Gallery, Projects and General > The Design Shop

Automatic paint mixing.

<< < (3/5) > >>

tom osselton:
They made us use densitometers at work trying to get the pictures the same on their two presses as well as cut back on ink they gave you the colour breakdown in %.

S. Heslop:

--- Quote from: Joules on September 02, 2018, 01:05:49 PM ---It was more the feedback could be used to refine the dispensing.  A sample on a piece of clear tape, or mix within a glass dish.  It would also tell you if the mix was consistent...  errr maybe...  or near match to a sample colour.  Create colour swatches etc with defined values, that would change from paint to paint that are used as your primary’s.

--- End quote ---

Oh right. I do think trying to get some actual data would be the next best step. I wonder how accurate those colour sensors are. I was reading about them a bit ago, can't remember why, but they have an array of red green and blue filters over a bunch of light sensors if I remember right. Wonder if that'd affect how the colour is read.


--- Quote from: Will_D on September 02, 2018, 04:36:53 AM ---Surely the most accurate method to dispense paint would be by weighing the amount delivered. Try to integrate a digital scale into your system!

--- End quote ---

For small amounts i'd need an analytical balance. And I have exactly that, but it's display is toast and replacements are £120. I've been sitting on it since replacing the display is a bit on the expensive side, and since it's broken i'd be hesitant to try sell it. If I was smarter i'd try reading the display data and seeing if I could output that to a different and cheaper display.

I still think stepper motors on syringes would be the easiest way to dispense. I've just been looking at the world of solder paste dispensing syringes. They seem alot sturdier than the medical type.


--- Quote from: JerryNotts on September 02, 2018, 03:23:50 PM ---I nearly jumped out of my skin when I read the title of your thread. In a sort of dissolute way I have worked for nearly 50 years in paint laboratories.

One of the paintmaker's long term problems has been colour matching. I have had my share of frustration. I don't know whether I should follow this thread  or not.

My own colour perception is fairly adequate, certainly enough to frustrate any one who apporoaches me with an Ishihari book.

It's not just getting the colour that needs work.

Jerrry

--- End quote ---

This does seem to be a much more complicated topic than i'd anticipated. Matters aren't helped by everyone seeming to avoid answering the questions when asked on other forums.

That said I feel exact paint matching isn't so important for art. It'd be a big deal if youre touching up a wall or your car where you're seeing the two shades against each other in differing light, but for an entire painting they'd only need to be consistent to themselves. But it'd be nice if its accurate too.

rowbare:

--- Quote from: S. Heslop on September 01, 2018, 01:04:43 PM ---Mechanically i'm also considering if there's a better way to dispense. The plastic syringes tend to flex a bit and trapped air bubbles up at the plunger end don't help either. Ideally i'd want something that'd really jet the paint out to minimize strings and droplets hanging on the tips of the nozzles.

--- End quote ---
Stepper driven peristaltic pumps would be a good way to go. They are easy enough to DIY or you can get pump heads on eBay or Banggood for cheap. With a stepper driven pump, you could calibrate them to steps/ml or mg. You can also get small motor driven peristaltic pumps that might do the job. These you would calibrate by time ml or mg per second.

As for scales, you can get cheap mg electronic scales quite cheaply.

Good luck with the project and keep us posted!
bob

JHovel:
I think you are on the right track.
There are plenty of stepper-driven syringes in use in hospitals for all sorts of things including pain killer drugs, even epidurally. So the accuracy of even very small doses of liquids can be controlled well that way.
For this purpose, I actually think you don't even need 5 pumps, you could swap syringes in the same pump, since it won't get contaminated. You just have to load the syringes in the same order each time  :nrocks:

S. Heslop:
I did think about peristaltic pumps but i'm not sure how you'd handle the non linear flow rate. There's probably a way to map it. Plus for viscous fluids I wonder if there's a chance the tube could sort of... collapse. I assume they work on the elasticity of the tube returning it to roundish to draw more fluid in at the start of the stroke, or from pressure on the reservoir to force it in. It seems like alot to go wrong, or change over time, or with different paint thicknesses.

Yeah I think syringes seem most promising. The main issue is how i'd handle (and eliminate) stringing. I've got some stuff to read on that but I think what might work is retracting or suck back. That at the end of a dispensing stroke the stepper can pull the plunger back a touch to hopefully break the flow sharply and maybe pull whatevers hanging on the end of the nozzle back in.

This might be another slow project though. I've been looking at finally building that big 3d printer first since some arrangements I had fell through, but ones that've freed up a bit of money I want to try spend before it slowly disappears on all the Good Stuff I shouldn't be buying.

Navigation

[0] Message Index

[#] Next page

[*] Previous page