tank size vs acid concentration

[whistule]

Ah-ha. Yeh I tried anodising again at 4.5 A/sqft and got different results but as you point out that isn't really a problem.

What are the disadvantages of weakening the acid down further to minimise dissolution

Someone suggested using a welding machine as it has a current control - would this work as a current soirce or is it a bit crude?

[whistule]

Sorry - another question! I presume you can't measure the resistance directly because of the effect of the power supply in the circuit.

[Fibergeek]

Weakening the acid further will reduce dissolution and decrease pore size. I haven't tried acid concentrations lower than LCD, I would be concerned about the pore size becoming small enough to make a new set of dyeing problems. You can try it, but I don't know where the lower limit for acid concentration is. Lowering the temperature is a better but more expensive way to reduce dissolution, this doesn't affect pore size.

You are correct about not being able to measure the resistance directly, but since you can measure the voltage and current you can easily calculate the resistance:

Resistance equals the Voltage measured divided by the Current measured. R = V / I.

One caveat though:
If the anode connection to the work is allowed to degrade, this won't work, you would not know if the resistance is that of the anodize layer on the work, or the resistance of the degrading electrical connection. You need to have good electrical connections.

[whistule]

Yeh, recent runs have shown a restistance fluctuating between 3.4 and 3.6 ohms. I presume the peak in the restistance when the layer is at its thickest should be fairly obvious? This is perhaps where my electrical connections come into play.

P.S. I see what you mean about the benefits of a constant current source !!

[Fibergeek]

It sounds like you have about 1 ohm of contact resistance, which is too much to allow reasonably accurate measurements.
If it always stayed at 1 Ohm, you could just subtract it out of your calculations. The trouble is; it won't be constant, it could easily be much higher or lower, which make your measurements worthless. Even worse; a poor connection will show increasing resistance as anodization proceeds, just like the resistance of the anodize on the work. There is no substitute for good electrical connections.

You are correct assuming that the peak resistance indicates peak coating thickness. Provided that you were not fooled by connection resistance.

[whistule]

What I mean is how obvious do you find the peak to be i.e. what's the diffence between starting and ending resistances.

[Fibergeek]

How obvious the peak resistance (PAR) is depends on current density and dissolution rate. If you look again at Figure 2; Anodization Curves for Various Current Densities, at the rear of the LCD manual, you will get the idea. At higher current density (12 a/sq.ft.) the peak is dramatic and hard to miss. As the current density gets lower the peak becomes shallow and occurs very slowly. If the conditions are right, equilibrium between anodization growth and dissolution can occur (growth rate = dissolution rate) and the resistance rises to a peak value but doesn't then go down. If this occurs anodize for 15-30 minutes longer; to be sure you haven't been fooled, and then stop.

The resistance observed in the anodization (PAR) starts at near zero Ohms at the start of the process, and finishes around 2.5 Ohms /sq.ft. at the end. Be sure to remember that the actual measured resistance is inversely proportional to the surface area of the work. In other words; if you see 2.5 Ohms when the work is 1 sq.ft., you will see 5 Ohms if the work was 1/2 sq. ft., and you will see 1.25 Ohms if the work was 2 sq.ft. This all assumes the exact same current density. This illustrates why it is important to know the actual surface area of the work.

[whistule]

Finally I can read current and voltage together. I cannot seam to get any sort of peak in the resistance. It starts around 3 ohms and stays there usually dropping off to 2.8 ohms towards the end of the process. I improved my connections between the hanging rack and the parts thinking it was that but it made no differnce. I have been anodising at 5A/sqft for 70 minutes and the dye is taking quite well but it's difficult to optimise the layer thickness without seeing a peak in the resistance. Could it be the wire > part connection? I thought this one wouldn't be so important as it is immersed in acid. I can't drill and tap the part so it's just a spring clip into a tube - not ideal but as yet the connection has never failed - it's just whether that might be affecting my results. Or perhaps acid concentration? Any ideas? It's got me stumped.

At 5A/sqft starting voltage is 15V ending at about 14.8V.

[Fibergeek]

Since this is an electrical circuit all of the connections are important, any one of them being bad will foul you up.

The one connection that will be the most difficult is the anode wire to the work connection. This connection is under constant attack from the anodization itself. If any electrolyte can get between the anode wire and the work; it will anodize this connection point, the resistance goes up, and results as you describe occur. If this connection wasn't in the electrolyte contact degradation would not happen. The 1 Ohm or so of contact resistance you are seeing means that the anode to work connection has a very thin film of anodization in the connection point, probably less than 1 micron thick. It doesn't take much.

The cathode connection does not experience this contact degradation because it is on the negative side, no anodization occurs here.

If you are using a spring clip for the connection, the spring tension needs to be much higher. I hope this clip is made of titanium, aluminum is too soft to provide much spring tension.

If you must use aluminum (cost and availability issues with titanium) can you make something like an all aluminum screw clamp?
It would be a very small C clamp; with an aluminum machine screw with a sharpened point, that bites into the work under the tension provided by tightening the screw. If the opposing side of the screw clamp (the anvil side) also contacts the work with a sharp point, the unanodized connection area is minimized. The anode wire is attached to the clamp by crushing it under the screw head with an aluminum hex nut. Now you don't need any holes in the work, and you made a real tight all aluminum connection. I have done this before, it does work.

I don't understand what you mean by "at 5A/sqft. starting voltage is 15V ending at about 14.8V". As you know by now, if you start the anodization at any voltage other than 0V (or close to 0V) the temporary short circuit can, and usually does, damage the connection immediately. I'm sure you aren't doing this, could you explain your statement further?