Removing Oxides for Alum Processing 1

[swertel]

I'm running into a surface finish problem due to, what we think, are oxides after heat treating aluminum.

The Part
The part is an aluminum tube 6061-O per ASTM B221 or ASTM B241. It is a thin-walled cylindrical tube used as a pressure vessel.

The Process
After initial forming, the 6061 gets heat treated to a final -T6 condition. No specification listed on how to heat treat it. After heat treating, the part is slightly warped so the manufacturer does another ironing process on the tube to straighten it out. Afterward, it goes to anodize, Type II Class 2.

The Problem
During the last ironing step, after heat treat, small scratches are dug into the outside diameter of the cylinder. They are not cracks and do not go very deep. But, this cylinder is part of a telescoping tube device sealed by an oring, so the entire length must have a suitable surface finish for the oring. The scratch may be too deep.

The manufacturer has tested numerous concepts to remove the problem, including modifying the ironing ring, different coatings on the ring, and different lubricants on the ring. We are stuck.

The Possible?
This is where we need some help, mainly in the form of ideas.
1) Is there a heat treat spec that would prevent oxides, assuming the problem is oxides, from forming on the surface of the aluminum tube?
2) Is there a heat treat process that would be less prone to warping the part so we wouldn't have to do the final ironing step?
3) Is there a method to clean the oxides off of the surface? We thought about etching, and may test it, but would prefer something less aggressive. Would plasma treating work?
4) Brainstorm ideas?

Thanks.

--Scott

http://wertel.eng.pro

 

[DesignerGuy16]

Sounds like you may need/want to polish the OD before anodizing. If it's a really long tube you might not be able to get it into a lathe to polish it.

If that's an issue, try bead blasting it. Any scratches on the surface will even out and reduce the stress concentrations at these points (since you mentioned it's a pressure vessel).

I don't think there's a possibility to remove the oxides from the tube surface as they naturally occur immediately after being extruded. I'd just add a finish operation before you anodize.

James Spisich
Design Engineer, CSWP

 

[swertel]

The anodizing doesn't seem to be the problem. Actually, is a psuedo-solution because the etching process during anodize removes the oxides. The problem is that the ironing process "grabs" an oxide (crystal) and drags it across the surface.

Bead blasting after heat treat prior to the last ironing step may be work. Thanks for the idea.

--Scott

http://wertel.eng.pro

 

[DesignerGuy16]

I mentioned the anodize that you want to finish the OD before anodizing. Otherwise you'll just strip the anodize.

The aluminum oxide layer is naturally rather weak and succeptable to chemical reaction, unlike stainless steel where the oxide layer offers protection from rusting and chemicals. That's why anodizing exists.

On that note, I'd doubt you could "fix" your ironing process used to straighten your tubing to not alter the surface. My best recommendation would be to have your ironing die be polished and nitrided. Probably wouldn't hurt if your die was made out of tool steel ast well. I used to design aluminum extrusion dies and this was our recommended adder to lower friction during extrusion.

James Spisich
Design Engineer, CSWP

 

[swertel]

I think we're on the same page. Polishing (even steel wool) may be too long a process to make it cost effective, but bead blasting could be economical.

I thought about bead blasting prior to the last ironing operation because the oxide layer will immediately return after any polishing or manufacturing operations. The problem appears to be large oxide particles causing a scratch as the die slides across the tube, pulling the particle with it. Bead blasting should make the particles smaller and therefore a)not get caught by the die or b)create a smaller scratch if it does. I read your prior post to iron first, then bead blast. Either way would probably work.

We hydrotest these tubes after anodize so I know the scratch won't affect burst strength. It's ends up being a cosmetic issue as well as a surface finish issue for the oring seal.

Our manufacturer has tested several lubricants and finishes for their die. They mentioned a specific finish to increase hardness of the die; I wasn't able to write it down in my notes. They are also making new dies with other base materials. I don't know the current die material, but they state it is harder than tool steel. That's why we are under the impression that the scratches are caused by oxide particles being dragged across the part with the die during the ironing operation. We're focusing on ways to eliminate or reduce the particles. But, as mentioned before, all ideas are welcome.

Thanks again for your continued support.

--Scott

http://wertel.eng.pro

 

[CoryPad]

1. None that I know.
2. Warping comes from uncontrolled quenching. The way to minimize warping is to add some controls. Can you use fixtures? Make sure not to drop the tubes sideways into the quenchant - drop them vertically.
3. Lots of cleaning methods available: mechanical, chemical and electro-chemical. Work with a surface treatment company like Henkel.

