Make:

Comments

Oldest comments listed first.

• The lack of isolation between the subject and the computer creates a potentially deadly situtation. A small ground fault could sent the subject into fibrillation.

  Never use ANY circuit that connects a human or animal to a line powered device without an optoisolator, RF isolator, or similar protection.

  Posted by: michaeljedelman on February 18, 2007 at 7:42 AM

• If you're a research engineer, connecting research animals to line powered devices without isolation is prefectly acceptable. You just need to convince your friendly local IACUC that you most likely won't harm the animal, say 1 in 10,000 odds.

  Humans are a different matter entirely, though this project is reasonably safe assuming the subject /isn't/ grounded.

  You can stick your finger in a 115v socket and not feel a thing as long as you are properly ungrounded. I've done it. Nothing happened and I couldn't feel a thing. 0.15 microamps was the fault current, standing on carpet, dry-ish skin, dry air (Arizona), and 5 feet from the walls. The same goes for this project.

  In order to get a serious shock, you need to have current flowing. For that to happen with this circuit alone (no other current paths or grounds), that would require the inputs to the computer's serial port to have dangerous voltages across pins which would subsequently have to make it through the opamps into the subject. Fairly unlikely, but not impossible. Even with a significant power line leakage current into the ECG system, no shock would be had without another ground return path through the subject.

  If you want to be safe, just run this experiment on a battery powered laptop. Preferably, put the laptop's powerpack in the next room.

  Sorry for the long response. I deal with these issues daily and there is so much misinformation out there.

  Posted by: CKThorp on February 18, 2007 at 10:25 AM

• As a follow up, people weren't really dying in operating rooms until the 1970's when electronic technology was proliferating to the point that more than one earth-ground-refered electronic instrument was connected to a patient at a time, which causes innumerable design headaches. (So many, that earth-ground-refered hospital electronics are essentally gone)

  Posted by: CKThorp on February 18, 2007 at 10:28 AM

• As a philosophical follow up, knowing how to accurately estimate risks and plan accordingly is the essence of being a Maker. Blanket statments of /never/ do "X" are as bad as labels of "don't open this product or you'll void your warranty" -- they stifle creativity without any increase in knowledge. If the product gave you schematics and mechanical diagrams you could make your own choices about whether you thought the risk of damage to the product balanced the gain in knowledge by taking it apart. The same goes for experimenting on one's self -- know the risks and make your own choices, similar to Barry Marshall's drinking of a dish of H. Pylori.

  Posted by: CKThorp on February 18, 2007 at 10:54 AM

Subscribe to MAKE Magazine!

Subscribe today, save 42% and get web access to MAKE free. MAKE Digital Edition is available only to subscribers.

$34.95 / 1 year
(4 Quarterly Issues)

Subscribe now