ADS1293: ECG and PPG synchronization using ADS1293 and AFE4400.

Hello Mohamamd,

Thank you for your post.

In order to synchronize the two devices, they will need to use external clocks derived from the same clock source and be configured for the same output data rate. The ADS1293 operates from an internal clock frequency between 370 kHz and 450 kHz. You would have to use the 4-MHz clock output from the AFE4400 and divide it down. The AFE4400 uses a typical external clock input of 8 MHz. If you scale this down to 7.2 MHz and divide by 16, you will get 450 kHz, which can be applied directly to the ADS1293 CLK pin. The AFE4400 gives you a "divide-by-2" buffered clock output, so an additional "divide-by-8" circuit will be needed. D flip-flops can be used to achieve this.

Once the desired clock inputs and data rates are configured, a sync or reset pulse will be needed upon start-up to ensure the data ready interrupts from each device are locked to the same clock cycle.

Best regards,

Did you have a look at the zip file that i attached?

I need to calculate the Pulse Transit Time.

But as you can see that the difference between the ECG and PPG is not consistent.

At first, i thought that it might be because of synchronization issue between ADS1293 and AFE4400.

But that doesnt seem to be the case.

Can you help me figure out the issue?

Hi Mohammad,

I did look through the images in the attached zip file.

How did you conclude that the device synchronization was not the cause? If the two devices output data at different data rates, their respective time stamps will not align.

Regards,

Hi Ryan,

The issue might be or might not be because of the synchronization, is what i meant.

I have not been able to pin poit the issue.

The ODR of ADS1293 is set to 200Hz. And the ODR of AFE4400 is also set at 200Hz.

The external crystal for ADS1293 is 4.096MHz and the external crystal for AFE4400 is 8MHz.

If the ODR is same, then what you mentioned above, that the time stamps will not align will not be true, Right?

Hence I have been confused.

I have attached the Pulse Transit Time Image that i obtain when i recorded using my System for about 20 sec.

Regards,

Ehshan Khan

Hello Ehshan,

What are the units on the y-axis of the PPT plot?

The tolerance and frequency drift of the two, independent external crystals may cause the two ADCs to output at slightly different data rates. While this may not be your root cause, it is important to achieve the most accurate PTT calculations.

Have you considered using a single device for this solution, such as the AFE4900? This device measures both PPG and ECG.

Best regards,

Hi Ryan,

Sorry for the delay.

Yes, I have gone through AFE4900 and yes I would love to use it.

But at present I want this system to work, hence i want your help on this.

I need to find the root cause of the frequency drift.

In the meantime, I figured to use CDCE913 for Clock synchronization and Divider.

Please any help and opinion is necessary.

Regards,

Ehshan

Hello Ehshan,

I'm glad you resolved the issue. Would you mind shedding light on what ultimately was the cause?

Regards,