 

[swertel]

Thanks for the response, Cory.

1) I agree that there are none that I know of. But there are several people more experienced with heat treating than me, and I was wondering if a lower temperature and longer dwell time would result in smaller oxide particles. It's aluminum, oxide forms. Is there a way to control how it forms while still getting from a -O to a -T6 finish?

2) I will follow up with the supplier on its quench and temper process, holding fixtures, etc. I have low confidence that we will ever get rid of the final ironing step because of the tolerances involved. To my knowledge, there is always some level of warping when heat treating (i.e. thermal expansion). But the potential is there and that would be a savings in all aspects of the process.

3) Can you point me to any specific ones? I'm familiar with lots of general processes, but without having a vendor in our pocket to get free advice on, we don't have the resources to do a full-on study, which is probably needed in this case.

--Scott

http://wertel.eng.pro

 

[DesignerGuy16]

You might also consult General Magnaplate regarding some of their finishes. They have a number of very high hardness, low friction coatings available.

James Spisich
Design Engineer, CSWP

 

[Ron]

Anodize only...don't polish.

 

[unclesyd]

You might try adding a lubricant during your ironing process. The lubricant could be wet surface or a dry film to be determined by a little experimenting. On some of our processes we used a boric acid based and cetyl alcohol based lubricant for severe AL forming. Natural waxes are alo good Al lubricants.

We also did a light phosphating on some Aluminum parts to prevent problems on deep drawing.

Another possibility, an additional step, would be to bright dip the Aluminum tube prior to the sizing step.

Instead of a bead blast you may want to try a soda blasting.

these people make some very good products for mechanically polishing Aluminum

https://micro-surface.com/index.php?main_page=page&id=15

A question or two.

Could you possibly form the tube in the t6 temper?

Could you use something like an industrial cylinder?

Is this a one, two, three, or many project?

 

[CoryPad]

Scott,

Here is one particular product from Henkel:

Deoxalume® 2310™

Here is a description from the Henkel site:

Turco® Deoxalume® 2310™ is a chromium-free concentrated acidic liquid product specifically formulated for deoxidizing and desmutting wrought aluminum alloys. Turco Deoxalume 2310 is extremely effective in removing surface oxides, discolorations due to heat treatment or thermal deburring, and smut which develops during alkaline etching and chemical milling.

 

[TVP]

Scott,

Even with your extremely thorough summary, I still think there are some missing details that would potentially identify the best way forward. What is the exact process used to clean/lubricate the parts after heat treat but before ironing? Because the part is now heat treated, the flow stress is considerably higher, and with the degree of surface expansion produced during the ironing process, the lubricant will be extremely critical. The process after heat treat should be as follows:

1. Cleaning - typically alkaline cleaning, 2-10% concentration at 70-95 C, 5-15 minutes

2. Pickling - can be either alkaline (2-20% NaOH in water, 40-70 C, 1-5 minutes) or acidic (phosphoric + hydrofluoric acid, 20-30 acid points, 20-30 C, 4-6 minutes.

3. Conversion coating - best coatings are either calcium aluminate or aluminum fluoride. If there is no conversion coating, then the only separation between the workpiece and the tools is from the lubricant, which is likely either a stearate soap or mineral oil/grease, neither of which has sufficient boundary protection under high normal pressures. There are some phosphates that may suitable, but I am not as familiar with these products.

In any case, I think that their tooling setup is less than optimal, and that is the true root cause of the defect. For long tubular parts, punch guiding is absolutely critical, not just during the insertion and forming phases, but also during the ejection phase. Have they determined whether the defect is occuring during the initial forming phase, or during ejection when the elastic deflection of the die (springback) causes the part to be "squeezed" between the punch/mandrel? Coating and lubrication will be heavily taxed if the entire tool system is not optimal.

How many ironing rings are being used? If it is just one, then maybe they need to split it into two successive rings. Definitely coat the rings. Is this operation performed in a suitable long-stroke mechanical press, or in a cobbled-together "draw bench" that is really just a cylinder with a punch attached and an ironing ring? I have seen the latter used with some success, but long parts in the heat treated condition with no conversion coating and just a oil or grease for lubrication could easily lead to problems.

 

[EngineerTex]

This may be a little on the wild side, but what about ironing it from the opposite face? If your tube needs the outer surface to be straight/flat, draw it over a mandrel. If your inner surface is the critical surface, try pulling it through another die.

