DIY ECG leads

[chefslot]

I have got a breakout op-amp AD8232 that has come with a 3.5mm headphone jack with 3 wires coming out of it. eBay auction: #https://www.ebay.co.uk/itm/AD8232-ECG-Heart-Monitor-Sensor-Module-Pulse-Single-Lead-Measurement-Kit-Arduino/163359225515?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

These use the button popper (pic#2) style clips that stick onto the electrode pads. I can get a load of these from China for quite cheap, the electrode pads they use in the UK have an alligator type of clip and appear to be made of better quality goop, more like tens machine goop. And I can get more of a datasheet from these. https://www.medisave.co.uk/ambu-whitesensor-ecg-electrodes-x-500.html?gclid=Cj0KCQiA7qP9BRCLARIsABDaZziaAeCJnDFNLIc8K8OIZIvVaaFWkTap4Kdmzy6IcS_rpaDW29eDpbQaAt2kEALw_wcB

I brought a couple of these because they were cheap and will come in handy for something eBay auction: #https://www.ebay.co.uk/itm/BNC-Male-Plug-to-2-Alligator-Clips-Adapter-Test-Cable-1-Meter-3-2ft-Long-C3B2/303568392689?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

That then got me down the rabbit hole of using SMA connectors, DIY scope probes and coax? Started seeing words like impedance and resistance to do with certain Hz of what the scopes input or working ranges and thought 'Does a £10 ecg kit need to be over-engineered'.

This chaps video, his waveform was very smooth, mine looked a little like the first picture. He did mention to upgrade the leads and it does look like I have got a lot of noise in my waveform. I wish I had a decent picture but whilst being entangled and not able to move to capture the screen shot, this is the closest I can find.

How I hooked mine up, two female jack outputs, one for if my soundcard is sensitive with a 10k resistor in serial and not two 5k's as a voltage divider, see picture #3 (sorry for the child's play). The other jack to use if my soundcard had some sort of limiting device on the input, passive or active I guess. 

Shall I go with the eBay BNC croc cables, or make my own from coax and SMA connectors?

I think I have covered everything. I'm also running from 2 AA batteries. Regards

   

[exe]

I suggest read this thread about safety: https://www.eevblog.com/forum/testgear/how-to-solve-the-ti-ads1299-eeg-demo-kit-is-not-save-for-humans-problem/ .

[chefslot]

I suggest read this thread about safety: https://www.eevblog.com/forum/testgear/how-to-solve-the-ti-ads1299-eeg-demo-kit-is-not-save-for-humans-problem/ .

Indeed, I was going for the multitool of ECG machines, with built-in defibrillator   

On a serious note, he does say somewhere not to hook yourself up to a power supply, that is why I have opted for batteries and yes I know my soundcard in my pc could create a massive short or some other act of god and fry me. I did think about adding a diode but might make excessive noise, also the likelihood of this, and if I was that paranoid I could use a laptop. Thank you.

Oh it isnt for medical diagnostics anyway. 

[chefslot]

Anyone else has any input?

[Ground_Loop]

Anyone else has any input?

The number of hand-wringing, sky-is-falling, safety sirens that come out of the wood work on these issues is just pure entertainment.  I cannot find any report of anyone EVER being injured by properly placed ECG equipment.  The single injury report I could find was someone being connected to electrodes that were then plugged directly into the power supply.  Recognize the risk, imagined or otherwise, and just move on.  Build it. Don't forget the magic fuse. Use it.

[fcb]

Galvanic isolation if you're plugging into a mains referenced PC. And test the isolation.

As for cables, www.cablesandsensors.eu, or probably something from alibaba - defo. use shielded cables.

[chefslot]

Anyone else has any input?

The number of hand-wringing, sky-is-falling, safety sirens that come out of the wood work on these issues is just pure entertainment.  I cannot find any report of anyone EVER being injured by properly placed ECG equipment.  The single injury report I could find was someone being connected to electrodes that were then plugged directly into the power supply.  Recognize the risk, imagined or otherwise, and just move on.  Build it. Don't forget the magic fuse. Use it.

