I'm concerned that there is so much connect your EMG,ECG,EEG electrodes to your amplifer and your all set to make measurrmets, control mouse pointers and soforth. What people forget is that when you connect to a mains powered item there is a risk of shock or injury. So your arduino or other widget that's powered from a wall brick, or a scope or a PC presents a real problem.

Its generally unwise to use anthing but battery power. Unless you understand and can control the current leakage paths your kit you may be in for a shock. The Leakage path is mostly via capacitance in power supply transformers and it makes no difference if its linear PSU or a SMPS.

I hope you wouln't deliberately connect yourself to the live mains through a small capacitor- so don't do it by mistake!

Its also important to limit DC currents that can flow though electrodes. An overloaded or damged ic input will provide plenty of current. even a few uA flowing through a skin electrodes can cause electrolytic burns. Not (usually) fatal, but very sore indeed.

Finally, don't test your thing out on those you love or hate. Just yourself, at your risk!

 

@DECELL
I strongly agree with you.
Thanks for a well written and needed caution.

I think this would get better attention on a different area of the AAC forum. Maybe the General Electronics Chat section. Even better might be a sticky in the Analog & Mixed-Signal Design section.

You can ask a moderator to move it by clicking the report button.

 

The simple solution is to only use medical grade power supplies, which have better isolation. The second solution is to let not knowing what you are doing STOP YOU!! Most of the damage that I have seen done is committed by individuals who were totally clueless charging ahead with no idea as to what they were doing. I will avoid naming the guilty ones so as to deny them any possible fame. Besides, probably nobody knows them.

 

Why is this a sticky? Concerns are not solutions.

 

Please name the guilty as I haven't seen any posts that imply anyone has been injured?

A medical grade power supply will not guarantee safety, granted their leakage currents are less but in reality don't differ much from your standard SMPS apart from price!

Isolation in medical devices isn't really done via the power supply but by the use of transformers and opto-isolators in the input circuitry.

You could argue that building your own test bench power supply is dangerous?

"I hope you wouldn't deliberately connect yourself to the live mains through a small capacitor- so don't do it by mistake!"

I'd like to think the members of AAC would have more sense and like jpanhalt don't see the need for a sticky. Using battery power only for circuits of this type is a good idea or failing that just ban them altogether just like other "unspeakables" on this site.

 

Please name the guilty as I haven't seen any posts that imply anyone has been injured?

A medical grade power supply will not guarantee safety, granted their leakage currents are less but in reality don't differ much from your standard SMPS apart from price!

Isolation in medical devices isn't really done via the power supply but by the use of transformers and opto-isolators in the input circuitry.

You could argue that building your own test bench power supply is dangerous?

"I hope you wouldn't deliberately connect yourself to the live mains through a small capacitor- so don't do it by mistake!"

I'd like to think the members of AAC would have more sense and like jpanhalt don't see the need for a sticky. Using battery power only for circuits of this type is a good idea or failing that just ban them altogether just like other "unspeakables" on this site.

The guilty parties are not ones who would visit such an elite place as this one. Their technical backgrounds do not come up to the levels of any that I have seen posting here. (I use "any" in the most broad sense)

 

You could argue that building your own test bench power supply is dangerous?

Yes, I would if it is connected to my heart through electrodes.

 

I'm concerned that there is so much connect your EMG,ECG,EEG electrodes to your amplifer and your all set to make measurrmets, control mouse pointers and soforth. What people forget is that when you connect to a mains powered item there is a risk of shock or injury. So your arduino or other widget that's powered from a wall brick, or a scope or a PC presents a real problem.

Its generally unwise to use anthing but battery power.

I disagree. That proscription assumes facts that are not universally accepted. Batteries can be up to almost any voltage. I have used 135V "B" batteries in transmitters in the 1950's. You could definitely get a tingle from them. It is not the source that matters, it is the voltage and current (ESR). Most medical instruments are mains supplied today.

Unless you understand and can control the current leakage paths your kit you may be in for a shock. The Leakage path is mostly via capacitance in power supply transformers and it makes no difference if its linear PSU or a SMPS.

