Tim Fake's Build Log

Hi,

My name is Tim. I have a Masters in Computer Science. I love hardware and I want to establish a circuit design consulting business. At the moment I am working on building my projects portfolio. My first project is to develop a sensor platform.

1 Like

Welcome Tim! Can’t wait to see what goes into the sensor platform! Any initial thoughts?

At the moment I’ve started out with the ESP32, a Sigma-Delta ADC (ADS122C04IPWR), an ECG sensor (maybe based off of this: https://www.mouser.com/ProductDetail/Maxim-Integrated/MAX32663AGTGFS%2B?qs=81r%2BiQLm7BTEDwSsf%2FqfAg%3D%3D), some sort of cuff-less blood pressure monitor (I haven’t identified a sensor for this), and a pressure sensor to detect and measure grip strength (I haven’t identified a sensor for this).

I just found this idea for a cuff-less bp monitor: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6292433/

1 Like

Hello Tim,

We explored some years ago a way the pulse transit time for blood pressure using images, and it works. For the PTT, you need one of the peaks of the ECG and the peak of the Plethysmography, i cannot reveal the way to infer the ECG from images but if you have a direct ECG is a very straight forward methodology since the Plethysmography can be obtained optically like the blood oxygen monitors . Using images, we got to a dead end since there is a point where phase shifts makes the process useless for some subjects.

Any way, welcome good luck with your project

3 Likes

There has a slight change of plans for the sensors. To make the project simpler for the first time out, I’ve gone with the AHT20 Temperature and Humidity sensor, Grove 3-Axis Accelerometer, and Grove Light Sensor v1.2. I’ve also decided to add on a 24 bit Sigma-Delta ADC (ADS122C04) for future devices that require higher precision.

Here is the schematic so far:

I haven’t completed hooking up the ADC yet. In the data sheet for the ADC is says to connect a bypass capacitor from DVDD to DGND but to connect a bypass capacitor from AVDD to AVSS, I’m not sure what that means or of it’s purpose. Help and any general suggestions will be greatly appreciated.

I think that I resolved this issue.

Here is the latest update on the schematic:

I think that I am still a little confused about the nature of the two separate grounds on the ADS122C04. Should I change the grounds on my power supply into either the analog or digital grounds (if so, which) or should they stay as they are? If they stay as they are, should the +3.3v for AVDD and DVDD be placed on a separate power supply?

Here is a copy of the relevant section of the datasheet:

I’ve seen EMC disasters resulting from people interpreting the data sheet too literally and trying to keep analog and digital grounds completely separate, with “star” runs back to the power supply. Basically you want to tie them together at the ADC. Do some reading on mixed-signal PCB layout best practices. Often you might start with separate ground planes, but with proper partitioning and attention to current paths, you will see that that’s unnecessary and end up with one plane and currents flowing where you want them to.

3 Likes

Yes, Julia is right, they say about 90% of datasheet recommendations are wrong,

There are some very good videos from Rick Hartley about grounding and ground splitting that could be of help.

Another one about differential pairs and stuff also very helpful.

2 Likes

Thank you both for the videos and suggestions. After I went through the videos posted, I went hunting for other mixed signal PCB design videos and found this series that I thought was pretty good so I wanted to share:

Hola Tim,

Great, I will check it up.

Saludos !!

The Amp Hour episode with Hank Zumbahlen (#185) has an excellent discussion on return paths.

It’s in my top-ten episodes list and I love the quote at 20:41 “Let me start out by saying that I wish I could find the guy who had the idea of making your power supply return and your reference node the same thing… …I wish I could find him and turn him into a speed bump”

1 Like