Low CTE castings: NiResist, Invar, or ? 1

[dmalicky]

We need to cast some ~16" diameter pump housings with good dimensionally stability over a temp range 0 to 400 deg F. To maintain required rotor tip clearances over that temp range, we need a material with a CTE of ~4.5*10^-6/deg F or lower (~8... /deg C).

The only alloys I can find that meet that requirement are
- NiResist type 5, D5, and variants
- Cast Invar

Other characteristics:
- Reasonably good machinability. I've read the NiResists are like mild steel, which is fine.
- Good dimensional stability over a period of years. I understand this is a problem for the machinable type of wrought Invar 36; I'm assuming it is also true for the cast version. But what about NiResist 5/D5--are they susceptible for the same reasons since they have the same % Nickel? Does annealing really solve the problem for either?
- Cost... I understand nickel has gone thru the roof the past year, so would like to avoid Invar or NiResist 5/D5 if possible. But if they are the only option...
- Corrosion resistance not very important (equiv. to cast iron is fine).

Does anyone have any experience with these casting alloys? Or can anyone recommend an alternative?

Thanks,
David

 

[metengr]

In reading your post, I guess I am somewhat confused about why you need such dimensional stability with a pump housing at a maximum operating temperature of 400 deg F. If the fluid you are going to pump is not corrosive, I would consider a cast steel pump housing. It can be stress relieved after fabrication to remove residual stresses.

As far as pump clearances, the radial and axial clearances can be accounted for in design between the impeller and pump housing seals and bushings. I see no reason to switch to materials that have low CTE, and could be difficult to obtain and repair (if the need arises)in the field.

 

[strokersix]

Be careful with the CTE for Invar. It has a low CTE at room temperature but increased CTE at higher temperature. I don't recall the exact numbers.

 

[dmalicky]

Good question, Metengr, sorry, more details: this is a solids pump--a rotary airlock feeder (pneumatic conveyance). Airlocks need less than 0.010 tip clearances to minimize leakage, ideally around 0.004. The blades need to be thin to maximize volumetric capacity, but this causes them to heat much faster than the housing at startup, and thus run into the housing (if the blades are not low CTE). If the clearances are increased to prevent startup bind, or if low CTE blades are used, then after the housing heats up the clearances will also be unacceptable. We can make the blades of Invar or NiResist, e.g., but still need a housing material with similar CTE.

This is an old problem in airlock design, and yes, there are a number of design "fixes"--pre-heated housings, conical bores with plunging rotor, and adjustable housing faces being the usual ones (and many others on the drawing board). We're looking for a totally passive sol'n.

Thanks,
David

 

[arunmrao]

dmalicky,
Your last post cleared some of the doubts I had in my mind about the need for such a close clearance( hence I was hesitating to respond). Well I have experience in manufacturing Invar castings and also Niresist D2 type(pump parts). But I have not had experience casting invar pump housing.

There are many issues related to this which if you desire we can discuss directly.

But please remember that you are dealing with a very expensive alloy as Ni is now very costly.

 

[EdStainless]

I have cast all of the various Ni-Resist grades as parts for centrifugal pumps. D5B is nice to work with, if you can afford it.
The chemistry control has to ge good and the heat treatment is critical. The stabalization/stress relief has a large impact on the expansion rates.
The D5B probably has better mechanical properties than cast Invar. It deffinitely has better corrosion and wear resistance. It also does cast nicely.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[dmalicky]

Thanks, everyone, for your tips.

EdStainless, I'm reading in the NIDI Pub (R. Covert et al., 11018) on Ni-Resists and it indicates the low chromium alloys machine easiest--e.g., 5 or D-5. But, I don't know if the difference is really significant or how the other factors play in. Any experience on how machining and casting compare between 5, D-5, D-5B? Abrasion resistance is also a plus. Machining is a big concern because we have to bore 16" diameter x 16" long.

Sounds like we need a foundry well-experienced with these alloys. Any recommendations for one in the midwest (we are in Kansas)?

On cost, does anyone know what has been affecting the price of Nickel the past year? I see it has gone from $4 to nearly $8 /lb and very volatile to boot.

Thanks again,
David

 

[EdStainless]

The Cr in 5B is to balance the CTE. The machinability isn't bad, as long as your anneal is good. I would suggest annealing the castings, machining and then a stess relief/stabalization treatment at a subcritial temperature.

Is the old Big4 foundry still in operation? What about Tonkawa(OK)? I think the one in Pittsbugh KS is closed. The others that I can think of in the region would not cast something that large.

I presume that you are casting a cylinder and then finish machining it.

As for cast, all metals prices have gone up at least 50%, some like Ni have doubled, and others are up a lot more than that (Mo has gone from $5 to $18). I wouldn't surprise me to see the raw material cost for D5B to be at least $5/lb without any casting costs. The best foundary practices for these alloys would be to buy master alloy that is melted in bulk and tehn re-melt that mixing it with a controled amount of scrap and minor alloy additions.

= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[dmalicky]

Looks like EdStainless's $/lb estimate was pretty good, based on the first quote we've gotten so far. Harris Industries in TX and Midwest Metals in MN are the closest ones I could find that would do these types of NiResist. Anyone have experience with either of those? Big4 is gone, Tonkawa has can't do our 400 lb weight.

EdStainless, I wasn't sure what you meant by the Cr in 5B balancing the CTE: More stable with respect to temperature, closer match to something else...? Yes, we are casting a cylinder then boring the ID.

Thanks again for everyone's help. David