Announcement

**Fibergeek** · 12-20-2003, 09:08 PM

If the cathode area is less than the area of the anode (the work) this will contribute to increased resistance.

I don't believe that either Whistule or NeoMoses are using beadblasting to strip off the native oxide, you aren't doing this either, right? How do you know that you got all of it off?

I don't accept the idea that just because a titanium clip is used, the connection must be perfect, and is beyond reproach. This is simplistic and naive. Pro anodizers use much higher voltages and higher current densities than LCD, they can afford to drop the voltage caused by a flimsy clip, titanium or otherwise.

Is this a solution? Yes it is, but it makes any sort of a determination of the end of the process (PAR) impossible. You won't be able to tell the difference between the resistance of the anodize from the resistance of the substandard (for LCD) connection. In your case; if you drop 13V in the connection, you have 17V left to drive the anodizing current. Higher current density should be more tolerant of incomplete native oxide removal. Maybe you should try 6A/sq.ft., or more if you have enough voltage left. You know how to do the math.

My gut feeling is that the problem is mostly native oxide left on the work. Could you describe your method for removing it?

**NeoMoses** · 12-21-2003, 07:39 AM

potsked, I'm not seeing exactly what you are. When looking at my voltage vs. time profile, it resembles that of FiberGeek's graphs in the LCD manual. I see a very rapid rise from nearly 0 volts to somewhere around 13-14 volts, then a fairly constant voltage range around 13-15 volts for the remainder of the anodizing. The initial voltage rise occurs very quickly, in 10 seconds to 1 minute usually. Although I've never graphed it, it's easy to watch the voltmeter needle on my PS when it's in Constant Current (CC) mode.

Are you sure you're using your PS properly? For mine, to use in CC mode, you must max out the coarse and fine voltage adjusters, then set the current. My method has always been this:
- Attach all wires/electrodes with PS off
- Maximize the voltage knobs, minimize the current knobs.
- Turn on the PS. If you followed the previous step, you should see max voltage readings, but 0 amps.
- Adjust the amperage to the desired setpoint (for example, 4.5 A for 1 ft^2)
** As you are adjusting the amperage, you will see the voltage start to fluctuate. Mine usually drops significantly, then raises quickly.

I am not fully satisfied that I've cured all of my consistency problems. I've had a couple of batches turn out well, but I need to get about 10 in a row good before I will say I've completely solved the problem(s). All of my parts are (currently) prepped as follows:

1. Sand/steel wool to desired gloss finish.
2. Polish with mother's mag polish
3. Wash with hot tap water and dish detergent using a toothbrush.
4. Wash with simple green/toothbrush
5. Make sure part passes water bead/sheeting test.
6. Attach electrode
7. Re-wash with simple green
8. Thoroughly rinse.
Part is now ready for anodizing.

As you can see, surgical cleanliness is implied. I believe most of my problems were from oils/greases left over on the surface, but not the native oxide. Why don't I think it's native oxide? Because I successfully anodized certain parts that had been left out in the air for months without any special preparation to strip off the native oxide. Although this may be a problem for some, I don't think it was mine.

I'm taking a break from this for the holiday season, so I hope this helps you as I won't be around much after this.

Happy Holidays,
NeoMoses

**Fibergeek** · 12-21-2003, 10:17 AM

NeoMoses,

I can use some feedback in the form of LCD data. You don't have to plot anodization curves, but could you post the current density you used?

Your setup starts at near zero volts, like its supposed to. But the rise to 13V (like Potsked) implies additional resistance or a currrent density of around 5 A/sq.ft.

Starting at near zero volts is the important thing. You may be right about Potsked not setting up his power supply correctly for CC mode.

Question: You aren't using a desmut/deox dip before anodizing, this would probably improve you surface cleaning. This stuff is very effective in removing the native oxide, provided all grease or wax is removed first. Have you tried this?

