Your leaking thatched hut during the restoration of a pre-Enlightenment state.


Hello, my name is Judas Gutenberg and this is my blaag (pronounced as you would the vomit noise "hyroop-bleuach").


decay & ruin
Biosphere II
dead malls
Irving housing

got that wrong

appropriate tech
Arduino μcontrollers
Backwoods Home
Fractal antenna

fun social media stuff

(nobody does!)

Like my brownhouse:
   data poisoning?
Tuesday, March 24 2015
For the last several days, the weather has been so unseasonably cold that today I was plagued by a cold weather phenomenon that usually only affects me for a few days at a time in the coldest part of a normal winter: the freezing of my flushless urinal system. When this happens, I am forced to go outside or urinate in bottles like some kind of interstate trucker jacked up on Gatorade, crystal meth, and antibiotic-resistant Gonorrhea. You're probably wondering why I don't just use one of the house's four conventional toilets. But I have a compulsive policy of never using the household flush toilet system (though I will violated that policy occasionally, usually for reasons having to do with timing).

At various times today, I did further tests with the hub circuitry designed to switch an I2C bus from an Atmega328 between a number of BMP180 barometric sensors. I kept running into a vexing problem that I'd first detected about a week ago: the only stable readings that seemed to reflect the reality of the environment (especially for the temperature part of the sensor) seemed to be coming from barometer number zero. All the other barometers (and I tested them at different addresses) seemed to be off by some amount. I wondered if perhaps some noise data from barometer zero was leaking through my multiplexing system to the other barometers. If so, this could have been a side effect of the ADG508A multiplexer. There are data sheets for this device online, but I've found no sample applications showing it in use. So I've had to guess at what voltages it requires. The data sheets all speak of power voltages as high as 40 volts, but nowhere do they specify the minimum voltage that the ADG508A requires. Since the BMP180 is a 3.3 volt device, I'm trying to run the ADG508A at that, but perhaps it needs more. If it does, though, it seems odd that my multiplexing system would be working at all. The only thing that seems to be wrong with it is that the data is skewed somewhat from where it should be on barometric sensors on multiplexed I2C buses greater than zero. I would think that if the 3.3 volts is too low, the ADG508A wouldn't be multiplexing well enough to send an address down the desired I2C bus and then receive the resulting data, something it seems to be doing reliably and exclusively (in other words, the data always shows up, and it comes from the correct sensor).
I did a number of other tests as well, including sending rapid streams of random data into the ADG508A's address wires just before doing a sensor reading to see if it had any effect on the resulting data. As far as I could tell, it did not, suggesting that "data leakage" or "data poisoning" (vague hypotheses I was entertaining without really understanding how they could happen) were not really my problem.

For linking purposes this article's URL is:

previous | next