M128 A2D Input Impedance?

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi there

My copy of the M128 datasheet lists the ADC analog input resistance as TBD Mohm.

Now my datasheet may be a bit outdated. Does anyone have this figure for me, or maybe a general idea of this impedance?

regards
Carel

www.pteq.net
Home of:
- Polygon Technologies CC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The copy I have shows a Min of 55M and a Typ of 100M.

Dave

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks!
Just the kind of answer I was looking for. Much faster than downloading the newest (4Mbyte) datasheet!

www.pteq.net
Home of:
- Polygon Technologies CC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Maybe much faster but also a bit incomplete.

You should look into the consequences of multiplexing, changing voltage levels, etc.

The DC input impedance may be 55-100M, but the dynamic impedance can be much lower. Search the forums for more information. This may or may not be of consequence to your situation or not but you should be aware of it.

Also, I wouldn't count on a true 100 Mohm input impedance. Your board, excess rosin flux, dust, moisture etc will significantly degrade that. And I am also not sure if that specification is for the actual ADC input itself and calculated or estimated, or if it also includes all the other pin logic, switching, etc, that is between the actual pin and the input of the ADC. The specs for I/O pins in general states leakage currents of 1 uA which implies a much lower actual resistance than 100 Mohm.

Please note - this post may not present all information available on a subject.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

To echo refields' comments:

"Impedance" is not a very complete idea when speaking of the ADC input. These inputs have an analog switch (multiplexer) followed by a sample/hold.

When nothing is changing, the input impedance may be very high. But, as soon as you change channels or make an ADC measurment, the input impedance may drop (for a very short length of time) to a MUCH lower value. The sample/hold is one of the big contributors to this.

This idea is enforced by the observations that you really need a low source impedance (less than 1Kohm preferred) for a signal source to the ADC.

Rather than paying a lot of attention to input impedance, you need to look at the source impedance required to maintain measurment accuracy. You will find that the required source impedance is a lot lower than what is consistent with a multi-megohm input impedance.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
less than 1Kohm preferred

Datasheet says 10k, are you saying that 10k may be too much?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

No, if the spec sheet says 10K, then that is the number you should use.

My point for the OP is that IF the input impedance is really many megohms, then there should not need to be a caution like "10K or less source impedance".

But, lets step back for a reality check here. Suppose that the input impedance of the ADC is 10Meg. A 10K source impedance results in a voltage divider. The loss is about 1:1000 and that would be equivalent to 1bit with a full-scale input. At 50% of full scale, it would be about a half bit and about one quarter bit at 25% of full scale input.

So, to even get close to the accuracy at tull scale, 10K is about it.

This is quite independent of additional voltage errors during the sample time. Depending on what happens to the sampler between samples, this error may be dependent on the previous measurment. But, with this kind of ADC, it is very hard to predict the magnitude or the sense of the error. The error may actually be positive at some voltages and negative at others. But, what ever is going on in this department, smaller source impedance is better.

There is, however, an additional caution about driving the ADC. A common response is to say, "Well, low imedance? Then use an op-amp to drive!" That is fine IF the op-amp has sufficient frequency response. Many low-power rail-rail op-amps don't. The sampling pulse is very fast and narrow. If the op-amp cannot hold its output stable during that time (because of limited bandwidth), you may be worse off instead of better.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
There is, however, an additional caution about driving the ADC. A common response is to say, "Well, low imedance? Then use an op-amp to drive!" That is fine IF the op-amp has sufficient frequency response. Many low-power rail-rail op-amps don't. The sampling pulse is very fast and narrow. If the op-amp cannot hold its output stable during that time (because of limited bandwidth), you may be worse off instead of better.

That's a good point, and one which had certainly escaped me. For low frequency signals, I've found an external cap can help reduce the effects of a high impedance source. If you have time to spare, you can also repeat the acquisition a few times to get a more accurate result from a high impedance source.

Four legs good, two legs bad, three legs stable.