Newbie question: Fail-safety of a simple DC circuit

[Margse]

Hi! The short backstory is this: I'm interested in what's called tDCS, trans-cranial direct current stimulation. To put it short, this means running a <2mA current through somebody's brain. Sounds wacky, is kinda wacky, but has turned out to be very safe from the medical point of view. And I mean neither me nor any masochist nutjob came up with this thing, it's a well known treatment that's been used quite widely for sometime now (if somebody's interested I can give some nice links).

Now some people have been looking into getting these to people's homes. There is a project (or actually several, but the most interesting one is this: https://flowstateengaged.com) that's been trying get a cheap home-assemblable kit for sale, but they understandably ran into some bureaucratic trouble. They have however released the schematics for the thing they use.

I don't have much of a background in electronics (I'm a master's student in physics though, so I'm not totally ignorant when it comes to electricity), but even I should be able to build one of these, it's pretty much dead-simple. The thing is, your brain being one of the resistors, you want to be as sure as possible that the thing is fail-safe.

So after all this babbling, this is my question: With the schematic attached, is there any conceivable way that thing could fail and produce a current spike? That's what the fuse is there for of course, but how reliable are those? And if I get what's called a fast-acting fuse then how fast is fast? Is there something else that should be taken into account?

 

[KeepItSimpleStupid]

With a 12 V power supply, 12 V may even have a hard time conducting the currents.
What I think is missing is a LED that indicates that the correct amount or some current is flowing . e.g. that there is a significant drop across the regulator. It's too easy for nothing to happen without that. The currents may be hard to achieve.

This part: https://www.electro-tech-online.com/custompdfs/2012/06/3092fb.pdf might be a better part to use, but the resistors may have to be recalculated. Another suggestion is a DC-DC converter which would give you better battery life and a higher consistant voltage. The important issue is that it be battery powered.

if you go to www.digikey.com, you can look select an appropriate fuse and look at the data. When a fuse pops is dependent on many factors.

7 to 10 mA is considered a lethal current when it's applied across the heart muscle so damage to the skin could occur (burning) if the current was exceeded by lots. Adding a small flame proof resistor such as 100 ohms would offer more protection, but make sure the fuse will blow.

 

[ronv]

Yikes! What are you trying to do?

 

[strantor]

This tastes like Royal Rife, but some prefer the taste of colloidal silver.

 

[canadaelk]

Please test the circuit. After that, come back (may be) to us and explain why the LED did not turn on. E

 

[Margse]

With a 12 V power supply, 12 V may even have a hard time conducting the currents.
What I think is missing is a LED that indicates that the correct amount or some current is flowing . e.g. that there is a significant drop across the regulator. It's too easy for nothing to happen without that. The currents may be hard to achieve.

This part: https://www.electro-tech-online.com/custompdfs/2012/06/3092fb.pdf might be a better part to use, but the resistors may have to be recalculated. Another suggestion is a DC-DC converter which would give you better battery life and a higher consistant voltage. The important issue is that it be battery powered.

if you go to www.digikey.com, you can look select an appropriate fuse and look at the data. When a fuse pops is dependent on many factors.

7 to 10 mA is considered a lethal current when it's applied across the heart muscle so damage to the skin could occur (burning) if the current was exceeded by lots. Adding a small flame proof resistor such as 100 ohms would offer more protection, but make sure the fuse will blow.

Thanks a ton for the reply, this was of great help. For starters I was thinking of just putting a current meter after the fuse to keep on eye on what's going on make sure the circuit is doing what it's supposed to. I'll check the stuff from your second paragraph a little later, it's gonna take some reading. The digikey link lead me to some nice spec sheets which had blowing-time/current plots. Precisely what I needed.

The current getting exceeded by lots is a big no-no, apparently it's safer than one might think to have currents through your head (electroconvulsive therapy aka electroshock therapy, which by the way is still in use contrary to common belief, uses pulses of a few seconds with 800mA), but I'm still taking no unnecessary risks what-so-ever. If it gets to the level of burning my skin something has gone wrong.

Can you briefly explain what's advantage of having a flameproof resistor in addition to a fuse? Is it that if the current regulator would fail and give out a big voltage it would limit the current to some sensible level untill the fuse pops?


