Evaluating Silica Alternatives for Iron Melting
In the refractory monitoring business, we get involved in numerous discussions about silica refractory performance. And the most common issue is cost vs. performance. While I personally wonder, “does a 20ish-cent-per-pound material deserve this much attention?” the fact is it gets the attention. So why does it get so much attention, and how do we decide the best or better material for a given melt shop?
Answering the first part of that question is relatively easy. Silica is cheap and performs well in virtually all iron melting applications with coreless induction furnaces. Historically, it is very stable and predictable given that it’s been installed, sintered and used within the guidelines of the various manufacturers. I can’t recall anyone telling me during a customer or sales visit that they didn’t have a planned reline schedule; always weekends, never weekdays – and after a predetermined number of weeks or tons of throughput. That’s how it’s done. Nearly the perfect fit for human nature, with no surprises or changes.
Yet people want more from silica. It has to last longer, wear better, be more tolerant to temperature variations, and not run afoul of PELs. So we’ve developed technologies and material variations to help us with most of these demands. A short list includes (in no particular order):
Fused and Mixed Element Materials
PLC (automated) Controls
When it comes to installing dry rammed or vibrated lining material, my personal favorite is the Gradmatic® system. I believe that the use of it with bulk bags is not only efficient, consistent and ergonomic; but also the best at limiting silica exposure. That’s my opinion; and furthermore, I expect the new OSHA PEL of 50µg/m3 will make the Gradmatic® an even more common tool in the coming months.
As lining removal systems began to be retrofitted on coreless furnaces, I was fortunate to be involved with some of the very first so-called PUSH OUT units. While I don’t claim to be an expert, the errors made during those first few installations are the kinds of lessons that stick with us best – embarrassing and costly. There are other methods that reduce the manpower and physical dangers of manually digging out linings, and I can’t find an acceptable excuse not to employ them.
We’ve played with silica in countless ways to get what we want from it – larger grains to improve wear, less binder, more binder, attempts to make it “dust free”, and so on. When lining push out systems came out, the coil grout got tapered to make it easier to push out the spent lining. Again, the binder got modified. Did we need more binder so the lining wouldn’t break too easily during the push out? Less so it didn’t sinter too deep where the grout was thickest? It depended on who I talked to. The end result was that silica endured. The cheapest of the refractory commodities is remarkably understanding to human intervention. In retrospect we’ve done more than we needed to. Experience taught me that the grout taper, while helpful, didn’t need to be much at all; about 1% - even less in furnaces >10 tons.
Lately we’ve been in a number of shops in the U.S. and Japan that are experimenting with fused silica and mullite mixes. In a few cases they’ve gone through 2 or more years of trials to figure out what works best, and there’s a myriad of variables. While users believe that fused silica is more resistant to thermal shock, cracking and volume change, they need many trials to prove to themselves whether or not it lasts longer. It costs a little more than conventional silica but, since it’s silica, it’s still pretty cheap. My personal observation so far is that, even after many trials, they aren’t getting much greater life. In fact, what greater life they have gotten is more than likely the result of increased attention to detail. The philosophical pattern goes somewhat like this:
Invest extensive time with vendors and internal experts to convince ourselves to try something slightly different.
Commit to spending several cents more per pound for the requisite trials. (Sorry, but 1 trial is no trial at all.)
Install and sinter new lining with more care and more people involved than in the past decade.
Control bath temperature, chemistry, cool-downs and cold-starts like we’re bathing a new born babe.
Measure the lining diameter every weekend or down day that permits.
Reline after the same tonnage or weeks as before.
At some point in the process the law of diminishing returns takes over. Slowly, people lose interest and we end up exactly where we started.
It is likewise with even more costly materials like alumina/silica (mullite) refractories. Knowing the cost of the lining material has increased by a factor of 3 or more, the user has to get more life from the trial material. But now the rules have changed! This new material gets treated differently so our process changes. It’s not as forgiving as was plain silica. In addition to being costlier, it requires more attention from our personnel during down time – it needs to be kept hot enough to avoid shrink cracking. This isn’t necessarily bad; it probably doesn’t cost any more either. But it’s something new added to the process. Then what? See item 6 above.
Well keeping an empty lining hot isn’t new. For a couple decades we’ve used burners with programmable controls and thermocouples to do the job. The same can be done using induction power and a full charge in the furnace. It makes your cold-start go a little faster and has been proven to help extend lining campaigns.
It’s time to move on to the second half of my question, “how do we decide the best or better material for a given melt shop?” Anecdotal evidence suggests the answer is a semi-educated guess at best! I said earlier that 1 trial is no trial at all. We have to consider all the variables (i.e. things that can go wrong) during a trial. We also have to accept that a trial of one doesn’t begin to qualify as statistically significant. Given that a silica lining can last from a few weeks to a few months, testing just one new material can take up to a year more.
Let’s say that instead of testing a slightly different silica (fused, larger grains, etc.), a completely different and much costlier mullite is tested. Now the pressure is on to get results sooner. Time to repeat steps 1 through 6 from above I guess. But we’re stuck – weekend relines are the norm based on how the foundry operates. Practically all shops measure and record lining diameters on weekends, provided they’re not running. It’s good practice. Eventually someone has to decide that you’re going to risk an extra week on the new material. Do you want to be the one making that call?
Think about it, you’ve cautiously installed and ran the lining for the normal cycle, and now it’s time to justify the extra expense of this “wonder lining”. You’ve done the due diligence to the best of your ability and within what existing technology provides (which isn’t much). Should you decide to add one more week to the campaign, you get to sweat it out alone. You might as well be jumping off a cliff, because the risk is way higher than the reward. No amount of money spent on a new refractory can guarantee its performance.
There are two things lacking here in spite of all you’ve invested – data and support. We say that data is objective, but we only have it as far as when we last measured the lining. But your furnace is full of molten iron again. Now what? Few are the people or vendors that will say “keep going.” They’ll hedge; there will be disclaimers. From every source except 1, you can’t get what you truly need – data, proof, security.
Hearsay and comparisons to what happened in someone else’s furnace aren’t valid. With data and support you can make decisions confidently. And with day-by-day, and even minute-by-minute, knowledge; testing lining materials becomes a matter of science and fact. That extra investment in new material and your company’s resources can finally get you past step 6. Time and tonnage on your lining will go up.
Evaluating the different silica alternatives can take weeks, or it can take years. Without reliable data, the final decision is still going to be a semi-educated guess.
This information is the opinion of the author and is the legal property of Saveway USA Corporation. Any use of the information contained herein is at the sole risk of the reader, and Saveway USA Corporation and its representatives are held harmless from any consequences. This information is not provided as professional advice or consultation. 26 July 2016 Copyright Saveway USA Corporation, North Canton OH, Myerstown PA www.savewayusa.com