I can't work out if you are being serious or sarcastic Ground_loop aha

[chefslot]

Galvanic isolation if you're plugging into a mains referenced PC. And test the isolation.

As for cables, www.cablesandsensors.eu, or probably something from alibaba - defo. use shielded cables.

Good point, better still i may be able to send the voltage to the soundcard via wifi!

Edited, it will actually be easier to run it on a linux based laptop running xoscope. Will keep this post updated.

[chefslot]

Galvanic isolation if you're plugging into a mains referenced PC. And test the isolation.

As for cables, www.cablesandsensors.eu, or probably something from alibaba - defo. use shielded cables.

Just one more thing, if I was to use shielded cables would I ground them and where would you suggest grounding them too? I was thinking to the boards ground correct? I was looking at using some oscilloscope cables as they are shielded and grounded too right?

Regards

[fcb]

Galvanic isolation if you're plugging into a mains referenced PC. And test the isolation.

As for cables, www.cablesandsensors.eu, or probably something from alibaba - defo. use shielded cables.

Just one more thing, if I was to use shielded cables would I ground them and where would you suggest grounding them too? I was thinking to the boards ground correct? I was looking at using some oscilloscope cables as they are shielded and grounded too right?

Regards

Your AFE local ground.

[Wimberleytech]

I built one some years back using an AD624 which is just an instrumentation amplifier.  It kinda worked...was noisey.  I found that good quality pads made a big difference.

Ultimately abandoned the project in favor of a Kardia--a very impressive device.

[fcb]

I built one some years back using an AD624 which is just an instrumentation amplifier.  It kinda worked...was noisey.  I found that good quality pads made a big difference.

Ultimately abandoned the project in favor of a Kardia--a very impressive device.

I've also got a Kardia Mobile (AliveCor) - agree it's impressive.  You tend to have to wet your fingers or press very hard on mine to get a good connection.

One thing that's interesting (with the Kardia), is that the first couple of seconds of recording is very noisey, it then seems to retrospectively apply a filter and re-process the results (possibly some sort of syncing hum/notch filter).

[chefslot]

Your AFE local ground.
[/quote]

I'm sorry AFE local ground, is that the ground pin on the board. AFE has confused me a little, analog front end?

Also salty water has supposed to help on those devices. Obviously try not to get any salty water into the device.

Regards

[fcb]

I'm sorry AFE local ground, is that the ground pin on the board. AFE has confused me a little, analog front end?

Also salty water has supposed to help on those devices. Obviously try not to get any salty water into the device.

Regards

Yes, AFE - analog front end, be it an IC or a discrete circuit.

There are special sticky conductive gels used on the pads, they used to be silver loaded - not sure what they use now.

[hfleming]

From your first pic, it seems as though your electrode reference electrode was not connected. The bio electrical signals that you want to detect is very small (mV for ECG/EMG and uV for EEG), the common-noise that your body picks up is huge compared to the differential signal that you want to measure. Therefore correct placement of the reference electrode is crucial, as it is supposed to cancel out the 50Hz mains signal, as well as other common mode noise. (Your body acts as a huge antenna, picking up everything.) the reference electrode drives your body with the anti phase of the common-mode signal that comes from the 2 main electrodes, thereby supposedly cancelling out the Common mode noise, however, it is not always that simple. You can get your signal clean, will see if I still have some clean plots of my EMG patterns.  I have attached a pic showing an example of electrode placement. Good luck with your adventures.

[chefslot]

Thank you but the first picture was a reference to the noise I was having and not the real selection as when I went to screenshot it made more noise as I moved aha

[Wallace Gasiewicz]

The gue that is used in ECG leads today is a salty water gel. Called a paste.

Homebrew recipe:


1. Water, 1 cup
2. Salt, 2 tablespoons
3. Flour, 1 cup
4. Bleach, just a drop

Copied from: https://www.instructables.com/How-to-make-ECG-pads-conductive-gel/