Sciencemadness Discussion Board

DIY Anodising dye

rocketscientist - 18-5-2011 at 04:08

Since my friend owns a DIY CNC machine, I have been looking to machine a custom waterblock for my processor. However contact between copper and aluminium often leads to galvanic corrosion, the cheapest way to avoid which is apparently by anodising the aluminium components.

My friend has ample experience with anodising, however, the dye remains very expensive, and despite hours of google I couldn't find any decent references to its makeup. Looking at pictures it just looks like a mixture of some fine powders.

I see no reason not to have a go at making my own. I've so far established that some dyes (yellow) contain Ferric ammonium oxalate. Can anyone provide any more info?

[Edited on 18-5-2011 by rocketscientist]

Phosphor-ing - 18-5-2011 at 06:09

Google "Anodizing Dye Msds"

I found several different colors on the first page of results. Should give you some direction depending on what color you are after.

Squall181 - 18-5-2011 at 06:25

I've successfully anodized a machined aluminum piece and then dyed it red using Rit clothing dye. I am not sure if all of the colors will work. I've also heard of people using Dylon brand clothing dyes and they seem to produce good results.

Wizzard - 18-5-2011 at 06:52

Why not just use all copper? :)

watson.fawkes - 18-5-2011 at 08:34

If all you want is galvanic separation, you don't really need a dye. You might want one to ensure complete coverage of the anodizing process.

Alternately, tin the copper completely.

rocketscientist - 18-5-2011 at 09:02

Quote: Originally posted by Wizzard  
Why not just use all copper? :)


Several reasons: Firstly copper doesn't machine anywhere near as well as aluminium. Secondly, since the top half has to be quite thick (rigidity), the copper block would have to have large dimensions and therefore be expensive. Furthermore, I would have to machine copper outlets for the pipe to go on. In short I'm trying to save money.

I was considering anodising without a dye until I read the wiki article, in particular:
Quote:

However, these same pores will later permit air or water to reach the substrate and initiate corrosion if not sealed. They are often filled with colored dyes and/or corrosion inhibitors before sealing.


The MSDS sheets are a wealth of info, but considering Rit is extremely cheap, tried and tested I'll most likely just use that.

Xenoid - 18-5-2011 at 12:00

Yeah! I've used Dylon clothes dye (purple), worked fine. I've also used food dyes, although they are supposedly not recommended. Red (Amaranth, E123) and yellow (Tartrazine, E102) produce particularly good results, the latter, a lovely golden colour.

Experiment with concentrations and times as it is quite easy to get the colours too dark!

rocketscientist - 18-5-2011 at 12:12

Quote: Originally posted by Xenoid  
Yeah! I've used Dylon clothes dye (purple), worked fine. I've also used food dyes, although they are supposedly not recommended. Red (Amaranth, E123) and yellow (Tartrazine, E102) produce particularly good results, the latter, a lovely golden colour.

Experiment with concentrations and times as it is quite easy to get the colours too dark!


what concentrations do you use, just as a starting point for me, if you don't mind me asking?

phlogiston - 18-5-2011 at 13:35

This guy also describes using Dylon, and has some pictures of the results:

http://astro.neutral.org/anodise.shtml

Also, I believe a commonly used black anodizing pigment, nigrosin, is also available as old fashioned pen ink (look in art suppliers stores). It's not extremely expensive.

Although it is ofcourse Wikipedia, and therefore not very reliable I interpret the quote rocketscientist gives as 'you need to seal it, but not necesarilly put in a dye to close the pores'. The sealing is a different process (heating in steam is a common method).

Xenoid - 18-5-2011 at 13:38

I'm not sure about concentrations, the food dyes were just made-up liquids from the supermarket. The DYLON was in a packet designed to be added to a washing machine, I just made up a darkish-purple solution. I think it is better to use a weaker solution for a longer time as this gives you more control of the desired shade. The thickness and density of the oxide coat will also be a factor (voltage and current used for anodising).

Make some Aluminium test pieces and try out different electrical conditions, times and colours.

rocketscientist - 19-5-2011 at 18:16

Quote: Originally posted by phlogiston  
I interpret the quote rocketscientist gives as 'you need to seal it, but not necesarilly put in a dye to close the pores'. The sealing is a different process (heating in steam is a common method).


That's exactly what I meant, however I was completely unaware of using steam. In light of the fact that dye is cheap and easily available I could just as well use it, especially as it would result in a nicer finish.

Would it be beneficial to use steam to seal the pores after anodising (with dye)? I was thinking of possibly using a pressure cooker?

Xenoid - 19-5-2011 at 19:34

Steam..... I just put stuff in boiling water for a few minutes, seemed to work OK!