48000

Galvanizing ash or flux contamination of work [Australia] 

February 28, 2008

Greetings from Australia. After long holdups largely related to the shocking price of zinc, the hotdip galvanizing plant that I've previously posted about is up and running.

9 metre long tanks and kettle, hot caustic, hot inhibited sulfuric, hot quad flux, lead bottom kettle with 30 ppm aluminium in the zinc, quench. First batch of 190 items wired to a frame, total weight about 1.5 tonne, 100% acceptable.

But not perfect, hence this post.

Flux is allowed to dry, so work enters zinc with not a sign of spatter. Starting about 5 seconds after immersion and lasting about 10 seconds, slight bubbling which I assume is the flux doing its work. Surprisingly large amount of white smoke.

But within seconds of the slight bubbling, we got a thin film of light-coloured fines over the bath. When the work was withdrawn, this film was picked up by the top surfaces of the jobs giving a rough surface to the zinc coating on those top surfaces. The rest of each job was completely clean, as each job came up through the clear patch created by its top surface.

A quick run over with a hand grinder cleaned the roughness off, but the film shouldn't have been there in the first place.

Possibilities that have occurred to me are (a) fine dross suspended in the bath from the original meltdown and pushed to the surface by the flux on the jobs; (b) quad flux reacting with aluminium in the bath, more than would happen with triple salt flux - the great amount of white smoke suggests the possibility of aluminium chloride. The film was still on the surface of the kettle 12 hours later, with the clear areas still in it from where the jobs had come out.

But these are only thoughts about what might be the nature and cause - I hope somebody can suggest what actually is the nature and cause, so that we can stop it happening.

Bill Reynolds    metallurgical services Melbourne, Victoria, Australia

Bill

the bubbling after immersion sounds quite normal.

The white smoke sounds normal. Part of the flux is ammonium chloride - thats what fumes off. The higher the ratio the worse it is. Quadraflux most fume, triple next, double less etc. But don't let that cause you to think that quadra is no good, its actually good.

You didn't mention skimming before withdrawl.
Its normal to get ash float to the top during that bubbling. Residues of oxides etc. While material is immersed, you use a thin bladed skimmer on a handle to paddle (not vigourously) the ash to one end of the kettle, pushing it between the suspenstion wires. This creates a clean surface to withdraw through, without the ash etc depositing on the work.

In other words, creat a "clear patch" of the whole surface, then withdraw.

Geoff Crowley    galvanizing &    powder coating shop Glasgow, Scotland

 

Sir:
Have you had the zinc tested for aluminum? 30 ppm aluminum(e.g. 0.003% is quite normal). How many kg of "brightener bar" (usually 10% Al) was added and how many kg of zinc are in the kettle?

How many kg of quadraflux did you add and how many liters are in the flux tank? What is the baume' of the flux and what is the temperture? You are correct, quadraflux gives almost not zinc spatter as the steel enters the zinc.

Did you skim off the ash just before pulling the product out of the kettle? What type of skimmer do you use?

There is no dross initially in the zinc, the iron is fully dissolved in the zinc for at least a few days or even a few weeks. Quadaflux does not react more with aluminum than does triple salt. I do agree that the great amount of white smoke indicates aluminum (and too much of it in the zinc). At start-up it is often the case that too much aluminum is added due to a math error.

Regards,

Dr. Thomas H. Cook
- Hot Springs, South Dakota, USA

Sir:
Over the weekend, I thought of a few questions that I wanted to ask you: 1)How many liters do you have in your sulfuric pickle tank? How many liters of concentrated sulfuric acid did you add to this tank? How many liters of inhibitor did you add to this tank? From what country did you buy the inhibitor? How are the acid fumes in the plant? Are the acid fumes: Very Slight, Moderate, Strong, or Very Strong. 2)How are you heating the sulfuric acid tank and to what temperature? 3)How are you heating the quadraflux solution and to what temperature? 4)What is the heating system for the kettle? Is it flat-flame-side-fired, end-fired, or pulsed-end-fired? How many centimeters of insulation are under the kettle (e.g. between the kettle bottom and the concrete support pad)? How many thermocouples are in the kettle and are they in the zinc or on the outside of the kettle wall?

Your observation that the quadrafluxed steel goes into the molten zinc "...with not a sign of spatter...." is exactly correct. This is also my experience and it is interesting that a drying oven is NOT required and in fact NOT desirable.

Regards,

Dr. Thomas H. Cook
- Hot Springs, South Dakota, USA

Hi Geoff and Thomas

Thank you both for your responses. In our innocence, we anticipated that a clean zinc surface going in would give a clean zinc surface coming out - the concept of needing to skim while immersed hadn't occurred to us. Your advice, Geoff, goes direct to the heart of our problem.

