AFE4400 LED configuration and transmit supply.

Yuecheng Wang

Other Parts Discussed in Thread: CC2540, AFE4400, AFE4400SPO2EVM

Hello,

I'm using AFE4400 for pulse oximeter design. I designed firmware according to datasheet by myself rather than AFE4400SPO2EVM. By the way, I used NJL5501R and CC2540 in my design.

Unfortunately, I can not achieve ideal waveform and signals. Now, I have some questions:

1. Is it ok to supply TX_CTRL_SUP and LED_DRV_SUP pin with +3.3V though it's ok according to datasheet?

2. Why my RED light is still on when Bit 17(LEDCUROFF) is set in register LEDCNTRL (Address = 22h) ? Does that mean there are some problems with my firmware design?

3.(IMPORTANT) I found that Rf and Cf really make great difference in designing but I don't know how to choose Cf and Rf, can someone give me any advice?

Thanks a lot~

over 9 years ago

0 Amy Schnoor over 9 years ago

TI__Mastermind 18275 points

Yuecheng,

We have received your questions regarding the AFE4400. We hope to have a response back to you soon.

0 Praveen Aroul over 9 years ago

TI__Mastermind 49176 points

Hello Yuecheng,

We do not recommend phototransistor based pulse oximeter solution using AFE4400 for the following reasons:

1. Phototransistor has high sensitivity due to its high gain. Hence it pickups even minor movements, EMG etc.

2. We could not get consistent signals across subjects even after tweaking Rs and Cs in the biasing network.

(See following post for the biasing network: http://e2e.ti.com/support/applications/high_reliability/f/30/t/285068.aspx)

Regarding your questions:

1. Yes it is ok to supply 3.3V to TX_CTRL_SUP and LED_DRV_SUP so long as the datasheet spec. is followed.

2. The red light will be dimly lit when LEDCUROFF bit is set. This is because of the 50uA switching bias currents inherent to the timer function. Refer to section 8.3.5.2 of the AFE4400 datasheet for a detailed explanation.

3. As explained earlier, for phototransistor based design, we will not be able to recommend TIA BW settings since it is highly dependent on the mechanical setup, measurement site, spacing between sensor and skin etc.

Regards

Praveen.

0 Yuecheng Wang over 9 years ago in reply to Amy Schnoor

Prodigy 30 points

Thank you very much.

0 Yuecheng Wang over 9 years ago in reply to Praveen Aroul

Prodigy 30 points

Thank you Praveen for your detailed explanations, my NJL5501R design is base on the reference just as you mentioned in that post. (http://e2e.ti.com/support/applications/high_reliability/f/30/t/285068.aspx) I chose R=100k, C=100nF in NJL5501R design. I will try again, if I can still not achieve ideal signals I'll change my design by using photodiode.

Regards

Yuecheng

Data converters

Data converters forum

AFE4400 LED configuration and transmit supply.