As to what I'm trying to do, I don't really know yet. You can find tons of hype about tDCS around the web, claiming you can BOOST YOUR BRAIN POWER, which is of course ********, or rather a horrible oversimplification. But it's true that it has been found to have interesting effects on brain functioning, depending on how you place the electrodes. The basic idea is that the current, most of which is lost when it travels on the skin, changes the excitability of neurons. Near the anode neurons become more easily excitable, meaning their threshold for firing a neural impulse is lower, and the other way around near the cathode. Placing the electrodes differently you can try and affect different brain regions, with effects ranging from motor learning (https://www.ncbi.nlm.nih.gov/pubmed/19802336) to working memory (https://www.ncbi.nlm.nih.gov/pubmed/16843494) to depression treatment (https://www.electro-tech-online.com/custompdfs/2012/07/047.pdf). So if I ever get this thing built and have the nerve to turn it on I would just start by trying what kind of effects it can have, and see where this goes from there.

 

[KeepItSimpleStupid]

Can you briefly explain what's advantage of having a flameproof resistor in addition to a fuse? Is it that if the current regulator would fail and give out a big voltage it would limit the current to some sensible level untill the fuse pops?

OK, Carbon resistors typically increase in value when they are stressed. Metal film resistors "puddle" or a molten glob or get lower or higher in value. Flame proof/fusible, usually metal oxide but can be wire wound. Metal oxide resistors just dicintegrate if stressed. WW just opens. So, they do act as a current-limiting resistor and a hard to blow fuse. I will add, that fuses can be changed to any value by the user, not saying that you will. A resistor of the proper type could be an extra layer of protection.

What bugged me the most about that simple circuit was "how do you know the circuit is regulating? Asked another way: Are the electrodes passing current through the skin or are they in air or a high resistance path?

 

[Margse]

Sorry about my slow replies.

What bugged me the most about that simple circuit was "how do you know the circuit is regulating? Asked another way: Are the electrodes passing current through the skin or are they in air or a high resistance path?

If I have an ammeter that makes sure I've got the current I'm asking for and the electrodes are ~10cm (a few inches) apart shouldn't there be no other option other than the current going where it's supposed to, i.e. through the body?

This part: https://www.electro-tech-online.com/custompdfs/2012/06/3092fb.pdf might be a better part to use, but the resistors may have to be recalculated. Another suggestion is a DC-DC converter which would give you better battery life and a higher consistant voltage. The important issue is that it be battery powered.

The DC-DC converter is probably a good idea, I'll see to getting one. I also checked the other current source you linked, I can see that it's designed somewhat differently from LM334 but I didn't really understand why one would be better for me than the other.

 

[KeepItSimpleStupid]

If I have an ammeter that makes sure I've got the current I'm asking for and the electrodes are ~10cm (a few inches) apart shouldn't there be no other option other than the current going where it's supposed to, i.e. through the body?

Well, current is also used to charge the capacitor. Usually you don't put big caps on the regulated side of things.

I do think it's useless unless you have some way of monitoring the compliance. In a real-world scientific current source there would be a max voltage (compliance) associated with the current source. A light will indicate if the current source was unable to comply. e.g. When the source is disconnected, the non compliance light is ON or you wanted 1 mA but not allowed to apply more than 50 V.

 

[jjw]

The led is connected backwards in the circuit diagram and won't lit.
If the electrodes are not connected to brain when the device is on, the capacitor will charge to about 11V and when connected discharges with current limited only by load ( brain ) impedance.

 

[Margse]

Excellent, thanks for that. The simplest fix for this would be to just do away with the capacitor. I have to double check but I'm pretty sure it's only function is to create a gradual increase and decrease in the current to avoid tingling near the electrodes when you switch the thing on and off, so it's only a matter of convenience (I was thinking of getting rid of even before, if I have electricity running through my head I would actually feel more comfortable if I did feel it). I suppose this would partly solve the problem KeepItSimpleStupid was talking about provided I have the ammeter there. Otherwise I do get the point behind having something to indicate the current source is putting out what I'm asking for but I'm not sure what to do about. Have to keep reading into this.