The tanks and kettles are 9 metres to accommodate the large beams and columns that will be part of the work load. But on our first production batch we used all small items (190 of them), not long beams etc, and wired them to use the full area of the lifting frame and kettle. Paddling between the wires in such a case is just not practicable. Under other circumstances, with several large jobs on the frame instead of hundreds of small ones, skimming between the wires would be easily enough done.

My thoughts now are to both reduce the amount of flux on the work and avoid skimming between the wires, by...

(1) add a surfactant (pentaerythritol?) to the flux to get better draining and hence less flux on the work;
(2) try coming out of the flux at creep speed rather than normal hoist speed (same slow speed as coming out of the zinc) and see if that also gives less flux on the work;
(3) on small work, use only half the length of the kettle - go in clean at one end then move the frame to the other end while immersed.

Do these sound sensible intentions?

I chose to use quad flux right from the start in the light of Dr Cook's many postings and publications on the subject of fluxes, and hope to be able to stay with it (or at worst somewhere between triple and quad).

The details that Dr Cook requested about the other aspects of plant operation seem not to impact on our surface crud problem. All the maths were right, and the caustic, sulfuric and flux tanks are all run at 70 degrees C and are not giving any problems. The kettle has 220 tonnes zinc + 24 tonnes lead, is pulse-end-fired and held at 448-450 degrees C. Aluminium is added as 10 kg bars of 4% aluminium (diecasting) alloy.

Thanks again, and we'll be grateful for your comment on this post.

Bill Reynolds    metallurgical services Melbourne, Victoria, Australia

Geoff,
You say something like "quadraflux is actually good." Do you use it? If so what baume,' what ACN, and what temperature? If you do not use quadraflux what do you use? Do you have kettle enclosures?
Regards,

Dr. Thomas H. Cook
- Hot Springs, South Dakota, USA

Bill,
Your very clear description of the problem allowed a quick solution.

You need not reinvent flux usage. What is the density of your flux (in Baume' or g/mL measured at room temp.) If you get too little flux on the product it will not galvanize. If you get too much flux, the outer layer will be dry and the inner layer wet and spatter badly at the kettle. If your purpose in reducing flux on the steel is to reduce smoke (white smoke that rapidly travels upward) you cannot "win." With heavy baume' double salt you get brown/black heavy smoke at head height and it does not dissipate rapidly. With triple or quadraflux you get white smoke that goes straight up to the roof and out any open doors. The real solution to the smoke problem is a kettle enclosure and bag house.

1) I looked up "pentaerythritol." It is a white solid costing about $1 US/pound. It is listed in the pocket guide to chemical hazards with ratings of 5 to 15 mg/cu M for OSHA and NIOSH. It is an eye, skin, and resp sys irritant. I am not familiar with this product as a surfactant. It would not be my choice without testing. It is true that a good surfactant will reduce the surface tension of flux solution from 80 dynes/sq cm to 30 dynes/sq cm, which does reduce LIQUID flux solution in root threads and around weldments.

2) Coming out of the flux solution very slowly will not produce less flux on the product. The amount of flux on the product is adjusted by baume,' and temperature. Always the product should be in the flux at least 3 to 5 minutes so the product heats up and then when the product is withdrawn the flux dries from the heat within the product.

3) Using half a lift and going in half the kettle and then tractoring to the other end can be done. Remember however that the total time the product is in the zinc determines the thickness of the zinc. (e.g. longer in the zinc means thicker zinc on the steel).

I asked several unrelated questions so as to avoid common start up problems. Recently I was in a plant that used 3 barrels of sulfuric acid inhibitor per tank and the fumes were so bad that I could not even work properly. I used one US gallon per tank of good inhibitor and the fumes were reduced to ZERO. About 25 years ago I tested about 100 sulfuric acid inhibitors. Only about 5 were good. The others caused more problems that they solved. Some vaporize out of the acid, some coat the tanks with black goo, some are horribly toxic and smelly, some just plain do not work (e.g. protect the bare steel from further acid attack after pickling AND greatly reduce fumes in the plant).

If the heating coils are made of the wrong material then they will fail in a few weeks or months and can destroy the boiler.

Quadraflux until recently is not a commonly sold flux and I was curious if it is available in Australia. Is it? It can easily be made from double or triple salt or even from zinc chloride and ammonium chloride.

I trust that the 4% aluminum diecasting alloy contains only aluminum and zinc. Other elements (e.g. copper, Mg, etc.) would be very harmful. If the entire plant is filled with white smoke it is possible that you have too much aluminum in the zinc. It would be useful to know how many kg of casting alloy was used.