Same applies to the aforementioned batteries. Does that apply to battery powered kits too? What if the mains supply is through a double-insulated transformer?

I hope you wouln't deliberately connect yourself to the live mains through a small capacitor- so don't do it by mistake!

Agree, and that is already in the AAC policies.

Its also important to limit DC currents that can flow though electrodes. An overloaded or damged ic input will provide plenty of current. even a few uA flowing through a skin electrodes can cause electrolytic burns. Not (usually) fatal, but very sore indeed.

That statement with regard to burns from inputs presents information not widely agreed upon. That is, will a typical IC with a current of a few uA can cause eletrolytic burns? Moreover, "electrolytic burn" is an unfamiliar term to me. Please define and provide a citation.

Finally, don't test your thing out on those you love or hate. Just yourself, at your risk!

Very bad advice. Current medical ethics do not allow investigator suicide as an exception.

Edit: Since I asked for a citation, here is a link to the US rules on ethical research involving human subjects: https://www.hhs.gov/ohrp/regulations-and-policy/regulations/common-rule/index.html
There is no exception for the investigator(s) as subjects.

 

I am wondering a bit why this discussion is even going on. Medical stuff is an area that some of us, including me, have chosen to totally avoid because of the potential for liability problems, mostly because nearly any product can be misused. Non-isolated mains powered devices, such as medium power variable speed drives, are far less risky, but quite possibly off limits here, although I am not sure about that.

 

/--/
Current medical ethics do not allow investigator suicide as an exception.

Edit: Since I asked for a citation, here is a link to the US rules on ethical research involving human subjects: https://www.hhs.gov/ohrp/regulations-and-policy/regulations/common-rule/index.html
There is no exception for the investigator(s) as subjects.

I am not disagreeing with your input in this thread, but, those are not rules on ethical research involving human subjects. Those (meaning what the HHS link is introducing, all of the relevant CFRs and so on which make up the "Common Rule" as well as any specific agency regulations) are actual regulations (essentially "laws" i.e., CFR) and Policies. I think that a larger view perspective is helpful.
_

To do research with humans in the US, you are required to prepare a protocol (detailed description of what you intend to do, including all equipment), which is typically reviewed by peers (who presumably have SME) and then reviewed by SME and regulators that may/or may not offer an approval. Obviously, conforming to all relevant CFRs and Policies (as interpreted by the approving body). This applies to Federal agencies, academia and commercial agencies.

To market a medical device in the US, you need FDA approval.

In general, the specifics of the requirements for those two issues (medical research, FDA medical device approval) are well beyond the scope of the Forum, so what is the purpose of the thread?

Earlier, you noted:

Why is this a sticky? Concerns are not solutions.

True, but a solution requires a problem - no?

I don't develop medical devices or conduct research with humans, but, in my view, the "problem" that the thread is addressing is one of safety and risk assessment. I support promoting those and I trust that you do also.

Understanding risk is a big deal and many times I read about a project or attempted project and I think, %$@# that looks dangerous. Sometimes, I say something outright. Sometimes I don't, in part because I don't understand the project/circuit and I am not going to build an example. But, I still might read about it to increase my understanding.

In that regard, the thread is as the title says.

Here is a slide show from a biomedical engineering class on Biopotential Amplifiers

On slide #1: Typical bio-amp requirements

1. protect the organism being studied
2. careful design to prevent macro and microshock
3. isolation and protection circuitry to limit the current through the electrode to safe level

So, from my viewpoint and in the spirit of the discussion in the forum, how do you do that?

Here is a link to a short discussion of the typical patient protection resistors that I have seen on biopotential amplifiers that are used with electrodes. Note the reference to IEC60601.

So, that's my two cents.

 

I am not disagreeing with your input in this thread, but, those are not rules on ethical research involving human subjects. Those (meaning what the HHS link is introducing, all of the relevant CFRs and so on which make up the "Common Rule" as well as any specific agency regulations) are actual regulations (essentially "laws" i.e., CFR) and Policies. I think that a larger view perspective is helpful.

