Another anodizing question- Stan et al.

[BrendanP]

I'm going to anodize and dye some aluminium picaxe project housings. Anodizing (dependent on the surface area to be anodized) requires a 12V high I power supply. I'll need around 200 amps. High I 12V power supplies are big bucks and I can't afford to buy one as I'm going to lash out on a TIG. (I'd build that too if it was feasible).

As indicated in another thread I came across this article by a Ham and have decided to use this method to mod a microwave transformer to get the current I need.

http://www.users.on.net/~endsodds/psrewind.htm

Supposing I can draw 20 amps from each transformer it occurred to me I could make a sort of variable I supply by having, say, 10 transformers in parallel.

Each transformer is individually switched to the ~240 supply. The 12V output of each is rectified individually and connected to a common output. As I need more current ( because I'm anodizing larger objects) I simply switch on more transformers.

Is there any reason this approach wouldn't work?

I'm off to the tip tomorrow to scrounge dead microwave ovens and thought I better check in before I debased myself (even further).

(Hand wringers need not reply to this thread. ie. Ooohhhhh.... you might get fried by the nasty ~240.)

 

[Mad Professor]

Did you say 12v @ 200amps.

When I last looked into home Anodizing I am sure the kit I looked at used a 0-12v, 0-5amp digital power supply.

From memory I think it was somthing like 12volts @ 500ma per square inch to be anodized.

I have no idea about Dyeing so can't say anything on that.

 

[BrendanP]

I'm sure they do for small scale, two bit, operations.

I need to do larger objects. To achieve mil spec type two class 1 anodizing finish it requires 12 amps per square foot of surface area.

Here, read it for yourself.
http://www.focuser.com/atm/anodize/anodize.html

Whilst I appreciate all replies and input if you read my post you will see that my question was concerning the operation of multiple AC mains transformers in parallel rather than seeking advice on the anodizing per se.

 

[Minifig666]

A few car battries should easly be able to do that, I saw one a a garage promising 200A output. It will save on the mot modding (Which is likley to be dificult) but battries may be a bit hard to scrounge. Mabie if you find a computer or two an ATX might do the job and there is lots of parts to be found in a computer.
If you got the MOT way be careful with the 240V~

 

[Minifig666]

Oops double post, can this be deleted please?

 

[BrendanP]

Did you read my post before you replied? Do you know anything about AC transformers used in parallel? If not why reply?

 

[Minifig666]

Yes, I know about transformers, but I don't get why the angry reply? I was just trying to come up with some equivlants that may be more suitable for your application. Also, if you are asking for help with transformers why question my knowage?

 

[Rickharris]

High current transformers in parallel to increase the current Mmmm not a good idea I think.

If you have such a large scale need get it done professionally there are people out there doing this for a living.

If a one off you many not pay much to get in their next batch.

Personally I don't think I could advise anyone to mix mains, high current and dilute Sulfuric acid, caustic soda, Phosphoric or Nitric acid (just to show I did read it).

Where do the picaxes go? (please don't shout at me just because this isn't the answer you want (-: )

http://astro.neutral.org/anodise.shtml This guy looks like he is being successful.

 

[manuka]

Brendan: Your quest is out of my league,so good luck (& -ahem- take care...)

 

[MartinM57]

I can't see why 20 absolutely identical transformers wired in parallel won't work - but I wouldn't put twenty random transformers in parallel and you would have to take care that each mod was as near as dammit identical.

I also wouldn't (initially) switch one/more on/off whilst others are on.

...and you do seem a bit spikey about this BP - take a chill pill and we'll all get on better...

EDIT: I am, of course, not qualified to give this advice...(about the transformers or the pills )

 

[Dippy]

I'd go along with Martin's advice.

Surely, the best answer will be given by doing a bit of searching for proper, experienced advice.
I find Yahoo/Advanced search quicker than Google.
http://uk.search.yahoo.com/web/advanced?ei=UTF-8&p="transformers+in+parallel"&n=10&va_vt=any&vo_vt=any&ve_vt=any&vp_vt=any&vd=all&vst=0&vf=all&vm=p&fl=0&fr=altavista

I would assume , however, that if you have rectified each transformer o/p then you could handle slight mismatches.
Certainly with many types of isolated bench PSUs you can parallel them OK.
Maybe there is a parallel there haha.

