Brass shrinkage 5

[HilaryE]

Hi,

I would like to ask a question about the brass shrinkage (CZ121).
I have a brass component in the machine (approx. 50mmx50mmx1000mm) which experiences shrinkage over the course of its life-time (approx. 2-2.5mm on its longest side). The brass component, once fixed in the machine is periodically heated to approx. 230degC and then cooled down to 25degC.
I have made some research with regards to the actual cause of the brass material shrinkage but still not certain about the right answer.

P.S. Is there any technology (e.g. heat treatment) that could eliminate or minimize brass shrinkage effect?

Thank you.

Regards

 

[EdStainless]

If you take the parts, and before they are machined and installed and temperature cycle them, say between 15C and 250C slowly a few times you should never see any further size change.
Are these castings?
Part of it could be a secondary phase in the material, but most likely it is residual stress.

These parts are not tightly clamped are they? If they are then you are causing the distortion. A 1000mm piece of brass heated from 25C to 230C will expand by more than 2mm. If it is clamped tightly you will distort it.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Tmoose]

Some pictures would probably help others understand the situation. Especially me.

I think, if EdStainless's very plausible "upsetting" suggestion is correct, a shrunken part's perpendicular dimensions would increase a little as the shortening occurs.
If there are not convenient features to measure width changes, I'd machine a few datums on new parts to permit quantification in the future.

What is the material the brass part is "fixed" to in the machine.
Does it get heated as well?

 

[HilaryE]

Thank you for your replies.

@EdStainless- The brass bar is rigidly fixed to the machine. I am not actually certain what technology is utilized to manufacture this brass components- I will check this out with the supplier.

@Tmoose- I apologize Tmoose but the design of brass component is confidential. The brass bar is not of simple design. Otherwise I would be more than happy to present its pictures.
I will check the width of the brass prior to use and at the end of its life-time cycle (from the dumped parts stock). Because the width of the bar is relatively small I will check the width dimension on a larger sample size (5-10 pieces).
The brass component is rigidly fixed to the steel part through an aluminium gasket.
The steel component is heated (electrically). The brass part is heated by the steel component.
The brass component is taken out from the machine (disassembled from steel and aluminium gasket) every few weeks for refurbishment, and then fixed back.

Thanks guys. I will check the brass bars dimensions, and will post the data (it will be 2-3 days before I get an access to the bars stock).

Thank you.

 

[EdStainless]

If the steel and brass are both being heated you still have an issue.
The steel will only expand about 1/2 as much as the brass.
This will still result in a lot of stress being put on the brass part.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Tmoose]

Does the steel heating continue to a set temperature, then remain for hours, so the brass and steel achieve ~ equilibrium? I assume something flows into the brass part to require a gasket, or maybe the brass just covers a hole/opening in the steel. Anyhow, some flexibility in connection to the steel might protect the brass.

Is the brass replaced because it becomes too short?

 

[HilaryE]

@EdStainless - It's true, the thermal properties of fixed materials are different and this is part of the problem.

@Tmoose- At the machine start up the steel component is heated to 230degC and then it is kept on that level for a few days. After 2 days of shut down it is heated again. That is right, there is a molten material which flows through the brass component from the steel component (molten material temperature is approximately the same as steel/brass temperature).
I have also wondered on the flexible connection between brass and steel components. The problem may constitute fact of high pressure molten material flowing through them. I am concerned about a leakage if some sort of flexible sealing would be applied.
Sometimes the brass bar is replaced with a new one because the shrinkage is too big. Another time the wear of the brass in specific areas is too big and then the wear is responsible for bar replacement. The brass bar is fixed to the steel component with bolts. The front surface of the brass bar is literally filled with the holes for the blots. There is not much room for making slot holes to allow for the shrinkage. I cannot modify the steel component for this moment as well.

 

[EdStainless]

Why brass? You cannot engineer out this issue. Every time that you heat it it is trying to expand, but the steel won't let it. As a result it yields in compression slightly. And when it cools it is now shorter. This happens every cycle.
What properties are you after? You need a material that more closely matches the thermal expansion of the steel.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[HilaryE]

EdStainless - The brass has been initially (years ago) used because this material is relatively easy to machine. It has intricate shapes which require i.a. micro-machining. Therefore, mild steel or stainless steel are very expensive option. Besides the design of this part keeps changing (optimisation). Ultimately, it may be machined out of even 316 s/s and then the thermal properties of connect parts will be very similar.

 

[HilaryE]

P.S. To my last reply, I clicked submit button by accident
EdStainless - For this moment, however, I have the ongoing study which aims to optimize this part geometrically (process enhancement). Nevertheless, it will take some time, therefore, possible limitation of the shrinkage would be helpful.

 

[EdStainless]

HilaryE, in the long run I would suggest moving to a ferritic stainless (439 perhaps).
The CTE of these is similar to steel, and much lower than for a Cu alloy or an austenitic stainless.
If you need higher hardness the martinsitic stainless alloys (420/430/440) would be good options.
Another good option if you need higher strength than a ferritic, but you don't want to work with a Q&T alloy (with finish machining after hardening) you could look at using a duplex stainless. They are a lot stronger (and harder) than a 300 alloy, have CTE that is between steel and a 300 alloy, and they have better corrosion resistance than 316 (even the lean duplex grades like 2101).

