diglet wrote: Thu Apr 18, 2024 4:34 pm
darthjj wrote: Wed Apr 17, 2024 10:02 am
I just found those pages on your wiki, and it completely overthrows my leading theory of why some people were perceiving carrier frequency changes as changes in signal intensity while others weren't. Well, back to the drawing board.. Appreciate the thorough testing!
FWIW my theory is that this is caused by a relationship between skin resistance and frequency. Suppose we have skin resistance 500ohm at 500hz and 200ohm at 1000hz, then increasing the frequency causes a (more than double) increase in current.
But the box also has internal resistance, which may be 100ohm or 500ohm, also affected by the winding ratio, presence of parallel resistor, and electrode quality. The lower the internal resistance of your stimbox, the more the current increases as the frequency increases (due to lower skin resistance).
The tests I linked earlier show that the internal resistance of the box is mostly independent of frequency
I investigated the dependency between skin resistance and frequency here:
https://github.com/diglet48/restim/wiki/skin-resistance
And the dependency between current, frequency and nerve activation here:
https://github.com/diglet48/restim/wiki ... activation
I also brought out the old, almost antique, oscilloscope today and ran a bunch of tests. It's analogue so there's of course a fair bit of measurement error everywhere. And I had to add a 100Ohm resistor in series with the body since I don't own a proper current probe, I hope this didn't influence the measurements. The currents I measured were a bit higher than I had expected, but I think I got the maths right..
I could confirm that the parallel resistor doesn't change how frequency translates to perceived intensity, by trying to locate subjective levels (minimum noticeable power, light tease, pleasure, intense pleasure, and pain) and then comparing the current logged at each point. Accounting for the subjective error margin, values were more or less identical with and without the parallel resistor.
One thing that surprised me a little was how much more current was needed for each "level" after having gone all the way up to where it starts getting painful. When starting at 0 and slowly increasing power, I found the (light tease, pleasure, intense pleasure) currents at (32.5mA, 40mA and 50mA) and finally hitting the pain threshold at 80mA. Staying there only a few seconds before performing the reverse process, I now found the same levels at (50mA, 65mA and 75mA), and it seemed the tolerance remained as long as any signal was present. After I rested for 30 seconds with no signal the sensitivity was almost fully restored. Obviously I was aware of this effect, but I didn't expect the differences to be so large and that it built up so quickly.
With a little bit of "warm up", I've found these approximate relations between current and subjective intensity at 1000Hz:
- 20-30mA just barely noticeable
- 30-40mA light tease
- 40-50mA pleasure
- 50-80mA intense pleasure
- 80mA+ pain
Perhaps not useful for anything but I found it a little interesting that the range for what I would label as intense pleasure was so wide compared to the others. Though I suppose that's a question about definitions..
I also calculated body resistance by measuring current and voltage drop over the electrodes at the same time. Used a stainless steel head 'trode, approx 20cm², with electrode gel and a wet cloth loop around the balls, approx 15cm², I think I got similar values to yours though a little lower, perhaps due to the larger contact areas. With 1000Hz I got 740Ohm at 20mA, 643Ohm at 40mA, 594Ohm at 60mA and 507Ohm at 80mA.
Now here's the interesting thing, I did not get significantly different resistances at different frequencies. I set the current to 60mA, the signal generator to 1000Hz and measured 32V over the electrodes. Then I increased the frequency to 1500Hz, the current stayed the same and the voltage dropped down to 30V. I increased the freq. to 2000Hz and I could see that it dropped another 2V, but not to 28V.. it looked more like 30V.. almost 31, which confused me a lot. So I turned frequency down instead, and I could see the voltage jumping up each time, but somehow the voltage always stayed in the 30-34V range. Looking more closely at the 'scope, it seemed like the voltage always crept back towards around 32V peak-to-peak regardless of frequency, though very slowly, on the scale of tens of seconds. Wish I had a better oscilloscope so I could provide definite numbers instead of having to eyeball it.
Going back to my subjective intensity measurements, I tried to find each "level" for 700, 1000 and 1500Hz. It seemed like each freq. increase required an additional 5mA to reach the same level, i.e. "mild tease" hit at 32.5mA at 700Hz, 37.5mA at 1000, and 42.5 at 1500. It also seemed fairly consistent at all levels. Though it's hard to say how accurate this is accounting for subjectivity, measuring error and tolerance buildup.