Start with 2 and work up...

 

[Bchip]

Brendan

If I read your original post, you are considering connecting the DC ouputs of (bridge ?) rectifiers in parallel, rather than directly connecting the transformer secondaries?

I think this is OK.

Years ago I did the same thing to get 100VDC @ 15A to drive a Milling Machine motor. The primaries were connected to a 3kW variac for speed control.

This used 2 240V/80V rms @ 10A transformers ( same type ) in parallel, with IIRC, 2 3506 rectifiers.
Yes, a bit over 100V peak, but WTH

Worked for several years OK. So, I presume the arrangement is acceptable. No smoke, flames or anything else exciting.

Not anodising ... sorry.

Dave

 

[BrendanP]

Thanks guys.

I like and respect this forum because I know I can get solid advice from people with a can do attitude. When those people don't know they do what Stan did and just say so.

Yep, acid, HV AC and home brew power supplies can be/are dangerous, thats why I'd wear proper PPE when handling the liquids and wouldn't touch the live DC conductors nor let anyone else operate the system.

I did as Dippy suggested and did a bit of looking around and it appears that transformers can be run in parallel. There are issues concerning the paralleling of mismatched transformers.

Experimentation and innovation is inherently dangerous, personally, I try and reduce the risks to an acceptable level and then get on with it. A life lived without intellectual stimulation and challenge is not worth living in my view.

 

[BeanieBots]

It is perfectly OK to parallel transformers and/or rectifier bridges as long as the matching caveat is met. This often done for failsafe reasons as well as current increase reasons.

What bothers me about this particular application is the intention to switch them in/out. Why?
Say you have (just for the arguement) a 1A transformer.
Parallel ten of them and you have 10A ability.

Put on a 10A load and all is well.
Switch out two, you now have 8A but the load is still 10A.
Result = lots of smoke.

Switching out any parallel supplies simply reduces the supply's ABILITY to supply current, the LOAD determines how much will be drawn.

Overlaod a transformer and it smokes.

If you want the vary the current delivered to a load, you need to change the drive voltage.

Use a variac to supply ALL the transformers which should be PERMANENTLY wired together.
(and don't forget a fuse for EACH transformer)

Mechanical analogy = suspension bridge.
The number of suspension wires (transformers) is determined by the MAXIMUM load EVER likely to be seen.
Would you add/remove suspension wires for normal/rush hour time depending on how many cars are going over it?

 

[BrendanP]

Thanks BB.

I anticipate having to anodize objects of different size or varying multiples of the same object. (This isn't going to be a one off project, I intend to churn these things out.)

The standard is 12 amps at 15V per square foot of aluminium surface area to achieve the required finish.

I thought by having multiple transformers in parallel I could vary the current flow by switching transformers on or off.

I had presumed that the inherent resistance of the copper wire in the primary would limit the current flow there and prevent the transformer from self destructing.

In my mind I had thought the primary winding would be like the coil in a solenoid that doesn't melt or fuse even when you put the solenoid across a large battery like a car battery.

Obviously my theoretical understanding of how a transformer works is wrong. I will do as you suggest re. the variac with one transformer to get an understanding of what is going on and then build on that.

I had planned to fan cool the bank of transformers. I pulled my first transformer out of an old microwave oven today.

Anyone who does this should beware of the large capacitor in microwave ovens. It can hold a very high voltage for a long period of time and can kill you. You should short it out with, say, a screw driver with an insulated handle before pulling the oven apart.

 

[BeanieBots]

Your thinking is only very partially correct.
A transformer will have some resistance due to the copper but it will be very low. It is designed such that the voltage will only slightly droop under full load. Full load is the maximum power it can transfer from primary to secondary without having a "disaster".

So, if you load up a transformer to its maximum rating and then add in another one in parallel, you will see very little change in its output voltage.
This becomes more true to ideal the larger the transformer.

