectronics.co.uk
    PC Services
 Click to see the full signature
Reply to
Paul Carpenter

I didn't say anything about putting caps on feedback. Perhaps I should have said "limiting thebandwidth of the signal into the ADC"

TW

Reply to
Ted

BTW...this may come in handy for some, but I had a nosiy ADC to deal w/. I spent 2 weeks on it and tried every single source. It turned out that one of the ground pins on the SOIC package had a cold solder joint that looked fine even under a microscope. Just running the soldering iron over the connection reflowed the solder and everything was fine.

The bad connection on one of the ground pins, meant high inductance to ground..basically lots of local ground bounce.

Reply to
Correlious

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.