Problems with LCD, looking for suggestions!

[Sid03]

I switched to the LCD method and am having trouble. I originally thought it was just a problem with the red dye, but the blue gets no where near as dark as it did before either. The parts anodized with the LCD method dont seem to have as thick of a layer either(they scratch easier). This all seems to have occured since I switched methods. The big things ive changed are the current limiting, done through a light dimmer switch in line with the power to my battery charger. The amperage is calculated by measuring the voltage across (4) .47ohm power resistors in parrell(.1175 ohms total). The ano bath mixture was changed to what was called out for in the LCD instructions, but in a larger container then before. I dont have fume balls in the tank any longer, no mist suppressant(or very little), and I havnt been using any air agitation like i had before. I dont notice any bubbles forming on the parts, so im guessing the air agitation shouldnt be a large factor now.....unless it creates a more uniform tank temperature that makes a difference. I also measured the PH of a couple things...and the red dye is way off from what Caswell states it should be, which is 5.5-6

*red dye- 8.3
*orange dye- 8.4 (Also, the orange dye was mixed with Caswell's red and yellow dyes, and only used once)
*yellow dye- 7.7
*green dye- 7.7
*distilled water- 6.2
*sealer- 5.5

I used a digital Ph meter purchased through Caswell, and calibrated after purchase. The sealer is rather old...about 4 months. I havnt been using it though....since the problems seems to be occuring before that I just dye and usually end up stripping and trying again before sealing. The ano layer does seem to scratch much easier then some parts I have that were done with the old method. Any ideas? The last part I did was 2.75"X1.5" and 3/8" thick. I calculated 11.4375 sq. in. At 4.5A/sq.ft. puts me at .343A. I got that by measuring .040V across the 4 .47ohm resistors in line with the anode. I let it anodize for 90 minutes. Part passed water break test before ano. Ano tank was 70-72 degrees. Dye was 145-150 degrees. Im not sure what to try out to fix my problem....is it the ano or the dye? If somebody could outline an easy way of monitoring for PAR, I would be very interested in trying to monitor PAR more closely and see if the 90 minutes is the problem. As always, any help is appreciated. Sorry if the post is garbled.....it's 4:30am and im not totally with it

[Fibergeek]

You're right about the dye pH being way off and that will cause dyeing problems, but I think you have a connection problem.

Yeah I know; everyone's tired of hearing Mike and I harp about connections, we're tired of saying it too. You mention no hydrogen bubbles (it's hydrogen not air) forming on the work, and you have no agitation. No hydrogen forming means that no (or very little) anodization is occurring. The most likely suspect is the connections (again). A good test is to start the anodization and look closely at the work until you see some bubbles forming, then turn on the agitation. The bubbles will be visible within a few minutes of applying power. No bubbles means fix your connections. With some experience, you will be able to tell if this is OK with just a glance.

For a given anodic layer thickness, LCD will liberate just as much hydrogen as any other anodization process, but its spread over a longer time, so it looks like there is less hydrogen.

You should be using some form of agitation; it does promote even tank temperatures (always a good thing) but even more important, it will be tough to get an even coating without it.

Its easier to monitor PAR if you're using a constant current source rather than a voltage source, watching PAR is harder to do when you are busy constantly adjusting the voltage. The actual numbers you get are less important than having the right anodization curve; you should see a rapid rise in the voltage at the beginning, then a slow rise to PAR, and finally a slow fall in the voltage, indicating dissolution overtaking anodic growth.

[NeoMoses]

Have you always dyed at those temperatures? I used to dye parts at about 140 but found that when trying to remove the dye with bleach (for multi-colored dye jobs) that it didn't remove all of it. It seemed that the parts were sometimes prematurely sealing when the dye temperature was 140 and above. I've since gone to 'cold' dyeing, where my dye is at 70-110 degrees, but not warmer. The colder dyes will absorb slower, but they seem to reach the same end color as the 140 degree dyes.

[Sid03]

When I said there were no bubles forming on it, I meant that no bubbles were staying on the part.......they rise off the part. I was thinking that they would only have a negative effect if they stayed on the part.

So, do my other dye Ph values seem ok?The two that contained the red dyes(red and orange) have higher values then the other dyes, and they all have higher values then what caswell said teh red should be. Can anybody else measure their dye Ph, and maybe some distilled water also to compare?

As for connection, on this part I drilled a hole smaller then the aluminum wire I use to connect, and twist and push the solid 12 gauge wire into the hole. Its very tight, in fact sometimes you cant pull it back out and it ends up breaking off inside the part(which is no problem since im just doing testing). Im hoping this isnt the problem. If the parts HAVE to be screwed tot he wire....some of the parts I need to anodize have no way of connecting them.

I'll try it this weekend with agitation and i'll try to watch for PAR. How big of a swing in voltage will there be? Does it matter if you watch the voltage at the supply terminals or at the part(just checking)?

Also, I saw somebody made an agitation thing out of PVC on these forums. Does anybody know if the weight of the PVC itself will hold it down being filled with air?

Thanks guys

[Fibergeek]

OK you do have hydrogen forming, that's good.

I don't know much about colored dyes, I only use black. I'll have to defer the dye questions to Mike and some of the membership who work with it allot.

A few electrical observations and questions:

The description of the anodic layer you have does sound like its too thin. Connection problems are the leading cause of this, not removing all of the native oxide from the work will do it too. The water sheeting test doesn't tell you anything about the presence of native oxide.

Good connections can be made without bolting the wire to the part. The idea is to get a liquid tight contact. The entire connection doesn't have to liquid tight, just enough to pass the required current without dropping too much voltage. In your case (12 AWG wire) that would mean at least 5% of the wire surface in REAL TIGHT contact will do. Nicking the wire where it contacts the work, so that there are small raised areas in the wire to be mashed down by the hole will help this. Its a PITA to get right at first, but you'll figure it out and it will stop being a problem. I like bolting where its possible to do because its sure fire. If you can make up some very small all aluminum C clamps (or similar) to make the connection this works well and is quick to attach.

Can you provide some details about your battery charger and dimmer? I've encountered one recently that was too small even for 3A/sq.ft.

Measure voltage and current at the load side of the current sense resistor, as shown in the battery charger /dimmer illustration in the manual. For 4.5A/sq.ft. your peak voltage will be around 11.25 V if your connections are good. Measure the voltage when its driving your setup, you need to have at least this much.

Neilfj and NeoMoses have threads here on the agitation systems they built, these threads should answer your questions.

[Sid03]

Fibergeek- Thanks for the reply. To answer your one question, I am using a 40A car battery charger(2/10/40 Amp range plus 200A jump start setting to be more precise), and I made a box that plugs into the wall and contains an outlet on it which is run through a 600W light dimmer switch. Is that what you are looking for? Thanks again.