Flooded Sample Chamber
It is very easy for a qualified technician to accidentally flood the sample chamber on a 5000 or an 8000. All it takes is a reversal of the inlet and outlet couplings between the simulator and the Intoxilyzer. Please remember that simulators and Intoxilyzers are separate instruments that are frequently disconnected from each other. The result of a flooded sample chamber is devastating for the instrument. Instead of analyzing the IR light passing through the air in the sample chamber, the electronics must analyze the IR light passing through a weird mix of liquid, air, and the crud from hundreds of previous breath tests. The instrument obviously needs to be sent out for maintenance but sometimes that is not practical or is too expensive. The quick solution is to turn the instrument up-side-down, shake it, and leave it running as long and as warm as possible. Eventually it may dry out - maybe after a month or so.
After drying out long enough, the instrument will pass its start-up diagnostics barely and will produce a good stand-alone diagnostics check.
However, as the instrument is drying out, the DVM (digital voltage - the electronic signals) fluctuates wildly. At first the DVM may be 000 or extremely high but gradually it settles down to perhaps 350 and the instrument will seem to work. The problem is that as the sample chamber is still drying, the DVM is fluctuating. It may average 350 but it is constantly going up and down (348, 351, 350, 353) and every once in a while the DVM may dip by 30, 40, or 50 (320,310, 300).
It is important to remember that there is no true zero on an IR machine. A breath reading or cal. check is simply the difference between an arbitrary position on the DVM scale at rest (e.g. 352) and the DVM scale on the blow or cal. check (e.g. 452) resulting in a digital reading of the difference (e.g. 100).
As a sample chamber dries, it may be possible to get several good readings but every once in a while you get a bizarre reading. You can even get readings of a difference of 40 where there is absolutely no alcohol. This may happen because zero may have been set (by the machine in the portion of the sequence where it sets zero) on the DVM scale below true zero (e.g. 312 rather than 352, 40 below true zero) due to one of these bizarre fluctuations in DVM.
Last month I conducted a hands-on tutorial for defence lawyers. I accidentally flooded the sample chamber in one of my machines. I turned it up-side-down, shook it, and have been keeping it warm. Sometimes it starts and seems to be working OK.
This week I was running a continuous string of stand-alone cal. checks, mostly so that there were a lot of air blanks - it is still drying out. No simulator was attached. All the cal. checks were zero. But then I got a cal check of about 40. I did not have a video camera going during the cal check but started to record during the air blank. See video below.
Notice that the air blank has a hard time getting to zero even though there is no source of alcohol. There is no alcohol going into the instrument through the breath tube or through the cal check port.
This is phantom alcohol. The display is reading alcohol where there is none.
Phantom alcohol can be displayed at any time when zero has accidentally been set on the DVM scale at a position below true zero. In the past my experiments have demonstrated that this can happen as a result of radio frequency interference (RFI). This is an example of the same phenomenon happening on an instrument that has not received proper maintenance following flooding of the sample chamber.
I am still in the process of trying to rehabilitate this instrument so that it appears to be reliable and working properly even though it does not receive proper factory maintenance. I will continue to experiment.
Please note that these videos and blog entries are NOT evidence. They cannot be used in Court. Your lawyer will need to retain an expert scientist.