 

[flyingnunrt]

Just guessing, but possibly another way to get your 12V at 100 A may be to use a car battery and an altenator driven by an electric motor
OR
use an old stick welder some large altenator diodes on a big heatsink plus a big capacitor and a few mosfets parralled up.

Use a picaxe to keep the current regulated etc

 

[moxhamj]

I'm not able to comment on the specific application, but with respect to transformers, yes you can parallel them if they are both matched in design, wire diameter, number of turns, magnetics and also matched in polarity. Take two 12V secondaries and put them in series. You will measure 24VAC (if you measure 0V then one of them is out of phase with the other). Now mark the wires (eg with electrical tape) as plus and minus. If this was DC you would have plus at the top, then minus joined to plus then minus at the bottom, like two batteries in series. Of course, this is AC and there is no plus or minus, but this is a way of working out the phase.

If you join plus to plus and minus to minus then that is fine. Double the output amps. But if you join plus to minus and minus to plus everything will blow up due to huge currents going round and round. So - first thing is to put fuses in every secondary wire, just in case this ever happens.

If you use toroid transformers eg for audio amps you can see which way round the wires go so it is easy to get the polarity right.

AC can do other cool things eg in series, you can have a 12V in series with a 12V and put a 3V in series with the opposite polarity and you get 21V. 12+12-3.

Of course, as has been said before, transformers need to be matched to run them in parallel. So when you buy a big toroid for an audio amp, the number of turns on the two secondaries is exactly the same. Plus the polarity is marked, and the ones I've bought even have instructions clearly showing how to connect in series and in parallel. I'm not so sure this would work for pulling transformers out of old equipment.

I wonder if pwm a mosfet at, say, 100hz and run it off a car battery? Then you can get a picaxe into the circuit too...

 

[vshortt]

This isn't really a solve for your issue, but have you tried caswellpalting.com?

They have a number of small plating and anodizing kits you can purchase, though they're not necessarily the cheapest, I can say to DO work.

I have their chrome plating kit and it works great. I've used to several times on a Jeep CJ restoration, chroming everything from hinges and screws to a small custom made muffler for an RC boat. (using copper pipe pieces) and have been VERY happy with the results.

I woul imagine their anodizing kits would be the same quality and probably much safer than coming up with a home-brew high amperage AC power supply.

Anodizing is no fun when you're dead, ya know? There are some things better purchased from a company.

 

[BCJKiwi]

Perhaps that should be
http://www.caswellplating.com/

 

[vshortt]

Perhaps that should be
http://www.caswellplating.com/

Exactly. Sorry, early morning typing got the best of me. Thanks for the fix!

 

[BrendanP]

Hi Dr. I was just thinking of you this evening and wondering what you had been up to.

"....There are some things better purchased from a company...." Sure, like nuclear reactors. I leave that to GE.

Otherwise I get in there and have a crack, if I get fried so be it. Once you get past making LEDs flash and so on you tend to look for bigger technical challenges.

 

[Minifig666]

"If I get fried so be it"
I feel that is a very bad way to look at electricty. It can and (God forbid) will, given the chance, kill you. I always ware rubber gloves and try to discharge caps when working wth the mains. On the rare occasion, when working with live stuff (Don't try this at home kids (kids? I'm only 14, but have been trained)) there is always an RCD in the curcit. Sorry to rant but sometimes remarks you probably never thought about get me all ranty. I'm not trying to have a go at you, just make a point.

 

[slimplynth]

(edited/deleted)

 

[BrendanP]

Thanks Slimp for your detailed reply. I apologized for not replying sooner, I only just realized that you had written.

Most of my information on the topic has been gained from this guy:

http://www.focuser.com/atm/anodize/anodize.html

I have phoned him in the US and gotten verbal advice and also bought his book. Lest anyone think he is a cretin ,try hard or wannabe he is a graduate electronic engineer.