I am still wondering how you will use spray rinses to rinse the inside of the product and what you will do with this spray rinse water? (especially when production increases)

If you get by this way please stop in and catch some rainbows. They are now 2.5 pounds and will be 4 pounds by June.

Regards,

Dr. Thomas H. Cook
- Hot Springs, South Dakota, USA

Bill,

Skimming is uanavoidable. Don't try to dilute your flux, nor to get less flux on the job, or you'll encounter other problems. The flux is there for the time of immersion, not so much as a pretreatment before immersion. Insufficient flux crystals on the job (dried in place there after the water carrying it evaporates) and you'll get bare patches, lumpiness and other quality issues.

The zinc is continually oxidizing on the surface of the molten kettleful also. Not so much at startup and for a week or two, but in production use, you'll see the skin forming all the time. To get good clean bright galvanizing, you must skim before withdrawl.

It can be difficult if you have not arranged the suspension wires to facilitate, so think about this when wiring up the steelwork. Try to get the wires in straight lines across the bath. They can be as close as 200mm apart between rows of wires and still aloow skimming.

My design of skimmer is a blade of flatbar steel, about 40x3mm and about 400-500 long. Attached to the middle of this is a 10-12mm roundbar coming off at rightangles, then curving back to be about an angle of 30degrees off parallel. This handle must be long enough to hold from alongside the kettle where you walk along sort of floating the skimmer about half submerged and half exposed along the blade length. Push the skimming from one end of the kettle to the other. When you encounter a line of wires, give an extra push and the ash will float past the wires, to be picked up by the skimmer on the other side. Continue the full length.

After several dips, you could have a shovel full (or 3) of ash floating about at one end. Use the skimmer as a chopping knife to "chop it up" to dislodge any metallics, then use a perforated spoon (make one for this) to remove the ash. Put it in a drum or other suitable container (remember its hot). You can sell this and get perhaps 50% of LME Zinc price for re-use in a variety of industries.

Judicious arrangement of suspension wires will help faciliate skimming. With structural steel you'll be using multiple wires to get the combined tensile strength to support the steel. (If you need more info on this, ask)

Depending on kettle width, you might be tempted to make a single skimmer to span the full length. I don't advise it, its too hard for the operator. Rather have someone else attend and help at this stage, one each side, each with a skimmer. They should be able to skim the full length up to about a 12m kettle. Longer than that and you might need two goes, from the centre to each end, as the ash buildup will be too much to skim easily.

This design of skimmer also makes a good tool for getting off those drips of Zn that due to low angle, don't drain naturally. Try to get as much vertical component in the hanging angle as possible, but there'll always be something that doesn't drain well, doesn't hang well, and has drips. Left to solidify, they become spikes, and it takes labout to remove them. Minimise this work by removing before solid, as the steel withdraws from the Zn.
You'll also find that on some jobs there's ash forming on the surface again after a first layer has emerged, and you can skim between layers of steel as the emerge from the zinc, giving a nicer finish.

Hope this helps.

Geoff Crowley    galvanizing &    powder coating shop Glasgow, Scotland

 

Bill,

I am always concerned about the short and long term safety and heath of workers. In your design did you put in an enclosure around the kettle? If so what type? If not, do you have a shield with a safety glass window so that during immersion into the zinc the kettle crane man (and skimmers) can get behind it to avoid zinc spatter? Do you have audio and flashing lights alarms during immersion in the zinc? Triple and quadraflux are "sleepers" in that 99% of the time there is little kettle spatter, whereas a pocket weld or lack of proper product venting will cause dangerous explosions in the kettle. In the USA while a galvanizer was dipping a heat exchange coil it blew up spattering zinc onto a worker, who, until he died was called the "tin man." Also be aware that the hollow steel tube handle of an ash skimmer shot a molten ball of zinc into a workers head, he died. I have never figured out the physics of this "rifle" handle, but it did happen and perhaps by acid cleaning the skimmer. About one third of my clients use wooden skimmers, both regular wood and fire retarded wood. They last a few hours to a few days, and there are ways to make them last longer.

My article "Accidents and Hazards in Hot Dip Galvanizing is in the journal Metal Finishing. It was NOT a well-received article by the industry, but if one worker avoided serious injury it was worth the bad feed-back.

Most accidents hapen to "visitors." If most or all of your workers are new to hot dip galvanizing, perhaps you could hire a galvanizer to speed your learning curve.

Regards,

Dr. Thomas H. Cook
- Hot Springs, South Dakota, USA

Bill, Here are 25 year old, never published, test results of bag-house-dust (as percent).