Just a thought.

Engineering is not the science behind building. It is the science behind not building.

 

[swertel]

Thanks for all the latest ideas. They are really helpful and I will be relaying some of these to our manufacturer, as well as asking the questions posed. (Ah, the benefits of a 9/80 work schedule. We were closed on Friday so I didn't have to think about this all weekend, but our supplier was chomping at the bit knowing he's losing valuable time.)

And, Cory. I attended a bachelor party this weekend. Lots and LOTS of golf, but that's a little off topic. On topic is the fact that one of the groomsmen is a sales rep for Henkel. He and I had a long discussion on this problem. Looks like I have a vendor "in pocket" now.

--Scott

http://wertel.eng.pro

 

[MikeHalloran]

Anodizing leaves aluminum's surface very hard, and micro-cracked.

It's not a good mating surface for a dynamic o-ring.

Either polish and don't anodize,
OR
chrome plate and polish.



Mike Halloran
Pembroke Pines, FL, USA

 

[swertel]

Some more information from our supplier.

Original ring has a Chromium Nitride finish to 78 Rockwell.
They are having another ring made out of carbide.

They use water soluble lubricants:
Macco 472 (preferred because it is easier to clean) or
Polylube (Polydraw 847)

There is a history that the previous supplier - who now refuses to even bid on this part because it was such a PITA - used Okyte deoxidizer a pre-heat treat cleaner.

The press is a horizontal press using hydraulics to move the ring in the press plate and the stripping mechanism. The wall thickness prior to heat treat is drawn to .0720" and then .0705" after heat treat. It is on the 2nd draw that the scratches occur. They are difficult to see because the lubricant hides the scratches. They don't notice the extent until after cleaning. But, the supplier had done some testing and determined that the scratches are caused during the forward stroke and not the return stroke. When cleaning the ring, there is built-up material on the forward part of the ring. It is a single ring to create the form, but they do have a secondary ring that is used for engagement / alignment.

During development of their process, our supplier realized that heat treating and quenching was going to be a difficult issue in order to maintain shape. They worked out specific method to hold the parts while dipping them in a vertical orientation, including vent holes and other means to remove all trapped air and get a uniform quench.

Solutions:
Because these are scratches, mainly cosmetic, and have little reason to infer any performance problems with the part, we are going to do our due diligence with a burst pressure test (we already hydrostatically test these, but not to burst) and salt-fog test. These two tests should verify & validate an acceptable part.

To prevent the scratches from re-occurring, we're going to have to run some other tests using a variation of cleaning or addition processing steps.

Thanks everyone for your help up to this point.

--Scott

http://wertel.eng.pro

 

[TVP]

Scott,

You have confirmed the problem I suspected, which is (1) the ironing press/tooling setup is suboptimal and (2) the lubrication is suboptimal as well. In general, carbide ironing rings are better than PVD coated steel rings, but I think that this is secondary to the other two points. Regarding (1), it looks like you are stuck here, which is a shame. These horizontal arrangements are grossly inferior to true, long-stroke forging/extrusion presses, which can be either mechanical or hydraulic, especially for a tube that has already been precipitation hardened (artificially aged).

Regarding (2), there are two separate problems here, one is the lack of a conversion coating on the part prior to forming and the other is the lubricant being used. As I mentioned previously, this type of operation really should use a conversion coating, preferably Chemetall Gardobond® Z3900 (calcium aluminate), with some type of lubricant applied over it. The lubricant could either be a dry film (Chemetall has a number of different products, either stearate suspensions or water-based polymer dispersions) or some type of mineral oil-based product (straight or soluble oil). Polydraw 847 is a soluble oil product (oil suspended within water), which means it will have difficulty maintaining film strength under severe conditions, especially if it is further diluted. Without a conversion coating beneath it, I would not expect great results, but it should be fine if used with a conversion coating.

 

[swertel]

Thanks, TVP. It seems we may be on the market for a second source for this part. You wouldn't happen to know anyone who may want to bid it? I'll forward their name to my Buyer.

--Scott

http://wertel.eng.pro

 

[TVP]

There are two suppliers in the USA that I would initially contact:

Mueller Industries (Marysville, MI)

http://www.muellerindustriesipd.com/category/impact

Metal Impact (suburban Chicago)

http://216.69.150.249/

Just out of curiosity, is the closest international airport to your current supplier in Memphis, TN?

 

[swertel]

No, they are a bit north of the Mason Dixon line.

--Scott

http://wertel.eng.pro