Channels all displaying saturated milA after starting

[jmicrobe]

Kris and I have had this issue a few times, the details are:

• Happens in both debug and run modes
• Blank (water) tubes inserted in all racks
• all racks off of platform, propped at an angle on shelf (typically we don't use the shaker/platform in debug mode)
• Nearly all of the channels read ~27200. The expected value for blank tubes (after zeroing) is ~1600.
• Using the voltage meter, all connections appear to have the correct voltages
• In debug mode, removing the tube/obscuring the light with a finger does not change the ~27200 value for the channel. Shining a flashlight into the tube slot also does not change the reading.

The last time this occurred in debug mode I just restarted the ODin about three times and it finally worked - but this hasn't fixed it yet today.

[jmicrobe]

Update: This morning I started the thing in debug mode with blank tubes and again got the ~27200 milA readings. Then I removed all tubes and re-started the debug mode (with the channels empty) and I got the expected ~1600 milA readings.

The readings were responsive to testing with my finger obscuring the LED, as well as having light shined in the channel. I did notice that when placing and replacing the tube in the slot there was often a drastic change in milA readings, which sometimes wouldn't recover to the expected ~1600 range.

I'm not sure if I can say that baselining without any tubes in the channels is what specifically corrected the problem. But I do suspect that the movability of the LED/photoresistor pairs in the rack might be it's own issue, and worth reconsidering the housing design for these on the racks.