ads1299 hardware or software problem?

Stefan Erni

Other Parts Discussed in Thread: ADS1299

On my own electronic-board, I have the ADS1299.

Write an read register are working nice.But if I read the analog signal input I have a strange signal.

With a symmetrical generator (sinus 100Hz on channel 1) on output 1 is a bad signal and output 2 I have also a strange signal.

Thank You for a help

over 9 years ago

0 Brian Pisani over 9 years ago

TI__Mastermind 38745 points

Hey Stefan,

How are you applying the signal to channel 1? For instance if you are using a signal generator and have the signal generator ground connected to the negative input to the converter, you will not achieve a good signal because there is nothing actively driving the negative input so it is allowed to float with respect to the board supplies. I would say this is one of the most common reasons for the behavior you are describing. Instead the negative input should be fixed to some DC voltage within the ADS1299's supply range. The signal generator ground may also be connected to that voltage. Typically we recommend using the BIAS amplifier as a source for a bias voltage for the input signals or the patient, but really any actively driven signal near the mid-supply voltage will do.

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian

Thank you for the help.

I applay the signal to channel1 with a symmetrical NF Generator.

If I dont use the biasref,biasinv,biasin can I connect them to the ground?

and if I dont use the srb1, srb2 can connect it tho the ground?

Regards

Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

So are you saying the signal generator ground is connected to IN1N?

In the cases of the bias pins and SRB pins, you are safe leaving them all floating.

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian
thank you for the Infos.
The signalgenerator has a 3 pin connector, a noninverting signal, an inverting signal and a ground
The ground from the generator I connected to the "gnd4". (gnd4 is ground from 3.3V and 5.0V psu)
If I put a 20Hz signal and the gain from the ads1299 to 1, the signal looks well.
40Hz,60Hz,80hz is also ok.
about 100hz the signal looks very bad
160Hz looks ok
200hz the signal looks very bad

I'm using 2kHz samplerate.
Regards

Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stephen,

What do you mean when you say it looks bad? What specifically is happening? Could you collect some data and send it to me?

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

It looks as though it is being amplitude modulated in another signal. Can you identify the frequency of that signal?

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

that was a chart probelm. I made a new 24bit 2khz chart and the signal looks well.
just some new problem.There is a lot of DC offset changing slow. Is there a good input circuit with some R and C to make it more stable?

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

That could be due to the patient bias changing over time. How are you biasing the patient?

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian

Left side is bad signal from time to time, right side is the good signal. The cableand and hardware and software is always the same.

I'm not biasing the patient. I just put 3 electrodes pos,neg, and gnd

regards

Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

I don't think I've ever seen anything quite like that. Would you send the raw data corresponding to the image on the left?

Brian

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian

At the moment I can not save data, bud soon.There is another question. If i change the gain, the sinus-signal has more offset but not more signal(maybe more noise)

regards

Stefan

Gain1 (hex0) (ch1Set)

Gain6 (hex30)

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

What happens when you try to convert the test signal with a gain of 6?

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

its working.chart1 is gain6, chart2 is gain1

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

That seems to indicate a problem with your input, then, since the test signal works exactly the same as a normal input. Perhaps a drawn diagram would help me to understand how you have the signal generator connected to your board.

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian

The genrator is a Neutrik MR1 with batteri.

I connect it like in the schema.Did I catch problem withVcap1....Vcap4. Simple used 10uF...

regards

Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

Are you offsetting the signal from ground? Also for the best performance you should use the VCAP values recommended in the datasheet.

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian
From Vcap1, I can see in the manual what is he doing. Is it possible to have infos about Vcap2,Vcap3, Vcap4
Regards
Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

From Figure 53 in the datasheet:

VCAP1: 100uF
VCAP2: 1uF
VCAP3: 1uF || 0.1 uF
VCAP4: 1 uF

-Brian

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian

Thank for the infos.This I saw it .But for Vcap1 there is on page 23 a nice info what Vcap1 is doing. But no infos about the another. Anyway I think they are not the problem.

I tried to put fromt posInput and negInput to to ground a 10kOhm and a 1nF.But no change.Then I put a 100kOhm from +5V to posInput and the signal looks much bether. Do you have an idea why there is a dc problem on the input?

regards

Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

Yes that was my original suspicion. If the sig gen ground is connected to the board ground, the inputs will be out of range. Can you put an offset on the input signal using your generator?

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian
No not with this generator. But Neutrik will bring me on monday a another Generator. The pro version of my generator.
But what about later if I connect it to a person, to I have problems?
(and gain is working nice now)

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

Where this becomes a problem for biopotential measurement is biasing the patient. The BIAS amplifier on this device is our solution to this problem. Essentially that amplifier is meant to connect to a bias electrode that brings the patient's voltage bias within the range of the ADC. Without that, you will need to find another way.

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian

In the forum I saw some put 2pcs 10Mega Ohm resistor from VDD to the pos-input and neg-input, is this a way?

Or if I use the +-2.5Volt psu?

regards

Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

Yes that is a way to do it if you place a series capacitor in front of the biasing resistor pair. The capacitor is a dc block and the resistors set the bias at the input to mid-supply. Keep in mind that this forms an ac high pass filter (this can be seen by doing the small signal analysis of the circuit).

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian
A lot of part in low tolerace class like 1%....
If I use the +-2.5Volt psu, where is the ground for the elektrodes (for the generator)

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

The bias for the patient does not matter when you've ac coupled the inputs. Only the signal content will reach the ADC.

Regards,
Brian Pisani

0 Stefan Erni over 9 years ago in reply to Brian Pisani

Intellectual 560 points

Hi Brian

Ok. Next week I try with the serie C and the resistor and take a record with eeg and ecg.

The signal now looks very good. I can change the gain in steps 1.2....24 that great!

But with the generator I just give 10mV /10Hz and I get a full signal.

Why the gain is that high? the ref 4.50 Volt

regards

Stefan

0 Brian Pisani over 9 years ago in reply to Stefan Erni

TI__Mastermind 38745 points

Hey Stefan,

Which gain setting are you using? And do you mean the internally generated 4 V reference? Have you offset your signal to mid-supply?

Regards,
Brian Pisani

Data converters

Data converters forum

ads1299 hardware or software problem?