Guide to anodising aluminium at home

This is an amateur guide to DIY anodising with an emphasis on frugality (at least for the items that it's worth skimping on). It is a work in progess and will likely have updated recommendations as multiple parties gain more experience. Why do you want to anodise aluminium? The thick oxide layer is much harder and resists scratching, plus you can dye it to look pretty with an endless selection of colours

What do I need to get started?

• Sodium Hydroxide (Caustic soda) - Bunnings
• Sodium Bisulphate (pool pH decreaser, alternative to sulfuric acid) - pool supplies, Bunnings...
• Cathodes (cheap:6063 aluminium, gucci: lead) - Bunnings
• Anode busbar - whatever aluminium you can find along with aluminium wire, rod to hand the part from the busbar (cheap:TIG filler, gucci:titanium wire)
• Demineralised water - Bunnings 5L bottle are the way to go, can be used to store chemicals as well since HDPE
• Assortment of buckets/containers (polypropylene) - Bunnings
• Power supply (varies depending on outcome desired)
• Sodium Bicarbonate (baking soda)
• Dye (this is the bit not worth skimping on IMO, you can use some fabric dyes but they arent all that much cheaper than proper dye meant for aluminium) - Caswell AU
• Scale, dyes are on the order of 2-3g/l so get something precise
• PPE...

Chemicals

Etch Solution: 2-3% w/v sodium hydroxide in water (I used demineralised but I doubt it matters for this). Put this in a polypropylene bucket that you can seal for storage - don't forget to label.

Anodising solution: ~20% w/v Sodium bisulphate in demineralised water. Remember, add acid to water, not the reverse. Mix up enough for the size of parts you will be doing in a polypropylene bucket/container.

Dyes: I used the leftover from making the previous solutions to mix up dyes directly in the demineralised water bottles.

Setup

At a minimum you need 4 containers. It's best to do this all outdoors as you will have acid fumes being generated while anodising. Optional: tub of demineralised water for rinsing post anodising, tub of sodium bicarbonate solution for neutralising acid prior to dyeing

Etch tank - as above.

Anodising tank - as above along with cathodes (-) and busbar from which to hang your parts. I'm using 6063 aluminium for my cathodes. Surface area should be around 1/3 that of the parts to be anodised, a lot less and you'll have issues. If using aluminium you'll need to sand off the anodising. Lead is an alternative that's longer life, but $. I used flat bar and bent it over the edge of the container - ideally you want two, one either side. I drill/tapped holes at the top to attach wires, you want good contact here and keep the attachment as far from the acid as possible. Attach the cathodes to the negative of your power supply. It's best to drill your anode busbar so that you can clamp your part hangers for good electrical contact with a set screw, I have a plethora of alligator clip leads so I just clip directly to the part hanger. Agitation is another thing that can be considered to get a more even oxide growth but I so far have not messed with that.

UPDATE: anodising tank temperatures above about 21-22c have a negative impact on the oxide development, I had multiple runs that failed to hold dye during the sealing phase that I traced back to too high an anodising temp (haven't determined how low is too low). I have since been using frozen water bottles placed in the tank to keep the temperature from rising too much to good effect.

Dye tank - I have a polypropylene container from the bunnings storage section for this, I add to it a DIY immersion heater made from a block of aluminium with a couple 40w hotend heaters and a thermistor with a cheapo DIN temperature relay. Dyeing can be done at room temperature for most of the colours based on what I've read but works better/faster at the 60c most are speced at. (You can vary the dye concentration from manufactures recommendation to get stronger or lighter colours, you can mix colours too, or one colour on top of another to get a mix)

Sealing tank - can just be an old saucepan, you can get sealing solutions (haven't tried) or you can use water. I boil the parts for 30mins.

The Process

Surface -> clean -> etch -> anodise -> dye -> seal

• Surface prep

Most of your effort will be towards cleaning and finishing, the starting point effects the outcome, the better the part looks before it goes in the tub the better it'll look coming out. I prefer a satin finish and so far my process is to sand down to at least 240grit then scotchbrite. Have not tried a polished finish yet but probably will soon. I have also tried raw machined parts and if appearance isn't your priority they look fine.

• Clean

Once your surface finish is to your liking you need to clean the part thoroughly. After surface finishing I wash the part in dish soap to remove any oils and loose debris, followed by alcohol/brake cleaner to spray out any blind holes before putting it back into a fresh bath of dish soap with this time with agitation using a tooth brush. Rinse and it's ready for the next step.

• Etch

Etch the surface using Caustic soda, I've diluted down to 2% and only need a one minute dip to get a good etch on most parts. Before dropping the part in the etch tub, attach your hanging wire to it, I use TIG filler rods as they are cheap. Your options are aluminium or titanium (aluminium will be single use unless you remove the anodising from it after each use). Ensure that there is good electrical contact. Place the part in the etch tub, it will start to bubble as the aluminium surface starts to be eaten away. On clean parts this will not take long, before removing the part get ready with a spray bottle of demineralised water to rinse off the part into the etch tank. Give it a good spray down to get rid of as much of the caustic soda then drop it straight into the anodising tank and attach the hang wire to your anode (+) bus bar.

(IMAGE HERE)

• Anodise

Now your part is ready to build up the oxide layer. Place the rinsed part into the anodising tank taking care to ensure it does not short out on the cathodes. Connect the positive of your supply to the part and set your current limit (see next section). Let it run and once your target run time is reached, disconnect power and remove part from the tank, rinse off excess acid by spraying it down with demineralised water as per the post etch rinse. At this point you can dip into your sodium bicarbonate tank if you have one or just thoroughly spray the part down with demineralised water. (the pH of the dyeing step is important so you want to minimise adding acidic/basic additions to it). You now have a anodised part.

• Dyeing

You can now add colour to the newly formed oxide layer if you so wish. Dip the neutralised and rinsed part into you dye tank (see setup). Depending on the colour this step can take seconds or minutes, to ensure an even colour move the part around in the tank. Take it out occasionally to see the progress and keep a note of duration for future attempts. Once you're satisfied with the colour, rinse the excess dye off with demineralised water. The pores in the oxide layer previously formed are now full of dye but the tops are open. I will reiterate that rinsing prior to dyeing is critical, acid remaining in the pores will mess with your dye pH.

• Sealing

To make the part colourfast we need to close off the pores. Commercial nickel acetate products are available to do this or you can just boil the parts in demineralised water for 30 mins.

How much juice and for how long?

Depends on the desired outcome, higher current results in smaller pores and potential for a thicker layer in less time. Larger pores are easier to dye and need a less capable power supply. Smaller pores are harder so more desirable if you need surface resilience, larger are better if cosmetic appearance is the only goal (to a point, turn the current down low enough and the pore size can be too large to hold dye - from what I've read).

I run approximately 1.5mA per cm^2, at that current it takes 90mins to get a good ~25um thick oxide layer. I have voltage limited to 20v but it hovers between 16-18v for most parts (depends how accurate the CAD area calc was). You can drop down to 0.5mA/cm^2 if your intention is primarily cosmetic. 90mins will then net you ~8um. For tolerance do note that the oxide layer grows into and out of the part equally, that is a 10um thickness will make holes 10um smaller (5um from both sides).

Approx thickness in um = (26 * minutes * current) / (surface area cm^2)