**potsked** · 12-21-2003, 11:04 AM

First off, to prep I:

strip w/ caswell's stripper
desmut w/ the desmut/de-ox
beadblast
de-ox with the desmut/de-ox
anodize

i recently tested the connection as a problem by anodizing a part halfway out of the electrolyte and the same thing happened. started near zero but rose quickly to 13.8 volts. the voltage rating of 13.8 has been re-occuring in part batches of wildly different sizes, which lead me to believe that this resistance problem is independant of the sq. inchage of the parts being anodized.

actually neomoses what you described is 100% exactly what im experiencing, and perhaps it has something to do with our rectifiers. by your description i think we have the same rectifier, would this be yours perhaps? if not the same model, is it the same manufacturer?

http://cgi.ebay.com/ws/eBayISAPI.dll?MfcISAPICommand=ViewItem&item=2582162456

i am operating my PS in the exact same manner as you are, os i dont think that's the problem. since the part i just experimented on was hardly larger than the area of my cathode, i think we can rule that out as well lets make a checklist of what we know it isn't and then figure out whats left

native oxide - doubt it
Connection to part - nope
cathode size - no
electrolyte composition - doubt it
arcing the connection through misting - doubt it, never heard bacon sizzling
part preparation - dont think so, is there any step im missing?

the exact duplication of this problem in anothers setup rules out many accidental things, like improper bath make-up or too high/low temp ranges

im led to believe at this point that it may be indirectly linked to the lab power supply. has anyone had trouble with theirs?

thanks

*edit*

Neo-moses i saw a post where you linked to the PS you bought, it appears to be very similar to mine. i could expect a problem like this from one manufacturer, but since the come from different manufacturers i can only assume that the PS isn't at the root of the problem. in addition i was having this problem with caswell's rectifier.

im going to try and make some runs with the extra resistance and see if i can still have the parts come out decent

**Fibergeek** · 12-21-2003, 06:31 PM

Potsked,

Your prep looks fine.

The consistent 13.8V thing looks very odd. If it was doing this with the Caswell rectifier too, the power supply could be ruled out. These are vastly different designs.

This does look like another current path; one that wasn't intended, that's why its called a sneak path. You said you didn't hear crackling, so it's not shunting that much current. Try to suspend the anode connection, so that the anode and it's support is nowhere near the cathode or its wiring. Hang it from the ceiling temporarily. Electrolyte mist can cause a sneak path if the area of the tank between anode/cathode is even damp with electrolyte.

**potsked** · 12-21-2003, 09:34 PM

i did a test with one part, one run with my usual fabricated tank bar and another with just the anode wire being attached to the cables. both had similar resitances. as if it could get any worse i am now getting more inconsistent voltages.

one run had 18 sq." (at 4.5 a/sq.') and was drawing 7.1 volts
the run im doing now had 90 sq" (at 4.5 a/sq') and is drawing 10 volts.

it usually start at like 1 or 2ish but rapidly ascends to the aforementioned numbers.

i grow weary of reading posts of people that hooked up battery chargers with dimmer switches and have had no problems, while im sitting here with my $2000 setup with my thumb up my butt cause of a resitance problem.

i realize that you're nearing the end of possible causes and i simply dont understand how an ohm of resitance can be working its way in there.

i was checking some connections, and i found there to be 369 ohms of resistance from the anode and cathodes because of their connection to the power supply, is this right? or should there be none at all? i also found there to be 1.6 or 1.7 volts when theres no parts in the bath and no amps applied, does this mean anything? i know im just stabbing in the dark, but its all i got nowadays.

i know i asked about cathodes beofre, you mentioned size, but i was wondering about condition too. from ron newman's site i had been advised that leaving the cathodes in the electrolyte doesnt make a big difference, could this have been causing the extra ohms? im cleaning and desmutting them now. recently ron told me has has switched to 6000 series Al cathodes, whats the advantage of these?

how might one check the gravity of the electrolyte? is it possible for the electrolyte to affect the resistance that much?

if one was to methodically cheack every single possible type of connection that would be associated with the curcuit's resitance what might be on that checklist?

thx

**potsked** · 12-22-2003, 03:02 PM

as an addition to my last post;

some have menitoned the fact that temperature can affect the resistance greatly, i have some heaters in the bottom of my ano tank to get it to the right temp, is it posisble uncirculated electrolyte that has pockets of heat or areas of cold could affect the resistance in the way i've described?

also, neo moses, i would like to collaborate with you on solving this problem we share, can i get your email or something?

thx

**Fibergeek** · 12-22-2003, 10:04 PM

It's time to consider the improbable.