The process is neither difficult or dangerous. I'm 44 years old work as a rigger/crane operator in civil engineering/construction and do things with a higher risk factor than anodizing as a matter of course on a virtual daily basis. I have a a BA degree in humanities. I know my limitations. relax guys I'm not about to disappear off the forum "k.i.a".

Yes the parts are largish, 150m diameter cylinders and there will be multiples.

Ok, I just bought a NOS 500watt computer power supply on ebay for 30 bucks. It has two 12V outputs, one rated at 23 amps and one at 25 amps. My online research indicates that you can join the two outputs together together to make a nominal 48 amps output.

Can I then join multiple supplies together, say 5, to give me the 200 amps?

For current limiting I am thinking of using 12V loads in series (car headlights maybe ) on the low side of the parts being anodized. I would have a I meter in the circuit. I start with all the head lights on. I then start bypassing the headlights until the current flow through the anodizing bath is at the level I want (but below the rated capacity of the power supply).

Ideas/input anyone?

 

[moxhamj]

You can test whether current is being shared equally by paralleled supplies by putting an Ammeter in each power supply lead. Apply a load and if the amps are roughly equal, then you are in business. Try it with two.

Car headlamps sounds an interesting idea. They are roughly 60W (12V, 5A). I'm not 100% sure of the circuit you are using. Do you have the headlamp in series with the anodising bath then you short the lamp? Or do you have the lamp in parallel to the bath and then turn the andoising bath on?

The anodising bath ought to have a resistance you can work out. In fact, it would help a lot if you knew that number. Ok, maybe it is 0.05 ohms but you can still work out V=IR and 12/0.05 = 240 amps. 0.05 ohms won't work on a multimeter as the resistance of the leads will be too high and will upset the reading, but if you put a bath in series with a car headlamp, and then measured the amps (which will be roughly 5 amps), and also measured the milivolts/volts across the bath, you can work out the resistance. Do you have access to the anodising bath?

 

[BrendanP]

Hi Dr. I had thought to have the lamps in series after the bath.

Then as you say I short out or by pass lamps as needed to get the amperage to the required level to give me 12 amps per square foot of surface area to treated.

It worked ok on the circuit testing software I have.........

I can make up a acid solution to test resistance. I have concentrated sulfuric acid here.

Obviously the greater the acid concentration the greater the conductivity of the solution and hence higher current flow. Higher acid concentrations affect the quality of the anodizing finish however.

Specs for the power supply are below.

This Power unit is design for ATX and Micro-ATX Case.
Fully support all Intel & AMD series demand

High Quality 24pin Premium Solid Power Supply
EMI/RFI:FCC part 15 Class B,EN55022 Class B
• Input voltage range:90-135/180-264Vac(Switch select)
• With 12V/4.0A for LCD monitor
• Super high air flow -- 1 Cooling Fans
• Low Acoustic noise
• MTBF＞50,000 hours at 25℃
• Complies with ATX12V
• Surge:IEC61000-4-5
• ESD:IEC61000-4-2
• Dimension:155x86x150(mm)
Specification Input
Voltage 230Vac(180-264)
Frequency 47Hz-63Hz
Current 4.0A(MAX.) At 230Vac
Inrush Current ＜70A COLD START AT 25℃
Efficiency 70%MIN.at Full LOAD 230Vac/50HZ

Output:
Voltage +3.3V +5V +12V1 +12V2 -12V +5Vsb
Max Load 28A 32A 23A 25A 0.8A 3A
Min Load 0.5A 0.3A 1.0A 1.0A 0A 0A 0A
Regulation ±5% ±5% ±5% ±5% ±10% ±10% ±5%
Ripple 50mV 50mV 120mV 120mV 100mV 120mV 50mV
Noise&Ripple 100mV 100mV 200mV 200mV 200mV 200mv 100mv
Power Good Signal 100-500mS
Power Fail Signal 1mS MIN
Hold-up Time 16mS MIN
Line Regulation 1% MAX

Protection
Over Load +5V:26A-36A /32-40A/ 38-45A
Over Voltage +5V:6.8V,+3.3V:4.5V,+12V:15.6V
Short Circuit All output to GND
Hi-Pot Voltage 1500Vac, Time 2-3 Sec. cut off current 10mA MAX.
Chassis color: Dark Grey

 

[slimplynth]

I was trying to edit my post before an messed up, ended up deleting the lot - apologies for that. Don't think it was too important but sorry none the less.

