Piano World Home Page Forums Our Most Popular Forums Piano Forum Sand casting, vacuum casting and cast iron

I was following a recent thread discussing the various casting techniques for plates and looking for a place to post, but unfortunately, I missed the bus....

I will try and do justice to this topic by kicking it off with some metallurgy.

Cast iron is an iron-carbon-silicon alloy. Steel is similar, but with a lower carbon content (<2.1%). The iron atoms in cast iron have a certain order to them. At ambient temperatures, this structure is known as body centered cubic (BCC). At higher temperatures, the structure changes to face-centered cubic (FCC). Just below the melting point, it reverts back to BCC.

Now, imagine all of this in reverse. Your ladle of molten cast iron is poured into a mold, where it cools, first forming the BCC crystal structure. After further cooling, the crystal structure changes to FCC, and by the time you reach room temperature, you're back to BCC. This is just for pure iron, with no carbon present. When you introduce carbon atoms, they sit at interstitial locations, between the iron atoms. This gets interesting because carbon impedes the motion of the iron atoms (dislocation) as they try to change crystal structure. Those blocked dislocations give a material increased strength and hardness, but also increased brittleness. In rapidly quenched steels, this material is known as martensite. Reheating steel in this state allows the carbon atoms to relocate, resulting in tempered martensite, which has improved ductility due to increased dislocation mobility.

The surfaces of thick sections cool more quickly than the center. Imagine what happens when surface goes through the crystal structure change from BCC to FCC, while the center is still going to be stuck at BCC for a while, trying to catch up. Distortion. Yes, thermal shrinkage plays a role, but there's a ton of metallurgical activity here as well.

Check out http://www.metalpass.com/metaldoc/paper.aspx?docID=66

Now, on to casting methods. Sand casting is old school, good for small product batches. V-casting is much newer, more expensive to set up and much more suited to mass production. Del has noted this in several posts. Very very few new foundries are being built in North America. The few that are being built here are not about to begin casting piano plates, that's for sure! Faziolis would suddenly appear very affordable if this were the case. Both Europe and Asia are ahead of North America with this method. I'm sure that labor costs play a role, but keep in mind that many foundries on this continent are over a hundred years old, while both Europe and Asia had to rebuild after World War II.

OK, so MetalMan, from your perspective is there a difference in the crystalline structure of a sand cast plate that is slowly cooled from a quickly cooled vpro technology?

Please forgive me if it was answered or inferred in your first post. I am a little slow this morning.

Interesting read... very technical. (I like that )

This may be slightly OT, but I have restored a few old Ford farm tractors from the late 1940's and early 1950's. The steel and cast iron in those old tractors are tough as nails. I read that Henry Ford had the best metal foundries in the world along about that time.

Metallurgy is indeed an interesting topic.

Rick

Rich,

It's a battle between cooling uniformity and the useability of a cast iron post rapid cooling. I'm sure that the vacuum molds incorporate this into their design. The key is being able to control your cooling rate.

I found an excellent presentation which discusses some casting defects quite well. www.me.gatech.edu/jonathan.colton/me4210/castdefect.pdf

Prettiness aside, our sand cast plates (including my beloved Baldwin) are probably closest related to fire hydrants...

Cheers,
Manuel

Rich, I'm sorry. I was also a little slow this morning and didn't answer your question properly. Quick cooling would form a very fine microstructure, with relatively small grains. The arrangement of the atoms within each grain would be BCC for either method though. Generally, grain formationpn occurs instead of grain growth with fast cooling. However, additional alloying elements could affect these properties as well, just as other mechanisms can also promote grain nucleation even with slower cooling.

The only way to really know would be to get a cross sectional cut of a plate and put it under a microscope. I might just try to do this someday, with a piece of someone else's piano....

Just out of curiosity, what do your suppliers tell you about the advantages of one vs the other casting method?

This caught my eye, as I remember visiting a dealer who told me that vacuum casting is inferior. He had a chart that indicated which brands were what type of casting and so on.

I forgot what he claimed was the reason for sand cast being better. It didn't make much sense though.

Do you have any information about the process of seasoning castings, that is, leaving them outside through changes of weather?

If you Google "aging cast iron", you'll see a number of papers/abstracts discussing the strengthening effects of aging cast iron at ambient temperatures. The time window seems to be 6 to 10 days, with claims of up to 10% increase in yield strength. Some of these references state that elevated temperatures accelerate this age strengthening process. Yield strength is important, of course, but a lower yield strength can be mitigated by using more material. I would be surprised if a piano plate was using a factor of safety of less than 5, given the importance of stiffness here.

There are also findings of improved machinability with aging which I think might be more compelling reasons to let the plate age outside of the piano. Otherwise, why wait?

I'm going to guess that the "seasoning" outdoors is really a function of storage space and cost.

If you really believe that last statement then you are very much mistaken about the nature of costs in piano manufacture.

I think he meant that the costs of plates cast in the US would go way up.

Yes exactly, thanks.

Jurgen,

Actually, my comment was directed purely at the cost of new infrastructure in North America, the burden of which would dwarf the actual cost of a 200-300 lb casting. Offhand, do you know how much a new sand cast plate would cost for say a six foot grand? I'm curious because I have rough costs for similar weight castings (the aforementioned fire hydrants) and mold size aside, it would be interesting to know what premium one would pay.

My crude understanding is that the sand cast plates are better than the vacuum cast ones for pianos as the former have a denser molecular structure which does not absorb or distort the sound of the piano.
The best and most expensive pianos usually have sand cast plates while the mass produced pianos, for example Kawai and Yamaha have vacuum processed plates.

