Waters 440 Absorbance Detector for HPLC?

[LtStorm]

Not sure whether or not this should go here, but this seemed like the most accurate place since my question is about how an instrument works.

I have an old unused Waters 440 Absorbance Detector for HPLC that I've scavenged for parts to try to digitize an old polarimeter. What I'm trying to figure out is exactly what the light sensor in the 440 is. Removing it from its housing, it's a small silver tube about the size of a small laser pointer. On its front it originally had what I presume was a filter to cut down what it could see to the UV spectrum, as it seems to be set to detect at 254nm or 280nm normally.

Removing the little filter, I see the two little cells that act as the sensors behind another clear lens. The probe has two wires coming from it, ending in BNC connectors that originally plugged into 'Reference' and 'Sample' ports in the back of the 440. What I've done is get a basic post-to-BNC converter and wired the sensor up to a Vernier LabPro that's set to read the differential voltage from the sensor.

At a cursory glance, the sensor looks to act like a photoresistor. The brighter light is, the lower the voltage is detected by the Vernier Box. But, well, I'm not sure if it is a photoresistor, as discussing it with my supervisor he noted that the LabPro shouldn't be sending a voltage across the sensor, meaning it couldn't be a photoresistor. So we're back to square one on how this sensor works exactly.

The 440 is about 30 years old, so the sensor is also too old to have been a photodiode, I've been told, so that's out too apparently.

So, yeah, I'm trying to figure out what kind of light sensor this thing is. I've been able to find painfully little documentation on the instrument (the HPLC has every book for every single piece of equipment it does or has used in the past except the 440) that helps any.

Knowing what kind of sensor it is seems like it would be a great help in deciding whether or not it's actually useful. It seems to be sensing light fine at the Sodium-D line that's being used in the polarimeter this sensor has been attached to.

Of course, I'm also wondering if there is any fundamental difference in how an automated polarimeter works as compared to how an analogue one such that I'm using does. In what it sees to take a reading, that is. In this polarimeter I'm using it's read by adjusting the moving polarizing filter until a vertical line and the rest of the circular view of the source light are the exact same intensity.

Any help is appreciated.

 

[berkeman]

Would it be possible to post a picture of the sensor? Initially I was thinking PMT, but from your description that doesn't sound right. It only has two wires coming off of it? Or are there more contacts? If you hook a DVM to it and measure the output voltage, does it generate anything when exposed to light? Or does it seem to require some bias? Why does your supervisor think the LabPro instrument is not providing a bias? Have you measured what the contacts to the light sensor do when connected to the instrument and varying the light level?

 

[LtStorm]

Hmm, I'll try to snap some pics with the camera in my cellphone, though I don't know how good they'll come out.

Well, after toying with it more, I've come to the conclusion it is some sort of photovoltaic device. After getting the new rigging set up for it, I finally had it so I could connect a multimeter's probes to it directly. It does indeed generate a voltage when exposed to light. I'm still not quite sure what's going on with the readings though.

It seems that it generates less voltage the more light that shines on it. Which is sort of hard to grasp for me. Though, that may be just because LoggerPro is set to read the Differential Voltage, and is doing it backwards, or something... Either way, it's reading, and reading consistently, which is what I need.

Toying with the multimeter I could quickly check through everything; it gave a voltage reading in the millivolt range which changed when exposed to light (should've noted how it changed when the light got brighter or dimmer specifically, I'll do that tomorrow). It reads 0 amps, regardless of light exposure, and a consistent resistance regardless of light exposure. So, yeah, photovoltaic?

As for the wires/contacts; well, it's got two wires because there are two separate sensors in the little module. Each wire coming off of it is two contacts in a BNC plug.

I've also troubleshot some of the other problems with the experiment. Such as the fact the unshielded wire was laying against the metal stand the device is suspended from, so when I would rest my hand on the stand's base while adjusting the polarimeter it would cause the sensor's readings to deviate. I've gone back and made sure everything is grounded/not touching any outside conductors, and it's much more consistent now.

Of course, now comes the question of how to interpret the results I'm getting from it... Anyone know what kind of data modern digital polarimeter output? What I get from mine is essentially a sine-wave shaped graph of voltage vs. angle, with the angle varying from 0-360 degrees. From what I can see so far, optical activity causes the peaks/valleys of the graph to be down/upshifted based on how much rotation there is. So normally there's a peak at 90 degrees for distilled water/a blank; for a solution with a high rotation, such as 100g/100mL of sucrose, the peak is offset to about 50 degrees.

Going to try some more sugars at varying concentrations to see how their different rotations affect things, then probably going to try running some mixtures of the sugars to see how the two effect each other's readings. As the biodiesel I want to test may look weird in the polarimeter because of two or more optically active compounds in it.