Not going to get into a debate on the difference between what is in the CFR, a rule, and what we must follow. As you go on to state, there is a process that involves an IRB (Institutional Review Board). My point at the time was in rebuttal to the TS who said an investigator can do the experiment on himself without such approval. Maybe there is an exception in the CFR or Code of Ethics, but I am not a ware of it (see hypothetical below) .

While entities may exist in the US that are exempt from those regulations,* you would be hard pressed to find one if challenged. I suppose someone working in his basement by himself who does a test/experiment on himself might be exempt. That is, until he tries to publish it and/or influence others to do the same or buys/sales materials interstate, uses the USPS, etc. Realistically, they are what we must follow today. And generally, they make good sense; although, the process can be slow and influenced by power.

Let's not get too wound up in the nomenclature. My objection is not to the TS expressing his views, it is to AAC making it a sticky, which implies AAC's endorsement of a highly flawed position. That is what I meant in my first post and decided to come back to the thread as it seemed to be gathering traction.

John

*Technically, Federal regulations only apply to entities doing interstate business or with the Federal government. However, that distinction is virtually impossible to maintain. Witness Federal rules again discrimination. The "Commerce Clause" can be applied to almost anything. BTW, the rules change when dealing with forensics (i.e., forensics are not for diagnosis or treatment per se) and deceased people, which become chattel with a whole set of state rules/regulations of their own. Common sense is very helpful, but not always correct.

 

Not going to get into a debate on the difference between what is in the CFR, a rule, and what we must follow. As you go on to state, there is a process that involves an IRB (Institutional Review Board). My point at the time was in rebuttal to the TS who said an investigator can do the experiment on himself without such approval. Maybe there is an exception in the CFR or Code of Ethics, but I am not a ware of it (see hypothetical below) .

While entities may exist in the US that are exempt from those regulations,* you would be hard pressed to find one if challenged. I suppose someone working in his basement by himself who does a test/experiment on himself might be exempt. That is, until he tries to publish it and/or influence others to do the same or buys/sales materials interstate, uses the USPS, etc. Realistically, they are what we must follow today. And generally, they make good sense; although, the process can be slow and influenced by power.

Let's not get too wound up in the nomenclature. My objection is not to the TS expressing his views, it is to AAC making it a sticky, which implies AAC's endorsement of a highly flawed position. That is what I meant in my first post and decided to come back to the thread as it seemed to be gathering traction.

John

*Technically, Federal regulations only apply to entities doing interstate business or with the Federal government. However, that distinction is virtually impossible to maintain. Witness Federal rules again discrimination. The "Commerce Clause" can be applied to almost anything. BTW, the rules change when dealing with forensics (i.e., forensics are not for diagnosis or treatment per se) and deceased people, which become chattel with a whole set of state rules/regulations of their own. Common sense is very helpful, but not always correct.

I think we are responding with two very different perspectives.

There is absolutely NO doubt in my mind that if it was revealed that, as a medical researcher, you used yourself as a subject (unless that is what the protocol specifically stated), you would be severely penalized. IOW you would be disgraced, lose funding, get tossed, etc... No Institute in their right mind wants that kind of liability and notoriety. No respectable IRB would cut you ANY slack because you were the subject if it was so stated in the protocol.

I am trying to "steer" the thread into discussions and education about the risks and steps for mitigation of the risks.

What, exactly, do you mean by "My objection is not to the TS expressing his views, it is to AAC making it a sticky, which implies AAC's endorsement of a highly flawed position."

Specifically, what precisely is the "highly flawed" position?

I mean that literally, I don't see it at all. A position that is pro-safety and pro-mitigation of risk is what I see as the "position" and that is based in part on typical User Agreement terminology. I don't think that is a flaw.

I don't know why it was made "sticky". I would speculate to encourage ongoing discussion of risk mitigation - but I don't know.

 

Specifically, what precisely is the "highly flawed" position?

1) The TS said all medical devices must be battery powered. Correct? Yes/No
2) The TS said an individual could get an "electrolytic burn" with 1 uA from an input pin of an IC. Correct? Yes/No (While you are at it, please define electrolytic burn.)
3) The TS said disregarding the safe protocols is OK if the experiment is done on one's self. Correct? Yes/No.

