How to eliminate anodizing appearance variations

A. Hi Vicky. Because this is a known problem to you, you should probably run coupons at the beginning and end of the run and maintain them on a sample board, and not run production until the coupon is right.

The problem could be material differences; or anodizing thickness; or dye temperature, concentration or time; or be related to sealing. But coupons will allow you to eliminate the first possibility, check on the second possibility, and move on to the dye and seal steps.

Luck & Regards,

A. What grade of aluminum are you using, 606x?

Quite a coincidence, but I just began a consulting project for a client in a similar situation. They require, but do not yet get, a perfect match.

We are going to do some microscopic profile analysis to see if we can actually quantify in geometric terms what causes the difference in appearance. Sometimes such an approach works. However, sometimes it doesn't because no matter how hard you work, you can't see the forest when you're looking at the trees, and you can't see the trees when you look at the forest.

My first supposition, by the way, is the same as yours, i.e., that the etch time and activity is the biggest factor in the difference in appearance. I wouldn't attribute the activity solely to pH, however; temperature and dissolved aluminum concentration are surely very important.

A. Ted is correct that the etch bath parameters are very important. The aluminum content in particular is very important in determining the matteness obtained from the etch bath. One of the problems with using reclamation units, at least when they were new technology, was that they would clean up an etch bath too well, that is, took ALL the aluminum out, leaving the etch bath almost useless in producing matte finishes. Vendors came to realize that at least 20 g/l aluminum were required in the bath. If the anodizer dumps periodically, it is a good idea to save 5-10% of the old bath, or process scrap metal in a new bath. I'm going to assume that the anodizer has his etch bath under control in making some other comments. The etch bath is one of the "lifeblood" baths of an anodizing shop, as the appearance of the final finish is usually highly dependent on a good etch. Having run a job shop for several years, a good part of my missing hair can be attributed to etching problems, especially when the cause of the problem ended up as something I couldn't control, such as metal composition. Etching technology has been around for quite awhile, and there are very few papers these days which discuss problems with etching. In the seventies there were several papers given on the effect of metal composition on the matte finish, particularly for the major extrusion alloy, 6063, which I would bet you are dealing with, Dennis. One paper I heard at an AES Conference, by E. C. Franz of ALCOA, helped to solve the exact same problem we were having at our job shop as you are experiencing. Mr. Franz showed that the matteness in etching 6063 alloy is highly dependent on the iron content. He also showed that with an iron content less than 0.2%, 6063 would not produce a matte finish, no matter how long it was etched. The 6063 specs. list the iron content of 6063 as 0.35% max, but the iron is an impurity, and if it is low, the manufacturer doesn't have to add iron to consider the alloy as In Spec. Incidentally, 6463 alloy, a modification of 6063, is the primary bright dipping extrusion alloy and bright dips to a much more specular finish than most species of 6063. However, 6463 will not etch to a matte finish. The main difference between 6063 and 6463? The iron content of 6463 is 0.15% max. Mr. Franz also showed that temper is important in achieving matteness. T6 etches better than T4. A good extrusion shop ages the extrusions to at least a T5 temper, which improves appearance. I don't know if you have any control over the alloy composition or not, but specifying an iron content of >.2% is the ultimate solution. An alkaline etch bath with fluorides will produce a more matte finish than straight caustic. I once had a large job of extruded electric switch boxes where we had to acid etch the parts to produce a matte finish, but that bath had fluorides in it and required special handling. Good luck!

Phil Johnson
- Madison Heights, Michigan
1999

! Wow, Phil. Thank you for the detailed and knowledgable reply. In my case we are dealing with 6005-T5 and/or 6061-T6 extrusions. Presumably we could stick to one or the other, or even have the spec changed to 6063 and >.2% iron because strength and mechanical properties aren't critical for this office decor application.

We're constructing a virtual wall from small pieces, so visual match must be the closest thing in the world to perfect, both for the clear and for the black parts. We had a strange result, even before we've started our instrument work: we took some photos of the extrusions with a digital camera, just for the record before shipping the parts off to the lab, expecting to be able to see no difference at all between the parts on 'film' since they were a very close match by eye. Strangely, they look far more divergent on camera than to the naked eye! I'm suspecting that this may have something to do with polarized light; now that we stumbled on it, we'll see if we can learn anything quantitative in the lab.

A. Hi, Mostafa. It's because of the finely divided silicon and copper left behind when the aluminum dissolves out of the alloy.

A. I would imagine so. 6005 has a higher Silicon content. I don't know that I've come across any two differing alloys that anodize the exact same color (naturally), even if they belong to the same series. This is especially evident in the thinner coatings (you didn't mention the whether it was type 2 or type 3)

A. Hi Josh. It's hard to be sure what you are seeing from a verbal description, but is there any chance the sample parts are hard anodized? Have you, or can you, measure the anodizing thickness on those sample parts? Good luck.

Regards,

A. Like I've said before [Ed. note: thread 22139, "Satin Anodized Aluminum", for example], add sodium fluoride to the caustic soda [affil links] etch in increments of 1 then 2 then 3 avoir. oz/gal

A. Bob Probert's correct. A little sodium fluoride will produce a more uniform, faster, and whiter etch.

A. Tejes, it seems that this is the problem of extrusion and not because of anodise or alloy

A. Hi Akash. It's possible that I am misunderstanding you, but nitric acid is a desmut-deoxidize step, it's not a "bright dip". So the first problem may be a misunderstanding whereby you think you are 'bright dipping' and you're not :-)

Search this site for "aluminum bright dip" and you will read that aluminum bright dips are generally very hot mixtures of phosphoric and nitric acid. There are also proprietary bright dips available such that you can get help on the issue if you don't want to try mixing your own.

But do not attempt to do bright dipping or even design a line for nitric-phosphoric bright dipping until you've seen a couple of installations. It is so aggressive that you need double walled 316L SS tanks, and conventional exhaust systems are not sufficient -- you need a "garage" style containment hood for the bright dip tank and its rinses.

Regards,

A. 1. Inside of billet versus outside.
2. Work hardening by machining.
3. Inside/outside of an extrusion.
4. Agitation in the anodizing tank.
5. Position in the etch tank where gas evolution wipes different areas.
6. In Your case of two different anodizers: free acid, dissolved aluminum, current, temp.
7. dull machining versus sharp machining (read "work hardening").
8. Heat treatment.
The list gos on and on.