Sciencemadness Discussion Board

Proposed project: Waters Prep LC500 meets rPi?

3xp0nential - 29-12-2012 at 01:12

Disclaimer: I have never previously used an HPLC system but have a solid biolab bench background and a dozen people around me with relevant experience.

In addition I have a terrible soft spot for solid old machinery and clunky workaroundsTherefore I cannot be sure if the following project is worth undertaking. Your advice is greatly appreciated

____________________________________

I have an old Waters prep LC500 sitting in my laundry. It was checked to powerup stage a year or so ago and a mate locally is more than capable of checking it again- he used to design cool astronomy electronics.

The Waters unit had an old tape chart writer and paper tape, probably not working. I'd like to replace this with a digital/analogue converter and port the signal to something like a Raspberry Pi so I can save and export data for later use

I have pretty much all the original documentation for the HPLC, but if anyone out there has a service manual I'd love to get a copy

How practical is this as a project? I suspect I just like the wow factor :) The main consideration is one of consumables for the unit- ie if columns or seals are no longer available there is no point starting.

Waters have been contacted for availability of columns and spares, but I don't expect a reply from them well into January and it's my experience that there are always stashes of spares for instruments which are no longer in use in various labs which are only too happy to shed them from inventory- so informal acquisition is also an interest

As I said above I've not used an HPLC before, but there are dozens of cool applications for one here and I've experienced the steep learning curve in the biosciences over the last few years. I've got several people interested in helping with the restoration who have skills in other fields, and a few more who could happily run a workshop on safe and accurate use

Has anyone here worked with these old units lately? Any comments on their day to day operation and maintenance?

And has anyone successfully integrated old science equipment with the small open-source boards like Arduino or Pi? I'd be very interested in hearing about such projects

smaerd - 29-12-2012 at 06:07

Although I don't know much about HPLC or such instruments some things I would look in to might be OpenChrom(http://www.openchrom.net/main/content/index.php).

I guess that won't help much as it uses a chart recorder. My professor said the way they used to measure peaks from old GC-MS readings was they would use an exacto knife to cut under the curves then weigh the paper and multiply it by the density of said paper, thus giving the volume. Thats if you can get the chart to read out.

In this thread(http://www.sciencemadness.org/talk/viewthread.php?tid=15617) ozone mentions some hard-ware to convert old data acquisition into computerized data acquisition. Could be worth looking into before trying to program your own. Also in this thread piracetam mentions "LV", referring too LabView(http://www.ni.com/labview/).

Not sure if any of this will help but I figured I'd try.

3xp0nential - 29-12-2012 at 14:55

Quote:
Although I don't know much about HPLC or such instruments some things I would look in to might be OpenChrom(http://www.openchrom.net/main/content/index.php).


Many thanks for this! It could come in handy once we get the HPLC setup

Quote:
I guess that won't help much as it uses a chart recorder.


Chart recorder doesn't work and I would rather have a signal converter from analogue to digital so I can run the software you mentioned ( I have no idea how this goes, but a mate with extensive experience in this field got all excited about the possibility when he was here yesterday- so they exist )

Ah- other Arduino science-y things. This is a really good $4 e-book I found, one of the projects is an Arduino PAR logger. PAR loggers are terribly handy as our eyes aren't really good at evaluating and comparing light available to plants


http://www.diesel-ebooks.com/item/9781449338121/Justo-Patric...

radagast - 29-12-2012 at 21:42

I ran into a similar issue with an old Perkin-Elmer Lambda 3B UV/Vis spectrometer, which only contained analog outputs.

I hooked it up to a DATAQ DI-145 device (~$30 new), which included rudimentary data-capture software.

Although the UV/Vis's analog output only roughly correlated to its displayed wavelengths (on its LED display), I believe that this was due to faulty circuitry in the UV/Vis rather than the DATAQ device. In any event, I obtained rough spectral data, which could be in turn imported into R for integration.

I imagine you could do something similar for HPLCs. I'm presently working on restoring an ancient set of Gilson pumps, UV/Vis detector, and will try the DATAQ device there.

By the way, I believe Dr. Bob has some old HPLC columns. I've purchased a large amount of glassware from him and gladly vouch for his credibility.

3xp0nential - 30-12-2012 at 11:49


Quote:

I hooked it up to a DATAQ DI-145 device (~$30 new), which included rudimentary data-capture software.



Wow, thanks, this is a beautiful thing and exactly what I'm looking for!


