Re: Homebrew programmable slide stainer: opinions and suggestions wanted...

On Mar 25, 11:31 pm, "Y" <a...@xxx> wrote:
Hi all,

I'm thinking on building a programmable slide stainer, starting from a
liniar one. I own a Medite COT20 liniar stainer (see the leaflet @http://tinyurl.com/cgtxo5orhttp://preview.tinyurl.com/cgtxo5). It's in
working condition. It has no less than 27 stations, so it can be used to
perform even rather demanding staining protocols such as those often used
trichromes as AZAN, Van Gieson, Masson etc.

The working principle of liniar slide stainers is, that the slides are run
trough the baths and the time in each bath is the same for all baths. That
way it's possible to run basket after basket after basket... trough the
machine, resulting in very high troughput numbers (the COT 20 should be
capable of staining up to 1 000 slides/hour, according to Medite). Variation
is done by altering the concentration of the solutions used or by putting
several baths after each other.

An example. Let's say staining by hand:
---------------------------------------
...
hematoxylin: 6 min
differentiation 1% ²HCl in ETOH 70: 15 sec
water: 2 min
bluing: 2 min
...

would become something like this, using a liniar slide stainer:
--------------------------------------------------------------
...
hematoxylin 1: 2 min
hematoxylin 2: 2 min
hematoxylin 3: 2 min
differentiation 0.1% ²HCl in ETOH 70: 2 min
water: 2 min
bluing: 2 min
...

But protocols such as this example are impossible in a liniar slide stainer:
---------------------------------------------------------------------------­-

...
mordanting 0.1% chromium trioxide in dH2O: 5h
running water: 5h
safranin 1% in ETOH 50: 20 min
ETOH 50: 10 sec
...

For those you need a programmable slide stainer. The running water isn't the
problem, it's the variation in treatment times.

As I also have a few old PC's gathering dust on my attic (486/33, 486/100,
PI/166, PII/400...), I would like to use one of those to control the stainer
using good ol' DOS batch files.
I would prefer the 486/33 as it is a compact model with an integrated screen
(one of those old Compaq Presario cubes).

The general idea is to control the stainer using an external relay board and
a relay controller board in the computer. Given the age of the PC's I'm
looking for a second hand ISA (prefeberaly) or a PCI controller board.

Controling the stainer shouldn't be that hard as about the only thing needed
is a programmable timing cycle, a square wave with in time adjustable 1's
and 0's. Or am I missing something? I've noticed already that it's rather
simple to set up a timing cycle in NTFS-DOS by using ping 127.0.0.1 -n# but
I suppose it isn't usable in DOS 6.**.

In the next couple of days I will dismantle the Cot 20 and draw it's wiring,
but it looks like it's rather simple: the only thing needed is to control
one cycle of a stepping motor in the machine. It's currently controlled by
an Omron timer. There's no need to controll the other controls on the
machine (draining time controlled by another Omron timer and agitation
on/off) trough the PC.

Any thoughts?

Thanks in advance!

Y.

This is a fascinating idea, although in my experience, for the super
short timed stations, such as with acid alcohol, one aspect to
consider strongly is the "dunk speed". Even with the robotic
stainers, because they're really slow to dunk, you can experience a
"gradient effect" on your specimens, where the bottom of the slide has
sat in the acid alcohol for a relatively significant amount of time
longer than the top of the slide, especially considering the total
time inside the reagent.

You may want to speed up the "dunk" for that station.

Also, is this something that you'll want to make your circuit that
enables this stain type to be enabled and disabled? I cannot fathom
making it run linearly when one of the stations is an acid alcohol
station, but I CAN imagine you inserting a rack, hitting a button, and
at time point # whatever, it speeds up the transfer rate, then slows
back down. This will clearly restrict you to a single rack of this
type unless I'm missing something.

Perhaps it'd be easier to use something like the RA Lamb StainMate for
these special stains? The COT-20 is a great linear stainer (as is
pretty much everyting Medite builds), please keep us up to date.
Perhaps consulting also with histonet would be wise.
.

Quantcast