Type Use Zn NH4Cl Pb Fe Al Cr Cd
Wet Job 4.9 19.5 .012 .135 .037 .079 .001
Wet Spin 4.5 1.1 .010 .140 .088 .002 .004
Dry Job 2.5 1.3 .013 .125 .049 .021 0
Dry Job 12.0 12.6 .120 .400 .370 .002 .001
Dry Job 33.8 33.7 1.26 1.25 .479 .004 .006
Dry Job 37.5 28.6 1.03 1.20 .432 .004 .006
Dry Sheet 4.2 21.2 .009 .045 15.2 0 .001

These data need additional duplication before publishing.

Regards,

Dr. Thomas H. Cook
Hot Springs, South Dakota, USA

Again gentlemen, thanks for your very helpful comments.

As if anticipating your suggestion, Thomas, we were able to get the ex-foreman of a galvanizing plant that was shut down a couple of years ago, to advise on the tricks of manipulating things on the shop floor. He spent this last Friday and Saturday observing and advising, and the skimming after immersion that Geoff kindly described in detail is now described by the plant operator as "a piece of cake". I wasn't at the plant at the weekend, but believe the work was coming out quite clean.

He (the ex-foreman) also expressed surprise at the large amount of white smoke relative to what he used to get using straight triple salt, so we are adjusting the flux back a bit from quad (actually to about 3.2-salt) and dropping its density slightly. Over the months to come, we will play around with that a bit and find the best balance for that plant and type of work between the flux ratio and density and the white smoke and the flux consumption and the zinc coating thickness etc etc...

I'll bring you up to date with operating details later, if the plant owner permits me to.

Bill Reynolds    metallurgical services Melbourne, Victoria, Australia

Bill,
What is 3.2 flux?
Regards, Tom Cook

Dr. Thomas H. Cook
Galvanizing Consultant - Hot Springs, South Dakota, USA

Bill,
I figured out what you mean by 3.2 flux. This flux would be 0.2 units from Triple salt and 0.8 units from quadraflux. This would be a flux with an "ACN" of 1.258. "ACN" = %ammonium chloride/%zinc chloride. ACN for double salt is 0.785 (ZnCl2.2NH4Cl). For triple salt ACN is 1.18 (ZnCl2.3NH4Cl). For Quadraflux ACN is 1.57 (ZnCl2.4NH4Cl). It is not likely that a flux formulator would be knowledgeable in chemical "moles," which is why I like ACN terminology.

In 1972 I tested a galvanizer's flux who had never used anything but triple salt (ACN=1.18). The test results showed an ACN of about 0.7. During galvanizing some rejects had been brought back and only refluxed, dissolving zinc into the flux solution as zinc chloride. Thus the ACN goes to a lower value AND the pH goes UP. We added 2,500 pounds of ammonium chloride to the 5,000 US gallon flux tank and the galvanizing got much, much better (e.g. zinc coating thicknesses were greatly reduced as were drips, icicles, and many fewer rejects).

For an article in Metal Finishing, I made up a flux solution in which I put 1 kg of ammonium chloride and 1 kg of zinc chloride (e.g. ACN=1.00) and had 12 test companies test for ammonium chloride and zinc chloride (Fig. 4 in the article). NONE of these "certified" companies got a usable result for ACN, even though they charged between $50--$300/sample. I also had them test sulfate and iron and about 20% of these companies got usable results.

Your new guy (who worked for a previous galvanizer) was probably accustomed to a flux which he thought was triple salt but in reality had an ACN near or below double salt. The South African's had an ACN of about 0.5 and a baume' of about 40. I took out 3/4 of the old flux solution and added 4 tons of ammonium chloride and water and a good surfactant. I also heated the tank to 71C. The reduction of zinc used was $8,000/day (when zinc was cheap). They took me to Kruger Park on a long weekend for a look see. Your new guy will bring with him all of his good knowledge AND some knowledge that is NOT so good. For example, it is never a good idea to put zinc floor sweepings or dross or ash into the flux tank (to neutralize carried over acid). The technology is simple but you must know and understand what works.

In-plant ACN testing, in-plant iron, and in-plant sulfate testing are readily available. You need these for testing your flux and (sulfate testing) for your spray rinse water. I expect that your flux tank will become contaminated with sulfate within a year due to your spray rinse system.

Regards,

Dr. Thomas H. Cook
Galvanizing Consultant - Hot Springs, South Dakota, USA

Bill,

There are so many, many important items in constructing and operating a galvanizing plant, I am concerned, that things are likely to get screwed up fairly soon. The owners of the plant expect to make a profit above their investment and operating costs. The "floor" galvanizer that you just hired comes from a plant that was shut down a few years ago. Why?

Your owners have invested a few million dollars to build the plant (according to some of your recommendations), it is time to invest a few thousand dollars on the best galvanizing consultants you can find to make sure that it operates as you have designed.

Regards,

Dr. Thomas H. Cook
- South Dakota, USA