I was just wondering about cooling in your tank design

Temperature is considered critical for our line of work; employing oil heat exchangers and water cooling pipes to maintain a steady temp. Given the size and batch quantities this could be an issue for you too.

(re. original post of this thread - sounds like there's plenty of scope to be adding a picaxe or two... Data logging/flow control/...).
You'll have to take pix of the final product for us to see when you're done.

Best Regards

Lee

 

[BrendanP]

You're right Lee, cooling of the acid solution is important.

I'll probably initially just use a coil of ally pipe laying in the bottom of the tank and run cold tap water through it.

Another option is to run the water in the coil (with a pump) through a car radiator mounted in a frame with a thermo fan to push air thorough it.

 

[moxhamj]

Is there a reason it has to be 12V? Just looking at your power supply specs, I wonder what would happen if you used the 3V. Should be enough for electroplating?? Anyway, you have a few options there, 3V, 5V 12V.

Given a constant acid concentration, and constant temperature, and a constant distance between electrodes, I would suspect that resistance is proportional to surface area being electroplated. If so, you ought to be able to test with x area of electroplating and one power supply. If you use a 10A circuit breaker then the worst that can happen is it trips. If the amps look ok and within the range of the supply then you can grow it - eg if area x doubles, then put two supplies in parallel. Some simple experiments ought to confirm ballpark figures for amps vs area.

 

[BrendanP]

Not sure about electroplating Dr. but in this instance I want to have a crack at anodizing aluminium. It appears that you do need the higher voltages around 12-15V to get the correct effect.

 

[slimplynth]

Re. Cooling pipes...If you can get some you might want to try titanium piping instead of Ally, harder to weld but it will last longer.. or something else that will resist corrosion from a dilute solution of Sulphuric. Perhaps C263... C263 requires electrolytic etching in HF acid to be able to view the grain structure so would be a good if not expensive bet. But a Nickel based CRES steel would be better than Aluminium if you intend to process more than a few batches.

http://en.wikipedia.org/wiki/Sulfuric_acid
http://www.bssa.org.uk/topics.php?article=33

Electroplating is a different process to Anodizing. Anodizing grows the oxide layer much thicker than it would do under normal atmospheric conditions.

This is especially true of Sulphuric Acid Anodize, the oxide layer grows outwards and inwards. The acid bores down to the pure Aluminum layer where it creates a porous oxide that allows acid to bore down further and so on... (well the action is not completely understood but i'm sticking with the explanation given to me)

When the process is complete you dip in the dye which fills the pores, which are then sealed either by boiled distilled water (dipping/rinsing) or some other sealing solution, we use a Potassium Dichromate solution.

Cleanliness is the key, a good alkaline cleaner to start (or trike if you're really brave - i am not :0) followed by a deoxidising solution to remove the existing oxide layer>>> Desmut (physically and/or chemically) >>> Anodize >>> Dye >>> then Seal >>> dry at a sensible time/temperature.. enough to evapourate any remaining water.. 70°C for an hour or two but low enough not to alter the the physical properties of the Aluminium (depending on Alloy, typical precipation temperature = 120°C) - Probably not important but might be if the part will see significant stresses in it's life time.

edit: another good guide: http://www.scribd.com/doc/24050387/How-to-Anodizing-Aluminum

edit 2: Our technical manager informs me there's a large supply of no longer used anodize dyes in our stores... that and we're due to bring online a new SAA bath shortly. Just need to get to our drop hammer division to get some sheet material heat treated/formed and the little bot i've been making will have an awesome black (batman) anodised amouring.. sweet! (It's temporarily using cardboard)

 

[BrendanP]

Hi Slimp, thanks for the detailed post and links. I'll have a a look at the links when Im back in my lab. Im in Melbourne with work on wlan at the moment.