dsPIC AD converter input impedance

[Joerg]

So I have this project that uses a dsPIC, in this case a
DSPIC33FJ256GP506. Not my choice, and I wouldn't have used it in the
first place but ...

When I wanted to use the ADC I looked around in the datasheet. Not much
there to write home about.

http://ww1.microchip.com/downloads/en/DeviceDoc/70286C.pdf

On the bottom of page 289 in the datasheet it says under "recommended
impedance of analog source voltage": 200ohms max. Ohms! Is this a joke?
What were they thinking?

 

[Joerg]

A lot of high-speed ADCs seem to have roughly similar requirements.
The Analog Devices PulSAR ADCs are shown (in some of AD's papers) as
being drive by either transformers, or dedicated high-speed driver
chips, with perhaps 35 ohms in each leg of the differential drive feed.

I imagine that this level of studly low-impedance drive is required to
allow the ADCs to sample at high rates... the current to charge and
discharge the sampling capacitors at megasample rates isn't small.

If you're working with low frequency ranges and low sampling rates,
you might be able to use a higher driving impedance, and use a
stabilizing capacitor right by the ADC input pin (basically a low-pass
filter arrangement).

That's what I put in there. Essentially it's measuring voltages and I
didn't want to place expensive low offset opamps. The cap puts the
lowpass around 1kHz, should be enough.

I was surprised how little information there was in the datasheet. 320
pages and no meat for the ADC. On the chip itself things are rather
spartan as well. They didn't even spring for a bandgap. At least the
datasheet doesn't mention any.

 

[Joerg]

It's no joke. The low impedance allows for the fast sampling rate
compared to the 8-bit PIC's

Ok, but they could have been a wee bit more detailed. Many times all you
want to do is monitor the voltage on some lines, say, from potmeters.
You don't want 200ohm potmeters and ideally also no buffer amps that
would introduced additional offset errors. They change slowly. So one
would need to know the DC input impedance and then slap a cap across it
so the ADC is happy. But they fail to state a DC input impedance.

 

[nospam]

Microcontroller data sheets are often very rich in detail on the digital
parts, and skimpy on the analog.

The dsPIC33F family reference manual section 16 provides 66 pages of
documentation on the ADC and the 1st paragraph in the ADC section of the
datasheet tells you to look at it.

 

[Joerg]

The dsPIC33F family reference manual section 16 provides 66 pages of
documentation on the ADC and the 1st paragraph in the ADC section of the
datasheet tells you to look at it.

So where does it state the DC input impedance for the DSPIC33FJ256GP506?

 

[[email protected]]

Ok, but they could have been a wee bit more detailed. Many times all you
want to do is monitor the voltage on some lines, say, from potmeters.
You don't want 200ohm potmeters and ideally also no buffer amps that
would introduced additional offset errors. They change slowly. So one
would need to know the DC input impedance and then slap a cap across it
so the ADC is happy. But they fail to state a DC input impedance.

I find that to be a problem with most datasheets now. At one time, datasheets
had schematics equivalent circuits of I/Os, if not the entire device
(admittedly a little tough for a micro ;-).

 

[Joerg]

I find that to be a problem with most datasheets now. At one time, datasheets
had schematics equivalent circuits of I/Os, if not the entire device
(admittedly a little tough for a micro ;-).

They did have that for micros as well. Some mfgs do, some don't. All I
usually need to know is the first 2-3 devices behind a port pin.

 

[Joerg]

If it's a switched-capacitor ADC there probably isn't a DC input
impedance per se. Most likely the closest thing to a "DC input
impedance" is the effective series resistance of the switch in the
sampler, and that's most certainly not going to be on all the time.

That's also what I think. But "likely" doesn't fly for some designs

IME when you call in you get an answer but many times not in writing.
For many of my projects I do need it in writing, and it has to be
something that was properly ECO-released as a document. There have been
many parts during my career that I could not design in because of this.
Sometimes whole series of parts.

 

[Joerg]

I have a feeling that the 200 ohms max is just recommended value at
high sample rate. I have a 10K resistor between the battery and A2D,
although they recommended 5K. My readings are accurate enough.

I also think it'll work. In this case it isn't safety critical, it is
just so we don't deplete a rechargeable battery too far. There is a
secondary shut-off and even if that fails, it isn't a LiIon.

More like a convenience message to the users in my case, along the lines
of "Hey, you've only got enough juice for this many sessions".

 

[Joerg]

True, but the ancillary documentation is often quite complete. I've
never flogged a dead processor for days without ultimately finding the
one improperly set up "screw the processor bit" in some obscure section
of some manual dedicated to some apparently totally unrelated section of
the processor.

(Although it only took me about 15 years to realize that I need to read
EVERY WORD of any document that purports to explain how the system clocks
are distributed).

Lucky you. I had it happen where a whole team assembled at the
manufacturer during the phone conference and then someone at their side
exclaimed "Oh s..t!".

 

[Nico Coesel]

That's what I put in there. Essentially it's measuring voltages and I
didn't want to place expensive low offset opamps. The cap puts the
lowpass around 1kHz, should be enough.

I was surprised how little information there was in the datasheet. 320
pages and no meat for the ADC. On the chip itself things are rather
spartan as well. They didn't even spring for a bandgap. At least the
datasheet doesn't mention any.