I believe you electrical connections are fine; something else is causing this 13.8V business, Im sure you tested your connections thoroughly.

So what's next? A possibility:

What if your power supply will not go below 13.8 volts (or so) in CC mode, only CV mode. Power supplies made these days operate by series regulation; the variable element (transistor or MOSFET) is in series with the load. In shunt regulation (like my VCCS) the variable element is in parallel with the load. Shunt regulation will go all the way to zero volts and amps but it dissipates more power than series regulation; which means it costs more, hence is less popular. This doesn't apply to premium power supplies; Agilent (HP), Tektronix, Lambda, and many others. These have the electronics to force tight regulation in both modes, which also costs money. I know the data sheet for your Chinese PS says it does CC mode down to zero, but does it really?

This can be tested by putting a power resistor across the power supply (disconnected from the anodizing setup) put the PS in CC mode and see if it will go down to zero volts. The resistor should be about 10 ohms 10W or more watts. Increase the current, don't go over 1 Amp unless the resistor is higher wattage. If CC mode is working correctly, you should be able to vary the current between zero and 1 amp, and the voltage goes from zero to 10 volts. You can get power resistors from your local Radio Shack.

If your PS won't go to zero volts in CC mode, you can fix it by putting an appropriate power resistor in series with you anodizing setup, ideally dropping the voltage to zero or close to it at startup. This should not degrade CC operation if you have enough voltage available (and you do).

There could also be an isolation issue which if fooling the circuitry in your PS, but try the above first.

It is normal to measure a few hundred ohms across the PS outputs when it is turned off.

**NeoMoses** · 12-23-2003, 02:52 AM

Originally posted by Fibergeek

NeoMoses,

I can use some feedback in the form of LCD data. You don't have to plot anodization curves, but could you post the current density you used?

Your setup starts at near zero volts, like its supposed to. But the rise to 13V (like Potsked) implies additional resistance or a currrent density of around 5 A/sq.ft.

Starting at near zero volts is the important thing. You may be right about Potsked not setting up his power supply correctly for CC mode.

Question: You aren't using a desmut/deox dip before anodizing, this would probably improve you surface cleaning. This stuff is very effective in removing the native oxide, provided all grease or wax is removed first. Have you tried this?

1. Additional Resistance/Current Density? Except for a few instances, I am aiming for 4.5 A/ft^2 current density. It is very possible that I am slightly higher than that, due mainly to the fact that I tend to over-estimate surface areas if they are time consuming to calculate. Could this over-estimation lead me to 5 A/ft^2? Quite possibly, since some of my parts are quite small.

Could there be additional resistance? Yes. I have multiple junctions between the power supply and the anodized part. (Power supply to wire to mounting bar to alligator clips....) There could be a stackup of these resistances in series.

A question that has popped up recently: I routinely anodize more than one part. For example, I have a bar that stretches across my tank with alligator clips attached to it. I can easily attach the wires from each part to these clips. The electrical circuit would see these as resistances in parallel, correct? So, if the contact resistances were different from part to part, it is very possible that each part is seeing a different current, thus a different current density. Is this a real issue? I don't know at the moment, but when I get back home I plan to do some measuring to try to quantify this.

potsked, if you would also like to communicate outside of this forum, the best ways to reach me are:
email: bpryor @ umr.edu (remove spaces - I hate spambots)
AIM: bryanpryor
Yahoo Messenger: bkpryor

Like I said, I'll be away from the 'net for about another week due to holiday travels, so if I don't respond in a timely manner, just give it some time.

Happy holidays to all!

Announcement

Extra resistance??????

Extra resistance??????

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment