DIY xy-stage for i1 Pro 3 Plus (instead of i1IO)

[KaneTW]

I'm looking at buying an i1 Pro 3 Plus for profile creation (seems like the best price/performance I could find).

But if i want to automate it, I noticed the price of the i1IO is kind of crazy high for what amounts to an xy stage that you can mount a spectro on.
Is there something I'm missing that justifies the price?

Any reason I can't just DIY it with an off the shelf xy stage that I'll mount the spectro to, and a bit of software to automate the measurements? (apart from the time investment)

[digitaldog]

You're going to control where each patch is measured (once, multiple times) for any/all patch sets via software how?

[KaneTW]

I have the image file, so I know which spectrum to expect for which patch.
Then find the center of the patch (via scanning up/down and comparing spectrum to expected spectrum) and then just move +- <patch size> to scan each patch.

Alternately, the 16x16mm patches are large enough that you can just center the first one by hand.

[digitaldog]

Again, you're making it sound a lot easier than it might be: you have to control the Spectrophotometer to position, tell it to make a measurement, and move it correctly to the next patch, all using the software to control that hardware. To use an iO, you assign three patches in three corners, and the software then controls its movements correctly for any and all patch sets. How will you control your hardware to do this, you are able to alter the software code?

[KaneTW]

ArgyllCMS already has a working driver for controlling the i1 Pro 3 and is open source. The only work would be
* assembling the xy stage of a desired size
* designing/printing an adapter to mount the spectrophotometer to the xy stage (not too difficult)
* interfacing the xy stage motor controls to ArgyllCMS (likely via an off-the-shelf motion controller and sending G-code to it)

ArgyllCMS is already prepared for XY stage instruments (xy_* functions, read_xy) so I'd just need to implement those.

[digitaldog]

ArgyllCMS already has a working driver for controlling the i1 Pro 3 and is open source. The only work would be
* assembling the xy stage of a desired size
* designing/printing an adapter to mount the spectrophotometer to the xy stage (not too difficult)
* interfacing the xy stage motor controls to ArgyllCMS (likely via an off-the-shelf motion controller and sending G-code to it)

ArgyllCMS is already prepared for XY stage instruments (xy_* functions, read_xy) so I'd just need to implement those.

And

Is there something I'm missing that justifies the price?

Yes, being forced to use ArgyllCMS, not say i1Profiler.
Sounds like you answered your initial question: just build this goodie and use ArgyllCMS.

[KaneTW]

Makes sense. I don't think i1Profiler justifies the extra 4k (i1IO + i1 Publish Pro 3 Plus vs i1 Basic Pro 3 Plus), so if that's the only downside, I'll go with the DIY solution.

[digitaldog]

Makes sense. I don't think i1Profiler justifies the extra 4k (i1IO + i1 Publish Pro 3 Plus vs i1 Basic Pro 3 Plus), so if that's the only downside, I'll go with the DIY solution.

A. So when did you test i1Profiler?
B. Sounds like a plan, go ahead and then report back how it all went.
If B doesn't work out, I have one I'm not using (make an offer). 

[KaneTW]

Heh, sounds good.

I haven't used i1Profiler for printed files before, but for displays I was much more satisfied with Argyll/DisplayCal.

I'll do a prototype using the movement stage of my existing 3d printer once the spectrophotometer arrives.

[Lessbones]

Unless you're looking to measure thick substrates (piece of dye sub metal for example) or backlits, you could look into getting an old DTP70-- you have to use older software to actually drive them, but they're still quite capable, and much faster than an IO ever was for measuring sheets.  If you're trying to do backlits, then I know there's a way to do it with an IO (I think you measure the backlight by itself first to create a compensation overlay for any differences in brightness/color then measure the sheet on it and i1profiler does the work to combine the two) but the best devices use a single point of light and move the substrate over it for consistency's sake.  Shouldn't be too complicated to do such a thing with a bed slinger 3D printer-- but there are so many good devices still out there that can be had for a song if you're willing to deal with ancient software.  Even the old gretag iccolor devices are still relatively viable, as long as you're good with M0.

[bteifeld]

http://www.colorpartner.de/html_english/cs.htm

You might find this to be an inspiration for your work because
unlike the i1io, this device can accommodate a broad diversity of
measurement devices through the use of specific holders. It is also a
gantry-like design.

[KaneTW]

Interesting!

I was busy with my main job for a while, buy my i1Pro3+ arrived last week. I'm using a cheap China laser cutter as the X-Y base and designed an adapter from its "toolholder" (34x3mm piece of sheet metal) to the i1 system with 2 pins in the back.

Going to try out some scanning on Monday.

[GWGill]

ArgyllCMS already has a working driver for controlling the i1 Pro 3 and is open source. The only work would be
* assembling the xy stage of a desired size
* designing/printing an adapter to mount the spectrophotometer to the xy stage (not too difficult)
* interfacing the xy stage motor controls to ArgyllCMS (likely via an off-the-shelf motion controller and sending G-code to it)

A word or warning - I've been asked quite a few times about this idea, and the stumbling block is that there is no easy way of triggering an open ended scan progromatically (that I know of), apart from pressing the button. No problem if you want to do it SpectroScan style - one patch at a time, but a problem if you want to scan. Now it's probably possible to adapt the low level code to start a very long scan and abort it as a way of emulating an open ended scan (which presumably is what X-Rite's software does), but someone would have to actual make those code changes. The most pragmatic thing to do is add a solenoid or servo to press the button.