I wonder if modern vacuum processing can replicate the more desirable density (for pianos) of sand casting.

Robert.

The interesting thing is that when you go to the top models including each concert grand made in Japan today, they suddenly change form vaccum cast to grey sand casting.

So, they would be perhaps in a better position to explain things.

However, one doesn't need to pay a premium price.

Most of the better Chinese maker I know, use them...

Norbert

The density of the plates by either method will be the same. Porosity would be the only way to change it, but this constitutes a casting failure and is highly undesirable in any casting. Grain sizes may be different due to cooling times, which could be the driver for a different sound. Honestly, with the excitation of all the other reverberant components in a piano, I couldn't see the casting method playing that significant role in the sound. That cast iron plate has a density of 7200 kg/m3, vs 800-900 kg/m3 for all the wood that surrounds it. If someone has data to suggest otherwise (comparisons between these two methods, not alternate materials like steel or aluminum), please share!

Molds for vacuum casting are costly. Prices for concert grands and elite instruments are already high; I'm sure that there are compelling economic reasons to use a smaller batch appropriate casting method such as sand casting for these reasons, even in Asia.

Greetings,
I think the sand cast plates are less likely to have their own resonant frequency. At least, not enough of one to produce a noticeable sound. This is not the case with the vacuum castings. There are a number of C7 Yamaha's that exhibit a distinct "boing" anywhere from the upper fifth octave to the mid 6th. I have seen the same in other brands, but never in a sand-cast plate. This may be coincidence, but I don't think so.
My logic tells me that the sandcasting produces a less consistent product, not only on the macro scale of plate accuracy, but also on the microscopic scale of internal stresses occurring in a more random manner. That may be wildly off base, but in any event, I think the reliance on sand casting, particularly for the high end concert grands, springs from acoustical reasons more than economic ones. The manufacturers don't go to extra effort for nothing...
Regards,

It is unlikely that anyone has set up vacuum casting foundries for the purpose of testing whether the result sounds better than plates made in sand casting foundries. This has got to be a purely economic decision.

I posted in relation to this topic recently even though I didn't share most of my notes. Piano Technology in a way is a contemporary subject, since piano construction has been a partially empirical process. The application of modern technology in piano manufacturing has been always slow due to the common resistance offered by the industry; since the industry is filled with romanticism, tradition and ignorance (plus also lack of financial resources). All these previously named factors are part of *slow* progress and resistance to the use of technology.

With support of a PhD in metallurgic from Stanford University here in California; I conducted a small research about casting techniques for piano plates manufacturing. The objective of this investigation was simply to clearly understand the process and what kind of effect has in the finished product.

The following is a compound of some of my notes:
In the V process the sand is dry and vacuum is used to assist the filling of the sand mold with the liquid metal. The dry sand causes the metal to cool more slowly allowing easier metal flow in narrow sections. That, along with the vacuum assist permits large slender castings to be made such as is the case with the cast iron frames in Japanese pianos. Normal sand casting uses wet sand to permit compaction and to increase metal cooling rate. Typical sand castings are also poured in air (no vacuum). The result is that typical air poured castings must be have larger section thicknesses due to the faster cooling (metal can’t fill small section molds when it is cooling rapidly) and there is typically much more porosity.

The result is that typical sand castings are bulkier, have more porosity and other defects, but the metal in places may be slightly stronger due to the more rapid cooling. On the other hand the vacuum castings are much more slender, have less defects, but the tensile strength is slightly lower due to the lower cooling rate. In most situations the overall strength of a “more perfect” casting is equivalent or higher than that of a casting containing porosity.

Lastly the cast iron piano frame is not a “strength critical” component. It is a “stiffness critical” component as most engineering applications are. This tension of the strings is only 30-40000 lbs. which does not stress the cast iron very much. The stiffness, or rigidity, or elastic modulus (different terms for the same parameter) which is a constant physical parameter and not affected by strength, is unaffected by any casting method. It is this stiffness which resists the tension of the strings and is pretty much insensitive to casting method. If you were to examine the cast iron frame closely before and after the strings are tensioned, you would note that the frame compressed slightly. Upon releasing the tension, the frame would return to its original size and shape. (We are talking about very small changes in dimension). If you could measure the stress on the frame at any one point, you would find that the stress was far below that of the typical strength. That is why we say that this is not a strength critical component.

The iron-carbon silicon alloy is not conducive to amplify noise/sound frequencies to the contrary is a dampening material. There is no connection between the casting process and the plate's *lack* of ability to reproduce or conduct sound. Those frequencies delivered by the hammer/strings/soundboard/rim; are very distant from any frequency that can be capture or reproduce by the plate in order to resonate with each other. The plate cannot sympathetically vibrate to these frequencies nor has the capability of radiating energy back to transform mechanical energy into air motion.

In the history of the development of the piano, there has often been a conservative reaction to technological changes in piano manufacture.

Even the iron frame itself was initially regarded with suspicion as it was believed that it compromised the sound quality of the wooden framed piano.

The introduction of over-stringing was also controversial as many believed that straight stringing was better. Even a great manufacturer, Bechstein, was still producing its straight strung model 10 vertical piano in the early 20th century.

Now we are told that in fact the density of the plate from the two different types of manufacture, sand cast and vacuum processing, is similar and that the differences in their respective effects on the sound may be negligible.


Regards,
Robert.

Well, now ladies and gentlemen, let's all sit down and play various pianos.

The we can all discuss the net results....

Norbert