The only thing the TS said that is generally accepted is the danger of non-isolated supplies (e.g., capacitor-coupled mains power supplies). And, that is already in the TOS for AAC. However, #3 directly contradicts that TOS.

One out of 4 is not even 70%.

John

 

1) The TS said all medical devices must be battery powered. Correct? Yes/No
2) The TS said an individual could get an "electrolytic burn" with 1 uA from an input pin of an IC. Correct? Yes/No (While you are at it, please define electrolytic burn.)
3) The TS said disregarding the safe protocols is OK if the experiment is done on one's self. Correct? Yes/No.

The only thing the TS said that is generally accepted is the danger of non-isolated supplies (e.g., capacitor-coupled mains power supplies). And, that is already in the TOS for AAC. However, #3 directly contradicts that TOS.

One out of 4 is not even 70%.


John

OK, you made those points and, basically, I agree with them (I add the "basically" to hedge my limitations, but not regarding #1 or #3 above).

But, you said ""My objection is not to the TS expressing his views, it is to AAC making it a sticky, which implies AAC's endorsement of a highly flawed position."

I think you stretched too far on that. Making a thread "sticky" does not mean that AAC endorses everything that is in the thread. So, we have expressed our opinions and that is good enough for me.

 

Its also important to limit DC currents that can flow though electrodes. An overloaded or damged ic input will provide plenty of current. even a few uA flowing through a skin electrodes can cause electrolytic burns. Not (usually) fatal, but very sore indeed.

A few uA???? Do you mean milliamps? Because I do not understand how just a few microamps can do real damage. Or am I missing something here?

 

Raymond said: "...many times I read about a project or attempted project and I think, %$@# that looks dangerous"

That's for %$@# sure! I haven't been around AAC for very long, but I have seen questions about projects that I have regarded as difficult and dangerous for someone with a wealth of knowledge and understanding, while the very questions have made it clear that the person asking them shouldn't be anywhere near the circuitry. On a couple of occasions I have said so, very bluntly.

MisterBill2 made an excellent point: "... solution is to let not knowing what you are doing STOP YOU!!" But ...
I was involved in a discussion on another site about the dangers of electric shock from many of the Chinese LED "light bulbs" that have non-isolated power supplies and row upon row of LEDs with exposed terminals. I had one person who was trying to make a case for "only dangerous if..." ask me if I knew what a bridge rectifier was. "Well, yes" (and a few other words) I replied. I think that was one of the last simulations I did before I packed in doing electronics. I had another try to "fix me" (his term) to say that DC at very low levels is more dangerous than much larger line-frequency AC voltage (it isn't, and I cited actual supporting research). But the real point is, neither of these people understood their limitations and ignorance, but were quite prepared to dish out "advise."

Recently there have been at least two threads about an EKG circuit that was published somewhere else. People came here for assistance - not surprising, because the published circuit was a rather bad design. There will be more instances. There is the possibility for injury. All is (usually) well when something runs on a low-voltage battery. Then someone adds an Arduino. Then they plug the Arduino into a USB port on a mains-powered computer. Well, that's usually safe, but if the AC mains ground is not intact (and I heard of a case where a real estate agent replaced 2-wire receptacles with 3-wire, but without the ground!), then "ground" suddenly is half the AC line voltage due to the Y capacitors at the input of the power supply - which are supposed to be low enough in capacitance to be at least reasonably non-lethal in exactly this scenario.

Do the TS's comments warrant a sticky? Maybe, but I agree with John at #13. I think something as a reference page with some well-written information and supporting citations would be better. I don't know how that sort of thing is handled at AAC. Does AAC, as an entity, want to engage in saying "this should be safe" or only in "don't do this"? It is the sort of thing that probably merits legal opinion - with the sure and certain knowledge that courts shred legal opinion with aplomb.

===
I have never done medical electronics as such, though I have done instrumentation used in medical research. I got to be an operating room tourist for one of the first defibrillating pacemaker implants done in this part of the country. Very cool. They are rather big on ground fault (leakage current) protection in ORs.