PSU calculations

[kickn]

Okay - can someone help me doublecheck my numbers since I don't think they are quite right.

I've got to anodise 4 parts at once - the total surface area between the three of them is approx 4.56 sq ft.

I'm working it out as a 20A supply - does that seem right

[Fibergeek]

Im not surprised that you are experiencing problems. Almost everything that you are doing is wrong. To divide up the work; I'll take the electrical issues, there are other members here who can address the prep issues, anodization details, dyeing & sealing. That way no one gets stuck with all of the typing. Don't worry, we'll get you on track, we've done it for newbies many times.

First off; you should reconsider doing the 3 or 4 pieces (which is it?) together; it will be a lot cheaper and easier, I'll show you why.

Using 4.56 sq.ft. as the example, we can calculate the required power supply current (Amps) and the power supply peak voltage required. I am also assuming the LCD anodizing process, which is ideal for learning anodizing, not to mention the the lowest cost and safest process.

LCD operates over a current density range of 3ASF (Amps per Square Foot) to 6ASF, with 4.5ASF as the median value.

The numbers are in this order, ASF, anodization time, power supply current(A), peak power supply voltage(V), and peak power dissipation in the tank in Watts(W).

For a surface area of 4.56 sq.ft.:

3ASF, 120 minutes, 13.7A, 8.8V, 121W
4.5ASF, 90 minutes, 20.8A, 11.4V, 269W
6ASF, 60 minutes, 27.4A, 13.2V, 362W

We can draw some observations from these numbers, and apply them to your example.

1st. you can forget your "20A" battery charger. If you actually measure what the charger is putting out (using the multimeter in your tool box) you will find out that it isn't 12V at 20A, its more like 10.7V at 10A. All battery chargers are rated for driving a partially discharged battery, which is about 6V. An anodizing or plating circuit is a grounded load. Notice that I said circuit, is is a circuit, and all of the electrical math and physics applies exactly.

2nd. Look at the power dissipation; these values in Watts show how much heat will be dumped in the tank, like there was an electric heater of that Wattage in it. You'll have some nasty problems with dissolution if you allow the electrolyte temperature to get warmer than 75 deg. F.. You will need some kind of cooling system.

If I assume that your 3 or 4 pieces are all the same size, you can reduce the required current and the power dissipation by 3 or 4 if you anodize them individually. If they aren't the same size, the reduction for each piece will be proportional to its surface are. This may allow your battery charger. However, you must have a way to measure the current and control it by varying the voltage. The lamp dimmer controlling the battery charger will work, but its a PITA.
If you don't control the current, it will be way too high when you start anodizing, this usually immediately damages the connections and you fail.

[kickn]

Thanks for this - It does clear some things up a bit. I'll go and re-check things.

The single biggest part is 1.98 Sq Ft so I would still have a problem with power as I need 15A @ 6ASF roughly. I do have a line on a 50A psu which if I can make a VCCS circuit for would be perfect. Any help on that would be greatly appreciate!

I'll go back to the manual as well and see.

Would you suggest I try another test with a meter on right from the outset and see if it spikes and then drops right at the beginning? If the connection was killed - wouldn't the bubbles aroudn the part stop?

and to you and everyone else I REALLY appreciate your help - Thanks!

[Fibergeek]

1.98 sq.ft. at 6ASF will require a peak voltage of 13.2V, an applied current of 11.9A, and will produce 157W of heat (peak value). If you used about 5 gallons of electrolyte, a small window fan blowing across the tank will probably keep you out of trouble with cooling.

Don't attempt another anodization until the membership here straightens out all of your problems, we are not finished. Let them get home from work and read your posts.

If you aren't fusion welding the anode connection, its likely that you will drop some voltage there so having 15V available would be shrewd.

