Getting my Pharmacia LKB Multidrive XL online... now with 3D printing!

Quite a while back, I bought a used Pharmacia LKB Multidrive XL on eBay. It's a gel electrohphoresis power supply that can output up to 3500 volts at several hundred milliamps.
There was a problem, though: Whenever I tried to use it, it just gave me the error "Invalid connector". The unit didn't come with a manual, so I was stumped as to what exactly this error meant.
Searching for the manual online didn't yield anything besides a few old forum posts with broken links. The manufacturer, Pharmacia, had since been bought be Amersham, which in turn had been bought by GE Healthcare. It seemed like a long shot, but I rang up the GE Healthcare customer support and asked for the manual. They were amazingly helpful and sent me a PDF.
Armed with the manual, I learned that the error I was getting was connected to one of the many safety precautions that were built into the device.
First off, the unit has four power outlets. All four need to be plugged in with something, either a power cord or a dummy connector - otherwise, the unit won't output anything. 
I eventually came across the three dummies that came with the device while moving. I thought that might fix the problem, but the device still gave me the same error.
Reading the manual further revealed that the device's safety measures include what the manual calls coded sleeves - specific pieces of plastic that are slid over the dummies and power plugs. In a dummy, they don't do much except be there, but for the power cord, they code for the voltage which the cord is rated for - 500V, 1000V, and 3500V, respectively.
And it turned out that this coded sleeve was missing for the power cord.


So in a leasurely Sunday afternoon session, mltn and I set about reproducing the missing sleeve. mltn attacked one of the sleeves with a slide ruler and rebuilt it in openSCAD. We then printed the thing on an Ultimaker at the Raumfahrtagentur, and - lo and behold - it fit and worked on the first try.

Rendering of the model for the generic coded sleeve. To accommodate the limitations of our 3D printer, all four legs of the sleeve as well as a bridge to hold them up (cylinder on the very right) were included; the bridge and the two middle legs were then snipped away after printing.

Well, almost. Since the recognition of the sleeves employs light sensors, and there happened to be transparent filament in the Ultimaker when we printed, we had to paint the printed sleeve with a black marker.
I hooked up a multimeter to the Multidrive and confirmed that it was really putting out the set voltage. Which is particularly satisfying as a search for replacement parts for this unit yielded - unsurprisingly - nothing.
We only made the lowest, 500V version of the coded sleeve since I was unsure what the cable and connector that came with the unit had been rated for, but it's nevertheless satisfying to turn a non-operational device into a pretty nifty gel electrophoresis power supply with a bit of 3D printing.


Granted, it would have been possible to make this work with a few well cut-out pieces of cardboard or acrylic, but somehow, making an actual sleeve seemed a little more robust.

Anyway, here's the quick-and-dirty OpenSCAD code for the sleeve:
 
$fn=100;
//main
difference () {
 hull() {
translate([14,0,0]) cylinder(r=7.5, h=31);
cylinder(r=7.5,h=31);
 }
 hull() {
translate([14,0,0]) cylinder(r=7, h=31);
   cylinder(r=7,h=31);
 }

translate([7,-8.5,8]) rotate(270, [1,0,0]) cylinder(r=1, h=5, $fn=100);
translate([7,-8.5,27.5]) rotate(270, [1,0,0]) cylinder(r=1, h=5, $fn=100);
}
;
//side
difference () {
translate([21,-0.6,0]) cube([10.5,1.1,33.5]);
  hull() {
translate([23,-0.7,6.25]) rotate(270, [1,0,0]) cylinder(r=1.75, h=1.2, $fn=100); 
translate([24,-0.6,4.5]) cube([8,1.2,3.5]);
 }
 hull() {
translate([23,-0.7,13.85]) rotate(270, [1,0,0]) cylinder(r=1.75, h=1.2, $fn=100); 
translate([24,-0.6,12.1]) cube([8,1.2,3.5]);
 }

 hull() {
translate([23,-0.7,21.75]) rotate(270, [1,0,0]) cylinder(r=1.75, h=1.2, $fn=100); 
translate([24,-0.6,20]) cube([8,1.5,3.5]);
 }

 hull() {
translate([24,-0.6,28]) cube([8,1.2,6]);
 }

minkowski() {
 translate([19,-0.6,31.5])rotate(270, [1,0,0]) 
    cylinder(r=0.5, h=0.6, $fn=100);
     cube([3,0.5,3]); }
}
translate([32,0,0]) cylinder(r=1, h=28);
 
 
Edit: Some commenters asked for the manual of the MultiDrive. It's available here.

Comments

  1. Hi,
    i have the same Problem.
    Can you sell me one ?

    Thanks

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete
  2. Please could you upload the manual somewhere as other people need it?

    ReplyDelete
  3. Hi
    that's realy usefull, just got one on ebay and trying to get it going.
    Any chance of sending me a link to the manual if you still have it?
    Thanks
    Kj

    ReplyDelete
    Replies
    1. If you still need it: I added a link to the PDF at the bottom of the post.

      Delete

Post a Comment

Popular posts from this blog

Charting a course to hands-on DNA sequencing with the Oxford Nanopore MinION

Google Xistence - a new approach to identity