Be sure you lie flat on the floor if you read about the ADC ST put in
their STR700 series microcontrollers. For starters (!!) it has an
internal reference voltage with an accuracy around 10% (IIRC). In
short the whole thing is a complete joke. After that I never
considered controllers from ST for any design.

 

[Joerg]

[...]

I was surprised how little information there was in the datasheet. 320
pages and no meat for the ADC. On the chip itself things are rather
spartan as well. They didn't even spring for a bandgap. At least the
datasheet doesn't mention any.

Be sure you lie flat on the floor if you read about the ADC ST put in
their STR700 series microcontrollers. For starters (!!) it has an
internal reference voltage with an accuracy around 10% (IIRC). ...

Probably a glorified diode

... In
short the whole thing is a complete joke. After that I never
considered controllers from ST for any design.

And then there are some Asian uC where the documentation consists of a
few pages, mostly listing local offices and their phone numbers.

I found that TI documents their stuff rather well.

 

[Joerg]

Make no assumptions. There are MCU families with the ADC equivalent
input resistance anywhere from 10K to 10M, not considering the dynamic
settling time.

Luckily, chapter 7 in this tutorial provides some more detail:

http://www.0x00f.com/files/ebooks/Programming dsPIC _Digital Signal Controllers_ in C.pdf

For example, that 2.5k source impedance can suffice. I am below that but
have a 1uF cap in addition.

Paradigm: programming for a chip is chip maker responsibility.

Or sometimes, you can get all the information. But only if you know the
owner for at least a decade and are married to his niece

???
1. TI documentation for new DSPs sucks.

2. TI has a habbit of telling one thing in exactly one place in the
datasheet. Such as if they describe a peripheral configuration register,
they describe only config bits that are relevant to this peripheral. All
other bits of this register are grayed. It is somewhere in the
datasheet, but you have no idea where to look for their definition.

http://www.ti.com/lit/ds/symlink/msp430f4260.pdf

Page 28 is what I call a good summary of ADC input characteristics. On
typical values give for input impedance but that's ok. The dsPIC
datasheet is a far cry from there.

 

[Nico Coesel]

I find that to be a problem with most datasheets now. At one time, datasheets
had schematics equivalent circuits of I/Os, if not the entire device
(admittedly a little tough for a micro ;-).

AFAIK NXP still provides an equivalent circuit for ADCs in their
micro's.

 

[Spehro Pefhany]

Ok, but they could have been a wee bit more detailed. Many times all you
want to do is monitor the voltage on some lines, say, from potmeters.
You don't want 200ohm potmeters and ideally also no buffer amps that
would introduced additional offset errors. They change slowly. So one
would need to know the DC input impedance and then slap a cap across it
so the ADC is happy. But they fail to state a DC input impedance.

The provided worst-case leakage current spec will limit you to a few K
of source resistance for +/-1 LSB error . If you like to play the
odds, you can try going higher..

 

[Joerg]

The provided worst-case leakage current spec will limit you to a few K
of source resistance for +/-1 LSB error . If you like to play the
odds, you can try going higher..

That won't be a concern because I am slghtly below 1k. But it would have
been kind of practical to know how much internal resistance is going to
be in parallel with my 1k. TI provides such information right in their
datasheets. Such openness has been rewarded with numerous design-ins so
far. Oh, and the pizza at the seminars was pretty good, too

 

[[email protected]]

True, but the ancillary documentation is often quite complete. I've
never flogged a dead processor for days without ultimately finding the
one improperly set up "screw the processor bit" in some obscure section
of some manual dedicated to some apparently totally unrelated section of
the processor.

I certainly have. We've had the factory folks in flogging these things. They
finally said "lets try all combinations". One problem was never solved. It
took hardware. We weren't the only ones who saw the problem, but according to
the vendor "no one else has reported the problem". Yeah, right.

 

[[email protected]]

I certainly have. We've had the factory folks in flogging these things. They
finally said "lets try all combinations". One problem was never solved. It
took hardware. We weren't the only ones who saw the problem, but according to
the vendor "no one else has reported the problem". Yeah, right.

BTW, the company lost upwards of two years trying to track these problems
down. Some of the problems are still only papered over. They come back to
haunt every so often.

 

[Spehro Pefhany]

That won't be a concern because I am slghtly below 1k. But it would have
been kind of practical to know how much internal resistance is going to
be in parallel with my 1k. TI provides such information right in their
datasheets. Such openness has been rewarded with numerous design-ins so
far. Oh, and the pizza at the seminars was pretty good, too

The model for an unbuffered switched-capacitor ADC input really does
not look like a simple resistor, so a model with parameters makes a
lot more sense to me.

Here, right from the data:

http://ww1.microchip.com/downloads/en/devicedoc/33023a.pdf

.... PDF page 391.

The details become especially important if you're using more than one
channel.


Best regards,
Spehro Pefhany

 

[nospam]

If it's a switched-capacitor ADC there probably isn't a DC input
impedance per se.

There is a model of the ADC input showing +/-500nA leakage and the 4.4pF
(18pF in 12 bit mode) sampling capacitor with an indication of the maximum
internal series resistance. The sampling time depends on how software
configures the ADC. The recommended 200 ohms probably covers the sample
time required for its maximum 1.1 MSPS conversion rate.