ADS1298: Input impedance certification test issue

Qianqian

Part Number: ADS1298

Dear TI Experts,

As discussed in another link:https://e2e.ti.com/support/data_converters/precision_data_converters/f/73/p/693255/2611200#2611200

I am now using ADS1298IPAGR for ECG projects~And now we need to pass the test according to YY1139-2013。But now we fail the test,and really really need your help:

Test condition/Method:

Condition: Connect a parallel RC network 620k || 4.7nF in series with tested electrode,The idea is to compare the voltage level with and without series impedance. The difference must be less then 20% to pass the test.The test signal frequency 0.5H/10Hz/20Hz/100Hz.~~Here is the test circuit：
Method:①ECG diagnosis equipment powers on,set the gain to 10mm/mV ②close S1&S2,set S3 to a, those used leads connet to P1 and P2, unused leads connect to P6,then regulate the sinusoidal signal generator to a 1Hz，20mm(peak-to-valley value) signal ③then disconnect S1,measure the change of amplitude.In the stable state,the change must be less then 20%. ④ Then use 0.5Hz，10Hz，20Hz，100Hz,and repeat step ①&②， The output amplitude's difference( disconnect S1 compared with connect S1) should be less then 20%.

Test results:

illustration：1. Sample rate is configured by the register of ADC1_ADS1298CONFIG1

2.Capacitance means the input capacitance of all channels, for example,C45、C46 in bellow picture:

3.pull-up resistor means the input pull-up resistor of all channels,for example,R49、R50 in above picture

4.for number 7 in above test results, R47 and R48 are shorted together~

From the test resuts,0.5Hz/10Hz/20Hz can pass the test in all conditions,but 100Hz cannot pass the test in any conditon~

Bisides， I have tested the EVM: ADS1298R, and test result is as followed:

100Hz 2mV signal,when the S1 is connected,the result is 1,91mV，and when the S1 is disconnected,the result is 1.96mV.

It means when external circuits is added,the peak-to peak value becomes larger. Compared with the original signal,their is distortion~

The register is configured as followed,and if you need the schematic,I can send you by mail~Hope I have made myself clear,If any problems,please let me know soon~thank you very much~

define ADC_CH_SET_ON_1 (0x0010u)

#define ADC_CH_SET_ON_6 (0x0000u)

#define ADC_CH_SET_ON_12 (0x0060u)

#define ADC_CH_SET_OFF (0x0090u)

#define ADC_CH_SET_MUX_OUT (0x0000u)

#define ADC_CH_SET_MUX_TSIG (0x0005u)

#define ADC_CH_SET_MUX_TMP (0x0004u)

//#define ADC_CH_SET ADC_CH_SET_ON_6 | ADC_CH_SET_MUX_TSIG

#define ADC_CH_SET ADC_CH_SET_ON_6

/* ADC 1 parameter data */

#define ADC1_TYPE ADS1298

#define ADC1_SERPORT (1)

#define ADC1_ADS1298DEVID (0x0000u)

//#define ADC1_ADS1298CONFIG1 (0x0085u)/* 采样率1000 */

#define ADC1_ADS1298CONFIG1 (0x0084u)/* 采样率2000 */

//#define ADC1_ADS1298CONFIG1 (0x0083u)/* 采样率4000 */

//#define ADC1_ADS1298CONFIG2 (0x0000u)/* 内部测试源 */

#define ADC1_ADS1298CONFIG2 (0x0015u)/* 内部测试源 */

//#define ADC1_ADS1298CONFIG3 (0x00DCu)

#define ADC1_ADS1298CONFIG3 (0x00EFu)

//#define ADC1_ADS1298LOFF (0x0003u)

#define ADC1_ADS1298LOFF (0x00E3u)

#define ADC1_ADS1298CH1SET ADC_CH_SET

#define ADC1_ADS1298CH2SET ADC_CH_SET

#define ADC1_ADS1298CH3SET ADC_CH_SET

#define ADC1_ADS1298CH4SET ADC_CH_SET

#define ADC1_ADS1298CH5SET ADC_CH_SET

#define ADC1_ADS1298CH6SET ADC_CH_SET

#define ADC1_ADS1298CH7SET ADC_CH_SET

#define ADC1_ADS1298CH8SET ADC_CH_SET

//#define ADC1_ADS1298RLDSENSP (0x0000u)

