Two rules you will never break, and the reasoning behind each. The seriousness with which you treat this is itself a credential.
You are skimming: the title, the first figure, and the short version. Switch to Read in the header for the full page, or Deep to open every deep dive.
This site puts electrodes on people, starting with you. Two rules cover almost everything, and the rest of the page explains why each one exists, because a rule you understand is one you will keep when nobody is watching.
Rule one. Anything connected to a body runs from batteries, with no conductive path to the wall.
Rule two. You never stimulate your own or anyone else’s nervous system outside an IRB-approved lab with a principal investigator.
Here is why. The wall is 120 volts. Every device plugged into it has a small leakage current to its own chassis and ground, normally harmless, because you are not connected to it with a low-resistance path. Electrodes with conductive paste are exactly that path. A fault in a cheap charger, a frayed cord, or simply the ordinary leakage of a power supply can put current through you, and current through the chest at the tens-of-milliamp level is how people die. The medical standard for equipment that touches a patient, IEC 60601IEC 60601The safety standard for medical electrical equipment, including how much current may leak into a patient (microamps). Glossary entry, limits leakage currentLeakage currentCurrent that flows into the patient through unintended paths; limited to tens of microamps for anything touching the body. Glossary entry to a few tens of microamps, a thousand times smaller than the fault current a wall supply can deliver.
A battery cannot do this. Its total energy is small and it has no path to the wall. This is also why hospitals use isolated power and why the AD8232 datasheet, like every biopotential chip’s datasheet, tells you the same thing in smaller print.
The reasoning. A milliamp through the scalp is not dangerous in the electrocution sense, which is exactly why people do it at home. The dangers are different: skin burns from electrochemistry at the electrodes, which are well documented in the self-experimenter forums; unknown effects of stimulating tissue you cannot see, in a brain that is still developing into your mid-twenties; and, for vestibular and magnetic stimulation, falls and seizures. The evidence that these techniques do anything useful at home is weak, so you would be taking a real risk for a speculative benefit. Learn the circuits. Read the literature carefully; the Phase 6 project has you build a stimulator and test it in a saline bath, where it belongs. Then, if you want to run a stimulation study, join a lab that does it.
The smaller rules
The test
If you would not put it on a ten-year-old cousin, do not put it on yourself.
Why this is a credential
Every device company you might work for lives inside ISO 14971ISO 14971The international standard for medical device risk management that every company follows. Glossary entry and IEC 60601IEC 60601The safety standard for medical electrical equipment, including how much current may leak into a patient (microamps). Glossary entry. An undergraduate who can explain leakage current limits, who has run their prototypes from batteries because they understood why, and who declined to self-stimulate because the risk-benefit did not justify it, sounds in an interview like someone who can be trusted with a patient. That is rarer than technical skill.
Recall
State rule one of the safety charter and the reason for it.
Anything connected to a body runs from batteries with no conductive path to the wall. Wall-powered equipment can leak or fault current through a low-resistance electrode path; IEC 60601 limits patient leakage to tens of microamps, which only isolation or batteries can guarantee.
Recall
Why is the Arduino the most common way rule one gets broken?
The amplifier is on batteries, but the Arduino is plugged into a charging laptop, so its ground, and therefore the amplifier's reference, is tied to the wall.
Recall
Why does the site forbid home tDCS even though the current is not dangerous in the electrocution sense?
The risks are electrochemical skin burns, unknown effects on developing brain tissue, and for other modalities falls and seizures, taken for a benefit the evidence does not support. Stimulation belongs under an IRB with a PI.