I am doing a project on portable 12 lead portable ecg device. We have to use raspberry pi or arduino to implement the application. We are considering ADAS1000 to do so.
Since you only need a SPI interface for connecting the ADAS1000 to the target board, you should be fine with those platforms too.
So, if I use only EVM board of ADAS1000 without the SDP board and interface it directly with raspberry pi or arduino, will it be possible? Or i have to make my own breakout with 2 ADAS1000 chips for interfacing? As our intention is to send the ecg signals to the portable devices and not to PC.
There is a header connector with the SPI interface on the ADAS1000 EVM, you should be able to interface to wire into this header in some way.
So I don't have to make my own breakout? Only EVM board interfaced with raspberry pi through SPI will do?
In the Renasas code example shared on our WIKI site (think you are already aware of this based on your questions), the guys sky-wired between the two boards. You could do this too - like in picture shown:
Of course it would be much neater to create a breakout board between the SDP connector and your micro board - would be much more reliable/robust than wires..
This was really helpful. But we are thinking about creating a breakout using 2 ADAS1000 for 12 lead ECG application and interfacing it with raspberry pi for further project i.e. sending the ecg signal on a portable device. So
1) How many pin connector will be needed?
2)Can you please provide us any schematic of breakout using 2 ADAS1000 or idea about how should we make it?
3)Also is the data given in datasheet of ADAS1000 is sufficient to make a breakout?
4)And will it surely work without the Analog Device's SDP board?
Apologies for so many questions!
We are really looking forward to make this project work.
Thanks for response.
When you say "breakout" what do you mean? Would this be your own hardware with 2xADAS1000 devices and whatever interconnection is required to the micro board?
The schematic, layout and bom for the ADAS1000 EVB are included in the CN0308 reference design (uses the adas1000 evb) which is available on the ADAS1000 product page:
Note: you will need to ensure your design is in compliance with the applicable medical standards - in terms of patient protection etc. Our EVB is for product evaluation purposes, not intended for connection directly to patient/person.
Yes, by breakout I meant our own hardware with 2 ADAS1000 devices as we dont want to interface it with PC. So is it possible?
Yes got the schematic and layout design.
My apologies, but I don't really understand your question when you ask "is it possible"? What are your concerns?
Do you have concerns related to the ADAS1000 itself? (I can only help with your ADAS1000 questions).
Sorry I wasn't clear before.
I just wanted to confirm that ,are there any libraries available to interface ADAS with the Raspberry pi?
And while creating my own breakout board with 2XADAS1000, if I don't wanna do the respiration measurements, can I leave the pin outs as 'no connection'?
If you do not need respiration measurements using the External paths, then you can leave unconnected, however, they might be better connected to ground (noted in the datasheet on page 78 - Unused pins/paths).
I'm not aware of any libraries (of schematics) available interfacing ADAS to Raspberry pi (or at least not something we have to share).
Can you please tell me what are the values of DNP capacitors and resistors used in the master ADAS1000 schematics?
I am aware that DNP means we have to omit that from PCB board, but as for my project I am referring ADAS1000 master schematic, I am curious to know about the values. And also, can I use the filters shown in the schematic to reduce the noise? There too DNP capacitors are used.
The DNP locations were placeholders - e.g on the power supply decoupling - extra footprints for decoupling if needed, we didn't need.
For the ECG line-up, there are component footprints for purpose of filtering - e.g. filtering for ESU protection/ESIS filtering if the instrument was going to be used in that environment. These components are not fitted on the board as the type of filtering that people use varies greatly.
You'll really need to decide if you need filtering and what is appropriate for your application needs.
Take care with any additional filtering in the ECG line-up if you need to use the Pace Detection feature - push the filter bandwidth up high enough that the pace pulse is not too heavily filtered.
So we need not place the DNP components in schematic.
Also, we are trying to make a portable device, so we want to use battery powered supply. But as per the options given in the user guide, we can get supply for only IOVDD pin. Then what about AVDD, ADCVDD and DVDD pins?
I know that ADCVDD and DVDD have on chip regulators. But how to give AVDD through battery powered supply?
As a minimum, you will need to provide supplies for AVDD and IOVDD. As you know ADCVDD and DVCC can be supplied from on chip regulators.
The evaluation board was designed to operate primarily from a +5V incoming supply (to suit the wall adaptor that was supplied with the board), that +5V rail goes through the buck-boost ADP2503 to provide a regulated 5V rail that feeds 2x ADP151 LDOs which provide 3.3V for AVDD and 3.3V for IOVDD (note IOVDD can run lower supply, the minimum 1.65V). The ADP151 were chosen because of their low noise performance.
The ADP2503 can accept lower voltages, 2.3V min. What kind of battery are you planning to use? what range of voltage do you expect?
There may also be power solutions from the Analog Devices/Linear portfolio that could also be suitable here.
We were planning to use 5 V lithium ion battery for the supply. What is your suggestion for the 12 lead operation? which type of battery should we use?
And are we supposed to give the +5V to ADP2503 from the battery itself or wall adapter?
Also what is exactly is the significance of the '0 ohm resistor' used?
Sorry if I wasn't clear, the wall adaptor was included in the kit for evaluation purposes. When using the EVB with Blackfin SDP board, the EVB powers the SDP board too - which consumes quite a bit of power, so the wall adaptor is useful then.
In your own design, you have the freedom to choose the appropriate battery/powering for your needs.
What 0 ohm resistor are you referring to?
Resistor R26 ,value is 0r in the ADP2503 circuit given in the user guide. So can we use the same circuits for power supply using a 5V battery instead of the wall adapter?
The EN pin of the ADP2503 allows the device be put into shutdown mode. There may be value in connecting the EN pin to something if you need to shutdown the system - e.g. a switch on the unit to save battery power etc (if not catered for elsewhere).
Yes, the same power solution could be used with a 5V battery. There may also be alternative options from the broader portfolio of products that could be worth considering (size, cost, efficiency)
Please review your overall power budget to ensure that you have the appropriate system solution.
Is the zener diode used with SP724 in the slave schematic 1N4728? If it is, why 3.3 V zener diode is used there? As we have observed that the bidirectional diode used is 6.8. Why bidirectional is not used in the slave schematic in the EVAL ADAS1000 board?
Apologies for delayed response over the holiday period.
The same TVS is used for both master and slave.
The schematic in the eval board documentation may need to be updated for the slave version.
We have designed a band pass filter considering ECG band channel 0.05 Hz and 150 Hz. Will this band pass filter help us remove the noise due to skin tissue interference?
And also, what is the gain of amplifiers in ADAS1000 ?
When you say skin tissue interference - are you talking about motion aritfacts or other interference?
With a HP of 0.05Hz, I would expect you will need to use other additional filtering to remove things like motion, baseline wander etc from the signal.
The gain of the ADAS1000 line up is stated in the datasheet, you have a choice of 1.4, 2.1, 2.8 or 4.2. Note that increasing gain results in smaller electrode input range, so choose the gain that is appropriate to your application needs.
Yes I was talking about motion artifacts. Will this band pass filter help us to reduce that noise? As we tested it for a 10 mV periodic signal, it is attenuating it to 7 mV.
And as the ECG signal is in several millivolts, what extra filtering would you suggest other than a band pass filter considering ECG band channel 0.05 Hz and 150 Hz?
Also which amplifier with the gain as you mentioned before will be appropriate for signals in few millivolts ?
Retrieving data ...