//#define ADC1_ADS1298RLDSENSN (0x0000u)

//#define ADC1_ADS1298RLDSENSP (0x0030u)

//#define ADC1_ADS1298RLDSENSN (0x0020u)

#define ADC1_ADS1298RLDSENSP (0x00FFu)

#define ADC1_ADS1298RLDSENSN (0x00FFu)

//#define ADC1_ADS1298LOFFSENSP (0x00FFu)

//#define ADC1_ADS1298LOFFSENSN (0x0002u)

#define ADC1_ADS1298LOFFSENSP (0x00FFu)

#define ADC1_ADS1298LOFFSENSN (0x0001u)

#define ADC1_ADS1298LOFFFLIP (0x0000u)

#define ADC1_ADS1298LOFFSTATP (0x000FFu)

#define ADC1_ADS1298LOFFSTATN (0x0003u)

#define ADC1_ADS1298GPIO (0x0001u)

#define ADC1_ADS1298PACE (0x0001u)

#define ADC1_ADS1298RESP (0x0000u)

#define ADC1_ADS1298CONFIG4 (0x0002u)

#define ADC1_ADS1298WCT1 (0x0009u)

#define ADC1_ADS1298WCT2 (0x00C2u)

//#define ADC1_ADS1298WCT1 (0x000Au)

//#define ADC1_ADS1298WCT2 (0x00E3u)

over 8 years ago

0 Ryan Andrews over 8 years ago

TI__Guru 69076 points

Hello Fawn,

Thank you for your post.

Why do you say that your system fails the test at 100 Hz? The worst result that you show is 17 / 20 = 85%, which is greater than the passing criteria of 80%.

Multiple factors, including noise coupling, filter frequency response, and sampling rates will affect the magnitude of the signal you measure at the ADC output. Leave R47, R48, C45, and C46 installed. Remove R49 and R50. These components should not affect the input impedance measurement.

Please continue to monitor the original thread for updates. We understand the application issue that you are all trying to pass and I believe the same feedback will help everyone conducting the same test. We know that the input impedance of the ADS1298R is more than sufficient - with resistor lead-off enabled, the input impedance of 10M will still preserve approximately 94% of the input signal when forming a resistor divider with a series 620k. Current lead-off mode increases this impedance to 500M and preserves 99.88% of the input signal. The remaining factors that must be accounted for are input signal conditioning, digital filtering, and sufficient sampling rate.

Best Regards,

0 Qianqian over 8 years ago in reply to Ryan Andrews

Genius 5010 points

Hello Ryan，

Thank you very much for your reply~~

Sorry that I didn't made myself clear~

Our requirements are much higher than 20%~

We have to design according to 50M input impedance, requiring attenuation within 2%~

0 yi li2 over 8 years ago in reply to Ryan Andrews

Intellectual 615 points

Hi Ryan，

Thanks for your reply

I am a manufacturer that needs to solve the 50M input impedance. As Fawn said, we need to pass the 50M input impedance test. From the test results provided by Fawn, the amplitude is almost no attenuation at 0.5HZ, 100.5HZ, 200.5HZ, and the attenuation suddenly becomes larger at 100HZ. This is what we doubt. Also for the three factors you mentioned in the reply:
How much is needed for a sufficient sampling rate?
All digital filtering has been turned off in our test
The input signal is a problem that bothers us, because from the spectrum analysis, the signal source we purchased can't get a relatively pure output signal. What advice do you have for the input signal?
In addition, 1298 nominal 500M input impedance, what is your test method, how to test?

0 yi li2 over 8 years ago in reply to yi li2

Intellectual 615 points

Hi Ryan,

Sorry, the above description is wrong.

"the amplitude is almost no attenuation at 0.5HZ, 100.5HZ, 200.5HZ, and the attenuation suddenly becomes larger at 100HZ." should be "the amplitude is almost no attenuation at 0.5HZ, 10HZ, 20HZ, and the attenuation suddenly becomes larger at 100HZ."

I am sorry that this error has caused you inconvenience.

Data converters

Data converters forum

ADS1298: Input impedance certification test issue