If you aren't experienced with building power circuits and have the test equipment, you will be much better off hunting Ebay (or whatever) for a used Constant Current / Constant Voltage (CC/CV) bench power supply. You'll need one that's at least 180W, 0 to 15V, and 0 to 12A. Avoid those that are over 20 years old. Caswell sells a new 20A 30V 600W unit and ships internationally, if you want the easiest route.

The bubbles are hydrogen being produced by the anodization reaction, you'll need some kind of agitation to knock them off of the work, bad things will happen if you don't do this.

[kickn]

Thanks fibergeek!

This is what I was thinking of getting just to see if I could get it to work - what do you think?

http://cgi.ebay.co.uk/ws/eBayISAPI.d...e=STRK:MEWA:IT

I know its a bit overkill but if I was going to do a lot of this then I could do all 4 parts at the same time (as you say with a cooling solution) and save a lot of time. Also - could I push the time window a bit if I had amps to spare?

I was going to get some airstones for agitation as I already have an extraction fan set up so I'm not as worried about the fumes.

Just in regards to the anode connection - if the bubbles are there - would that not mean the connection is okay or is it just a 'partial' connection.

My test parts are small bits (approx 3in) of 1/8 x 1 1/2 bar.

Thanks again!

[Fibergeek]

That power supply would not be a good choice. You want something that will operate in constant current mode (CC) this will make the process way easier and produce better results. The power supply you choose needs meters to indicate both the voltage and the current, you will need to know both. Go and take a look at Caswell's 3 Amp and 20 Amp constant current rectifiers, you want something like that.

Yes, increasing the current density reduces the anodizing time, provided that you can hack the cooling. All of this is easily calculated, you don't have to guess at it.

Aeration (air bubbles) works, but has the disadvantage of producing acid mist. This will rust anything iron or steel that it can reach. You will find agitation method threads here, which illustrate a better way. For now, just stirring the electrolyte for a minute or two every 5 minutes will work OK. Not so vigorously that you splash electrolyte.

You need to be careful with what you read about anodizing on the Internet. I've seen most of it, some of it is downright hilarious its so wrong. If you haven't surmised this by now, this is one of only two sites (that I know of) where you will get tested and verified facts, not hearsay and old wive's tales. That's because some of us here post there too. I am including the anodizing "pro" sites in this statement. These people don't understand their own process, its needless to say that they don't like us. Albert Einstein said "Everything should be as simple as possible, but not simpler". That has what has been going on in anodizing for decades.

The only "pro" that I know personally who understands the math and physics of anodizing is a PhD. electrochemist and materials scientist, and in much more detail than I do. She has helped me develop the formulas I use here and has verified them for me. The other members who will respond here won't be as mathematical; its not their style. But rest assured that they know what they are doing, we all know each other.

[kickn]

What if I just use a couple meters I have here (or look out for some analog ones? I understand that a CC psu is a better option - but they are also £60 or £70 quid over here. Ordering stuff from the states isn't really an option for cost.

Do you think that would be okay?

I'm just re-reading the caswell manual and hope I can increase my learning curve - and reverse some of the bad knowledge I've gained

Thanks for your help anyways though

Would a couple meters on that PSU work?

Edit: don't get me wrong - i'm not set on this PSU - it just seems like such a good deal for what it is (possibly because of its own limitations) If its a definate no please say so - if it needs a few extras to make it work - i'm fine with that

[Fibergeek]

You can use separate meters if you want, digital or analog makes no difference. That PS isn't capable of CC operation. If you get into anodizing for more than just occasional use, you'll get a CC/CV supply anyway. Why buy a power supply twice? Look around, CC/CV power supplies are pretty common in the UK, and everywhere else. These are common general purpose test equipment in any and all electronics labs. Neilfj got one off Ebay a while back, it would sail through your job, he paid around 20 bucks for it.

[kickn]

how about this then?

http://cgi.ebay.co.uk/ws/eBayISAPI.d...e=STRK:MEWA:IT

[Fibergeek]

That one is not CC/CV either. You can tell if its CC/CV if it has two meters (current and voltage) and two adjustment knobs, one for current and a separate one for voltage. Some CC/CV will have two knobs for each function.

