Consistency Problems

[NeoMoses]

I'm nearly at the end of my wits. I have been anodizing using the LCD method for about 2 months now. When I first started, I had no problems with consistency. Parts that were anodized took color well and were great. When I first converted, I used my existing electrolyte and just watered it down to get close to the LCD concentration. After about 10 batches without problems, I decided it was time to get a larger tank (which I had been wanting) and make a fresh electrolyte bath. I used a 30 gallon rubbermaid tote as the tank and prepared 15 gallons of electrolyte solution according to the LCD instructions.

My power supply is a 5A, 30V lab power supply that can run in CC or CV mode. Obviously, I use CC for anodizing.

So, with the new setup, I'm getting about 50% of my parts to turn out correctly, and I'm getting really tired of stripping and re-anodizing parts. When parts are good, they're great. Here's a pic:



However, when parts do not turn out well, they always have spots on them that do not dye properly, as in the pic below: (you may notice that there is some etching on the part... that's not the problem, the problem lies in the large discolored areas...)



I don't have this tank set up to measure PAR, so I'm anodizing according to the LCD instructions, 4.5 A/ft^2 for 90 minutes. Dyeing is done in Stainless Steel pots on the stovetop. Sealing is done in boiling distilled water in a SS pot.

Initially I thought these problems might be caused by bubbles sticking to the surface, so I tried adding an agitation system to the tank. It does not appear to help, as I'm still having the same problems with or without agitation. Here's a pic of the agitation system in action:



My last possible idea is that this may be something left on the surface, some sort of grease or something. I do not use Caswell desmut, I simply wash the parts in hot water/dish detergent. Does anyone else have any ideas to help me out? I'm stumped here.

[neilfj]

I've run into a similar problem and found that the washing in dishwashing soap would clean the surface, but wouldn't do a complete job of degreasing it. I also suspect that some soap brands actually leave deposits.

An easy way to determine this is to pick up a small bottle of Simple Green. First, wash the part in dish-soap as you currently do and after rinsing make note of how the water sheets off the part. If the rinse water beads up, even slightly, then you have a surface contaminant present.

Take the Simple Green and spray it completely over the surface and scrub it with a toothbrush and rinse. The water will completely sheet off the part, and if it doesn't repeat the simple green treatment. If you notice an improvement in the sheeting action between the dishsoap and the simple green cleaning, then you can reasonably guess that the dishsoap is the culprit.

I'm not sure if Simple Green itself will cause any issues with the anodizing process, so I follow it up by using Caswell's degreaser.

My process is to wash in dishsoap, rinse, wash in Simple Green, rinse and then use Caswell's degreaser to make 100% sure it is thoroughly cleaned and degreased.

Another potential area to investigate is that you may be getting some shadow effects where the electrical field is weak in one area due to placement of the anodes/cathodes, or any other item that may be interfering with the electrical field.

[potsked]

just out of curiosity, you wouldn't happen to be anodizing these parts for logic cooling would you?

[NeoMoses]

just out of curiosity, you wouldn't happen to be anodizing these parts for logic cooling would you?

Nope, those are personal pieces (notice the DTek Whitewater... Definately not a Logic product). Anodizing is mostly a hobby for me, I haven't been paid to anodize anything in about 9 or 10 months. For those who don't know, I used to own Logic Cooling, so if you see me anodizing lots of Logic stuff, it's most likely left-overs.

[Fibergeek]

Regarding the white spots; this can also be caused by not removing all of the native oxide from the work, or waiting too long between cleaning and anodizing. "Native oxide" is the naturally occuring aluminum oxide film that forms on raw aluminum by exposure to air, kind of like rust on steel. The presense of this film will hamper anodizing or block it entirely. I see in your pictures that you aren't beadblasting, this is the fastest and easiest way to remove the film. You probably used sanding to remove the film; the white spots are shallow depressions in the surface where the sandpaper could not reach, these depressions failed to anodize properly since they were blocked. A strong alkai like sodium hydroxide (lye) or any other strong alkaline chemical will remove this oxide. I think simple green does, which is why Neilfj's cleaning method works, he isn't beadblasting either. Caswell anodize stripper would work well too. Desmut/deoxidizer is intended to remove this film, and is probably the best chemical method. In all cases the chemical should be used warm; between 100 and 140 deg. F. Through rinsing after this treatment before anodizing is essential.

I don't understand why you think you "aren't set up" to measure PAR; since your lab power supply has current and voltage meters/readouts, all you are lacking is a calculator (or a pencil) to do the arithmetic. If you just try it once you will know when to stop the process, you will immediately see the advantage and you will never anodize blind again.

[NeoMoses]

I've never used the readouts on the Lab power supply for calculations because they are analog gages, and voltage is displayed 0-30V in 1V increments... pretty coarse. I've ordered another multimeter that I can dedicate to anodizing, so hopefully I'll have a little better idea of my voltage profile.

Interesting thought about native oxide... I usually prefer a gloss finish over a matte, so I don't do much bead blasting. I think I'll give that a try and check my results. I just noticed that my pieces that have a simple geometry (AKA the red,flat waterblock tops) turn out well, while the more complex geometries have a higher chance of messing up... It's likely due to the fact that I sand all flat stuff on a surface plate, and the entire surface ends up being flat. Most curvy stuff is just sanded by hand, and could easily have spots that aren't sanded as much as others.


About the Lye/DeSmut... I'm gonna order some Desmut from caswell, but it'll likely take 'til Wed. or Thurs. to get here. I've got some Red Devil Lye here. What concentration of lye should I use? I'm assuming something pretty weak, because I've seen strong Lye concentrations start to pit aluminum. Also, how long should I leave the parts in the warm lye? Should I just use the same amount of time as is called for in the desmut part of the LCD instructions?

[Fibergeek]

OK, remember that you need to know what the current is too. You can do this with the 0.1 Ohm 10W resistor and another cheap DVM. You can get adaquate (for this) DVMs for $7.00 ea. now.

Using lye for a deoxidizing etch would use a solution much weaker than for anodize stripping. Start with maybe a teaspoon in a pint of warm (not hot) water and see what happens. You want it to just barely bubble when in contact with the aluminum. When its suficiently etched the aluminum will have a slightly frosted look, but you might be able to get the oxide off before it looks frosted. If you can scrub it while etching it will help. This is much easier to do with desmut. I used the desmut warmer and for a longer immersion than the desmut instructions, because I wanted a matte finish, start by following the desmut instructions.