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[HilaryE]

EdStainless,
Thank you for your suggestions.
Indeed, the martinsitic stainless alloys or duplex stainless are one of the best options.
I also considered little bit softer material than s/s, some kind of mild steel for example. The advantage of mild steel potentially is lower cost of its machining when compared to s/s (part has intricate geometry). On the other hand, mild steel in this case would require surface hardening (carbonizing or nitriding- not certain actually which would be better considering protective coating which would be given to mild steel later on) and protective coating (nickel-chromium or similar).
Nevertheless, the part has a large number of through holes (0.3mm diameter) with large L/D aspect ratio. Any coating, such as electroless nickel plating for the brass, would have to leave the micro holes evenly covered with approx 4-6 microns coating, on the entire hole length. The hole blockage with material coating would not be acceptable.

 

[SwinnyGG]

If you're currently using brass, why would an alloy steel version of this component require surface hardening?

 

[HilaryE]

jgKRI - The flow process running through this part poses a great challenge for its material durability. On the market, this part is often made (by various vendors) from high hardness s/s with the extra coating applied. The vendors offer that extra surface treatment to s/s with demonstrable benefits. If we think about the mild steel of some kind (as it it is easier to machine intricate shapes), extra surface treatment is absolute must.

 

[SwinnyGG]

Right... but you're currently using brass, alloy CZ121, with a maximum hardness in the range of 150-160 brinell.

A medium carbon steel alloy such as 1045 is cheap and readily available, has hardness in excess of the material you use now, and will have very closely matched thermal expansion compared to the other steel parts in the assembly, eliminating your problem.

 

[racookpe1978]

Granted you have a proprietary system, but the brass is only used now because it can be machined tolerably easily, but is causing problems due to its softness. So use the carbon steel just mentioned, but fabricate it with a 3D deposited technique and sinter to a solid mass.

 

[Tmoose]

I am leery of equating hardness with wear resistance, most especially if the wear results from erosion.

 

[WKTaylor]

HilaryE...

Differential expansion and contraction can be a killer... especially when a high thermal expansion/contraction alloy is rigidly sandwiched by a significantly lower thermal expansion/contraction alloy.

This situation is analogous to when water-freezes-to-ice [expands] in a rigid container. Forces by the Ice-expansion cause a tremendous compression force on the ice and a tensile stress on the container. In Your case the relatively rigid/low-expansion steel places a high compressive stress on the ductile/high-expansion brass, causing it elastically strain then plastically [permanently] in yield in compression. When the brass/steel subsequently shrink to their smaller static dimensions as the assembly cools off the brass would retain it's now-smaller width-size relative to the steel.

NOTE.
I'll bet that the unrestrained brass dimensions increase about 1/3 of the permanent contractile strain.

Recommend getting a copy of SAE AIR809 Metal Dimensional Change with Temperature to study this phenomena.

Alternately, I am sure there is Finite Element Software that could easily model this differential materials and elastic/plastic strain phenomena.

NOTE.May consider...
(a) adding a gap between the interior of the steel fixture and exterior of the brass 'part'; then adding very stiff high temp wave springs to restrain the brass during it's differential expansion within the rigid steel.
(b) shrinking the size of the brass blocks slightly, so that when it thermally expands within the steel frame, It fits with only a slight [compression] interference when max temperature is reached/stabilized with both materials.

NOTE. The old naval term "it's cold enough to freeze the balls off a brass monkey" originated then the temperature dropped well below freezing on the deck of a wooden ship. The brass cannon-ball carriages ['monkey', mounted to the soft wood deck] contracted at a much higher rate than the tightly fitted/stacked [at normal temperature] iron cannon balls. This could/did cause the pyramid-stack of cannon-ball to become loose from the carriage... possibly allowing cannon-balls to separate from the stack and roll-around the deck [a bad situation on a ship]. The naval-brass carriages had good strength and high resistance to corrosion and were a permanent deck feature; but were very expensive. The cannon balls were obviously disposable in high number and needed to be just tough enough to shoot reliably from a cannon... but be cheap/easy to make... hence good quality cast iron was used.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[Tmoose]

Hi Wil Taylor,

Did you verify that monkey tale ?

Snopes for one doesn't think it happened quite that way.

http://www.snopes.com/language/stories/brass.asp

all in fun

regards,

Dan T

 

[WKTaylor]

Rut-Ro ... looks like I was sucked into the abyss of an old [nautical] fairytale!!

However the thermal expansion contraction differences between various alloy is a major concern for aircraft and spacecraft designers/maintainers.

I DO know-of, and have a similar experience with, copper alloy [CuBe, Al-Ni-Sn-Bronze, etc] plain bearings press-fit into UHS steel [4340M, 300M HT260-to-HT300] landing gear components that accommodate steel pin rotations. What a nightmare of temperature ranges to deal with: -65F +160F 'design'... or possible -100F to +180F 'real-NS-world'. Interferences in bearing housings and bore clearances with rotating pin surfaces and the lubricant groove-designs and the lubricant selections and environmental seals are all super critical to LDG operations... without spinning or migrating the plain bearings.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]