I searched Ebay UK. Item # 3836936467 is a 2.5A 25V CC/CV supply, 6.5 Pounds Sterling. I'd prefer a larger one (more current) but it would do for this size work.

Item #3836743025 is a 3A 30V CC/CV unit. 16 Pounds Sterling, and I think its new. Again, a larger one would be better.

[kickn]

Okay - thanks for that - sorry if this is going to frustrate you but what do you think of these:

I know the voltage isn't variable but I thought it was the amps which impact how quickly the layer is created and volts is the possible depth of the layer?
http://cgi.ebay.co.uk/ws/eBayISAPI.d...tem=3836386869

a bit overkill
http://cgi.ebay.co.uk/ws/eBayISAPI.d...tem=3836557270

What I want to avoid if possible is getting say a 2.5A psu that takes 4 hours to anodise the part with. I don't have 16hours per unit to spend

I completely appreciate what your saying though about time and money. Unfortunately I have more time then money right now LOL

[Fibergeek]

Neither of those is suitable; the first one is an AC/DC type, the wrong kind for this. the second won't go over 8V, which is too low.

Wait a few days and look again. Or look at the for sale adds in your local newspaper. You'll find what you like, be patient.

[kickn]

Thanks fibregeek - I think my confusion is coming from not fully understanding the requirements. Is 12v the absolute target? or is there a range of say 10v to 14v?

The reason I put the one with only a current control in there was because I thought that it was the current you adjusted and monitored?

So what does varing the voltage impact? I think I understand how the current is working (in a very basic sense anyways).

If I did want to anodise all 4 parts at once (for time) then I would need a 14v 30a psu right? I know there are cooling issues but I would of course deal with that.

Thanks again

EDIT:

Okay - i'm working out the forumlas based on LCD right now to get a good handle on things so please ignore the rather stupid statements above.

Can you tell me how though you work out the potential voltage required? I can see how to work out the amps but as I see it PAR will increase and then decrease - is there a way to formulate the potential maximum? or is this just based on tests?

Just on the psu front how about this one:

http://cgi.ebay.co.uk/ws/eBayISAPI.d...e=STRK:MEWA:IT

I'm still looking around

Thanks

J

[Fibergeek]

A power supply like that one is what you want.

We determined at the start of this thread that you needed a minimum of 27.4A to anodize 4.56 sq.ft. at 6ASF. This corresponds to a maximum Voltage of 13.2V. We also agreed that you should have some margin, so we bumped up the maximum Voltage to about 15V. This will require 27.4A x 15V = 411W. This is going to get expensive.

We then agreed that a reasonable compromise would be to do the anodizations 1 piece at a time. Since the largest piece is 1.98 sq.ft. the power requirements are now 11.9A and 15V, which is 11.9A x 15V = 179W. This will cost less.

If you had a 10A power supply, you would need to reduce the current density to 6 ASF x 10A / 11.9A = 5 ASF. 5 ASF is within the LCD range so this will be OK.

When all of this makes sense to you; and you understand how to adjust the process without getting into trouble, you can optimize it to suit your particular application. At this point, you can make the necessary process adjustments to speed it up considerably, and you will know how. Or maybe you will decide that that it will be too expensive and you will live with the anodization time. Either way you will be successful.

[kickn]

Well - i just won the auction - a bit more then I would have wanted to spend but if it gets things going then hey!

My gameplan is to get this going with single pcs in a reasonable timescale and then move from there into 4 pcs (by wich I should have gotten around to buying/building a bigger supply)

I have a good idea now of how things are working - thanks for all your help

One question I do have though is how to calculate the potential max voltage. I was reading the manual but I think I was missing something.

By my numbers this should take about 70mins @ ASF 5 right?

Am I also correct in my understanding that the electrolite solution is 5% solphuric by volume?