distilled vs. tap water and anodize time too short

[M_D]

When I read the part about the lead cathode be cast on a piece of aluminum, I thought that is most likely the problem. If the aluminum had enough area alone, and the other connections were ok, it may not be a problem. But I am guessing that you are depending on the lead plate for enough area, and I believe chances are very high the connection there is your root problem. You could check it with an ohm meter, but it won't have the load on it and may show it is ok. If it shows any amount of resistence, it is bad for sure.

Also, it helps to have the cathode area split up, with part in between to cathodes with 2-3" minimum clearance. | x | The x would be a part, or parts, and the uprights the cathodes.

The fact that you got 1 part to turn out by messing with the amperage and time says the chemicals aren't totally messed up. They may not be optimum, but everything points to a bad connection and cathode arrangment being out of wack.

I could be wrong, but I bet if you found some aluminum sheet and made 2 plates that extended above the electrolyte with good connections that you would be steps ahead of now. The whole way you describe it says you aren't getting the right current to the part surface.

[blackcote]

M_D:
Ok. cathode is not something I checked on. I suspect it is good. but for the small pieces the Al part of the cathode has plenty of SA. It's 2" wide by about 12" in the solution. The lead lays on the bottom directly below the part. I will try one on either side once and see how that works. Lots of stuff to try this weekend again. Hope I find something.

neilfj:
If that's how the rule works then I am way off on anodizing time and amps. 2.5 min would be a dream come true. Maybe I'm doing it wrong though. This is how I thought it was supposed to be....

part .06 SF
desired CD @ 18 A/SF
18*.06 = 1.08 amps
720/1.08 amps = 666.67 min/SF
666.67*.06SF = 40 min (which coincides with most standard anodzing info I've seen)

Is this wrong?

agitation defintely helped get a more even finish. but not darker. it is a aquarium air pump with the tubing weighted down ant the air bubble direct to directly pass by the part.

connections are turned in untill the wire is twisting. How in the world do you do a part with no threaded holes or even no holes at all?

Ponder this with me... IF my connections are bad wouldn't the wire that I'm using for a connection get blacker then the part when dyed? And not the same shade? Especially since I'm in CC mode and amps are not being changed. Now you say the voltage would change, but I never saw the voltage drop on any of these experiments. It is possible that I missed it... but I wonder...

Thanks for all the help so far. I hope to be able to return the favor someday for someone else.

[neilfj]

Justin;
I think I screwed the formula. It didn't sound right when I posted it, but just figured the math was right, so the answer had to be correct. Let me double check and get back to you.

[neilfj]

Originally Posted by blackcote

M_D:
Ok. cathode is not something I checked on. I suspect it is good. but for the small pieces the Al part of the cathode has plenty of SA. It's 2" wide by about 12" in the solution. The lead lays on the bottom directly below the part. I will try one on either side once and see how that works. Lots of stuff to try this weekend again. Hope I find something.

That may contribute to the problem. On my setup I have the plates on the sides, with the connections above the level of the electrolyte to eliminate any potential problems.

neilfj:
If that's how the rule works then I am way off on anodizing time and amps. 2.5 min would be a dream come true. Maybe I'm doing it wrong though. This is how I thought it was supposed to be....

part .06 SF
desired CD @ 18 A/SF
18*.06 = 1.08 amps
720/1.08 amps = 666.67 min/SF
666.67*.06SF = 40 min (which coincides with most standard anodzing info I've seen)

Is this wrong?

I screwed up big time. I factored in the sq/ft of the workpiece twice. M_D pointed it out in another thread. As M_D stated, the size of the workpiece is already being taken into account when the amperage is set on the power supply. By including it in the 720 rule calculations, it gives the wrong result. I've seen the same rule as you that includes the sq/ft as part of the equation, but they aren't taking into account the size of the sample by adjusting the current setting on their power supply, which we are doing here.

The 720 rule is:

720 minutes divided by current density x desire layer thickness.

Using your example, your power supply would be set to 1.08 amps Constant Current.

To Get a 0.5 mil thickness (12 micron), it would be:
720/18 x 0.5 = 20 minutes.

You can adjust the time by raising/lowering either the current density, or the desire thickness. I wouldn't go below 0.33 mils though. Dyeing quality seems to drop off under this point. My recomendation is to go no lower than 0.40 mils (10 microns).

[b]
Ponder this with me... IF my connections are bad wouldn't the wire that I'm using for a connection get blacker then the part when dyed? And not the same shade? Especially since I'm in CC mode and amps are not being changed. Now you say the voltage would change, but I never saw the voltage drop on any of these experiments. It is possible that I missed it... but I wonder...[/quote]

As with everything else with anodizing, it depends. I've seen them do that, I've also seen them burn, dissolve, and remain the same color as the part being anodized. This type stuff drove Fibergeek and me nuts for months. Everything would look good as far as the readings of amperage and voltage are concerned, but the anodizing still failed. We have a number of theories, but no real evidence of what the reasons were. What we did determine is that it was a connection issue. I started to use 1100 alloy wire which was softer than what I was previously using, and the success rate jumped. This was because the softer wire conformed to the threads and tended to cold flow itself into the them. Once Fibergeek came out with the prototype of the CDW welder, I've had 100% success. It may not be much of an answer, but it is the best I have. Once we eliminated all the variables, the only thing remaining was the electrical connection. We were even monitoring voltages and amperages to 0.001 resolution to try to track this down.