Quote:

By the way, I believe Dr. Bob has some old HPLC columns. I've purchased a large amount of glassware from him and gladly vouch for his credibility.


Cool, will check him out

Some awesome tips in this thread


snackmonster - 2-1-2013 at 19:06

can be done highly doable. As like the marriage between old modular (analog) HPLC systems and their modern counter parts (Fibre optic).

They have analogue ports which they can support. So I dont see how you cannot transfer analogue to digital or infact just maintain analogue code and mine the info from there.

snackmonster - 2-1-2013 at 19:06

can be done highly doable. As like the marriage between old modular (analog) HPLC systems and their modern counter parts (Fibre optic).

They have analogue ports which they can support. So I dont see how you cannot transfer analogue to digital or infact just maintain analogue code and mine the info from there.

3xp0nential - 9-1-2013 at 15:46

Waters Australia just contacted me to say that spares are still available for that model- they seem really nice

The local rep will be in touch

I'm still very interested in consumables for this unit which may be available as surplus for reasonable cost ( just so they can leave inventory )

There is still a long way to go with this unit, next step is getting it powered up again and checking everything works before we start testing it with columns and solvents. I think I have people round here who can do this without too much inconvenience for anyone

More Updates

radagast - 17-1-2013 at 16:44

Since we're both working on restoring HPLCs, I thought I'd occasionally post updates here.

I tried powering up the main pump (Gilson 305), but the LCD screen flickered while the pump made alarming spastic movements. Since the circuit board appeared normal, I decided to swap out the power supply circuit for a similar working power supply circuit in a working slave pump (Gilson 306). Much to my surprise, the main pump booted up and I was able to input settings and start the pump.

Although the pump moved correctly, it would not pump water from one beaker to another. I removed and disassembled the pump head, and found that the seal was broken. After replacing the seal (again salvaged from the slave pump), I found that the pump would draw up water and spit it back out on each piston cycle.

I suspected a broken check valve, and disassembled the inlet and outlet check-valve cartridges, revealing a tiny pebble (!) in the inlet valve. I think the valves are still probably broken but will post an update after investigating further.

bfesser - 17-1-2013 at 20:08

This thread has gotten me a little excited. I have an HPLC pump in a garage somewhere that needs a little work. I purchased it surplus years ago, and opened it up for inspection. I noticed that there were glass shards and remnants of what appeared to be an optical encoder disc. I determined that it had 180 black sectors (and 180 transparent). I thought maybe I could replace it by printing a new pattern on transparency film (running through the printer again, if needed for opacity) and laminating it to a polycarbonate disc. Has anyone repaired or replaced one of these encoders before? I'll bring the pump home for info/pics this weekend.

[edit]
Quote: Originally posted by 3xp0nential  
The Waters unit had an old tape chart writer and paper tape, probably not working. I'd like to replace this with a digital/analogue converter and port the signal to something like a Raspberry Pi so I can save and export data for later use.

I've actually been looking into a way of emulating a chart recorder using a Raspberry Pi (RasPi) and a thermal receipt printer. I've been tinkering with used and surplus geophones as low-cost seismic detectors, and plan to to analog-to-digital conversion (ADC) to feed data to the RasPi for processing/storage. Eventually, I wanted to have a live printout from the thermal printer, with a trace for the sensor response like an old-school chart recorder. Hoping to develop it (open hardware & open source) to a point where underfunded schools could afford 'seismographs' for their geology programs. I made a post regarding this application at the Adafruit Inc. forum, but haven't chosen the ADC yet.

Something about the idea of distributed network of RasPi geophone monitoring stations makes me feel warm and fuzzy inside. :)

[Edited on 1/18/13 by bfesser]

More Updates

radagast - 24-1-2013 at 17:06

After replacing the check valves on the Gilson 305 pump and hooking it up a manometric module, I found that the pump could maintain a steady pressure of 150 to 200 bars, depending on the prescribed flow rate. (For others using the same device, I'd recommend NOT running the pump without the manometric module, as this could cause the pump to hit pressures outside of its specs).

I scavenged a Rheodyne 7010 injection valve from a broken automatic injector, and configured it for manual injection with a 20 uL sample loop. I could not find a suitable UV/VIS detector in our lab (all five of them gave unreliable readings (!) so I resorted to using a Gilson refractive index detector which outputted its analog signal to a DATAQ-145 A-to-D USB device (as described above). I was gratified to see a sharp peak after injecting a neat sample of polyethylene glycol through an old column. I plan to run more tests and then import the data into R for further analysis.