[quoteThanks for all the help so far. I hope to be able to return the favor someday for someone else.[/quote]

Sorry for screwing up and almost sending you in the wrong direction.

[Fibergeek]

I know exactly what you mean...Ask Fibergeek about how stubborn I am!

I don't think its accurate to say that Nellfj is just stubborn, he makes the proverbial 19th. century army mule look open-minded in comparison.

Allow me to comment on Caswell's position regarding non-LCD anodizing variants:

As has been stated many times; Caswell will only support LCD, because he has the documentation and knowledge base to do this properly. This is both correct and entirely fair, remember he is operating a business that has the LCD process as one of his products. Caswell is determined not to allow any of his customers remain unsupported when they are using his process (LCD). This demonstrates a solid business ethic.

As is obvious; there are countless varieties of the anodization process, and this gets tricky real fast once you have stepped out the baseline LCD method. Those participating in this thread have hands-on experience with developing certain variants, and are willing to share what they have learned. These guys are certainly not "internet armchair experts" pay close attention to what they say.

The one common thread; about all of them started with LCD "by the book" once they got everything working and understood why it works they were prepared to make modifications to suit they own specific applications.

Speaking for myself, this is how it should be: no one process can be everything to everybody. If you go off on a tangent BEFORE you really understand everything that is going on you going to get seriously confused; and it will take a lot more time, expense, and trouble to get to the particular anodizing method that's optimum for you. You have to learn to walk before you can run.

Caswell is prepared to fully support LCD; there isn't much in LCD that he hasn't dealt with already, seen and solved. Weird stuff will arise occasionally, he gets me involved privately if he needs assistance or just isn't sure.

Edited mostly for spelling.

[neilfj]

Originally Posted by Fibergeek

I don't think its accurate to say that Nellfj is just stubborn, he makes the proverbial 19th. century army mule look open-minded in comparison.

All this because I don't believe a word you tell me unless I can see and prove it to myself! A mule though? Like you can't be a horse's "patoot" yourself!

[Fibergeek]

Have I ever claimed to NOT be a horse's patoot?

[M_D]

Since the aluminum strip which the lead plate is molded to is large enough for the part, it may not be the entire cause of trouble after all, although it is best to have cathodes on at least 2 sides, at least I see the difference in my work.

Those really high current density numbers are still say something is screwed up big time, not that you didn't already know. But if the parts were actually seeing those numbers, you wouldn't have got the one good part with the time and method.

So I was pondering the possible cause of what might make sense. It all comes back to a bad connection somewhere, or the instrument readings on current are out of the ball park. If it were me I would look at that. Are you using a CC power supply with built in meters?

You did get a huge increase in tanks temp though, which indicates it saw some amperage. What size tank are you using?

I think starting with the LCD method is a good stepping stone also to get the basic process down. We use higher current densties now because it gives quickers cycle times, but started with 4.5 amps per square foot.

[blackcote]

ahhh. I feel the love on this board. You guys are great.

neilfj:
No need to be sorry. Thanks for takin the time to edit your posts.

Fibergeek:
I understand caswell's point of view, but thought that it was uncalled for. Maybe he thinks I'm just surfing around on the net and looking for a free place to get help. But I already bought my dyes and chemicals from him and was under the impression I could post here becuase I "paid" for the privilege. sorta. The idea that he wants me to purchase chemicals twice just rubbed me the wrong way. I apologize if I was out of line.

And I see your point about starting with LCD. However I thought that to be uneccessary since I had gotten it to work about a year or so ago with all else the same except power supply. I plan on giving the LCD a try so I can see how long it takes to reach PAR.

M_D:
I'm leaning toward the power supply guages being the culprit now w/ some possibility of the connections. But I will eliminate connection issues as best I can on Saturday. And try the plates on either side. I will also attempt to verify that the guages are at least close. I do remember reading voltage with a meter that was about 1.0V below what the PS guage showed.

PS is CC w/ guages and w/ voltage knob maxed. and current at 0. "HOWEVER", it should be noted that the current guage never shows zero (shows about .2 amps), except when it off or not connected to anything. I was not able to verify if there was current actually there. My meter didn't pick up anything. so I'm thinking verifying current would be a good place to start.

[Fibergeek]

Blackcote,
The chemicals used in LCD are exactly the same as are used in standard anodizing, the only chemical difference is that the electrolyte is of lower acid concentration, and Caswell doesn't sell sulfuric acid (the hazmat charges are murder).

Use the 720 rule, PAR has turned out to be more difficult to use than I thought it would be. If you go to racking (common in commercial uses) the parasitic resistance of the rack will make PAR totally undetectable. We need to reduce the level of confusion.

Regarding CC power supplies; take a look at a thread entitled Constant Current Issues further down this page.