Brass Annealing Properties 1

[RontotheB]

Hello,

I have a supply of half-hard CDA-260 cartridge brass that underwent bright hydrogen annealment (@800°F for 1:06 w/ nitrogen cool). This annealed brass as well as its half-hard cousin were both stamped into 0.006" thick, 0.56" discs, which were later burst under pressures above 2500 psi.

The annealed alloy burst at a higher pressure than the half-hard alloy. I read that annealing generally softens metals, so why would the softer metal burst at a greater pressure? Did the increased brittleness of the half-hard alloy hinder its performance?

I was also looking into a annealment spec similar to the bright hydrogen anneal. I don't care much about cosmetic properties, I just want an anneal that will provide a brass of the same strength as the other anneal.

 

[kenvlach]

"Did the increased brittleness of the half-hard alloy hinder its performance?"
Yes. From

http://www.MatWeb.com,

some values of elongation before fracture (tensile test data, but correlates to bending) and ultimate tensile strength for
Cartridge Brass, UNS C26000 (260 Brass):
Elongation at Break UTS
OS100 Temper (annealed) 68% 43500 psi
H01 Temper (1/4 hard) 43% 53700 psi
H02 Temper (1/2 hard) 23% 61600 psi
H04 Temper (Hard) 8% 76100 psi

I've included H01 since your annealed material was worked hardened by punching, and likewise, the 1/2-Hard was further hardened. Seems elongation decreases faster than UTS increases when work hardening.

If you want to get into annealing copper and brass, please be aware that the prior thermal & mechanical history needs to be known. Recrystallization and annealing occur faster in more heavily cold worked material, and excessive grain growth can occur from too much time at temperature. ASTM B601 gives multiple sets of annealing tempers for copper & its alloys, some based upon prior mechanical processing, some based upon grain size. E.g., OS005 = annealed, with final average grain size 0.005 mm. OS200 = annealed, with grain size 0.200 mm. O81 is an 'Annealed to' temper = Hard material annealed down to 1/4 Hard.
ASTM B601-07 Standard Classification for Temper Designations for Copper and Copper Alloys—Wrought and Cast

http://www.astm.org/cgi-bin/SoftCart.exe/DATABASE.CART/REDLINE_PAGES/B601.htm?L+mystore+eulz4392

For further information, see ASM Handbook Volume 2 ...Nonferrous Metals... or ASM Handbook Volume 4 Heat Treating. or Heat Treater's Guide: Practices and Procedures for Nonferrous Alloys at

http://www.asminternational.org/Content/NavigationMenu/Bookstore/ASM_Handbooks/ASM_Handbooks.htm

 

[EdStainless]

good response Ken,
My concern about the failure is the grain size. You need to do some basic work. You need to know tensile properties before and after anneal, and grain size also.
If the anneal was done right then it may have the finer grain size.
Look at failed samples under the microscope. Look for where the cracks started (surface defects?) and look for the amount of ductile deformation.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[RontotheB]

Thanks for the info Ken.

As far as the amount of work hardening due to punching, I don't think it'll be much of a factor, because punching will work harden the outer area of the disc. The burst pressure, however, relates to the disc's integrity closer towards the center. At least, that's what I was told by the company doing the punching. Could they be mistaken?

 

[EdStainless]

Is there a crease or notch in these discs to control where they fail? This is common in rupture discs. The harder material may not like being scored.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[RontotheB]

No sir, there is no crease or notch. The only work done to the metal after annealment is the actual stamping.

 

[EdStainless]

Do these actually rupture in the center? Often without a notch of some sort they will start to rupture at an edge or where they are restrained. This makes the exact condition at that point a critical factor which cannot be controlled.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[RontotheB]

Ed, you're right in that they are ruptured where they are restrained. Usually this process produces circular rings where the pressure "stamps out" an inner circular area.

 

[btrueblood]

Agree with Ed - if you don't control the edge, or use a scoring pattern in the center of the disk, it will not consistently break at the center. Instead, it breaks at the edge, and at very inconsistent pressures due to the variety of "failure modes" at work (degree of fixity/slip at the edges, stress concentration due to bending over a sharp corner...)

Are you manufacturing burst disks?

 

[RontotheB]

Yes sir, I am.

Actually, in some of my tests, when I hand-cut brass with titanium scissors (this sounds absolutely ridiculous, I know), I obtained impressively consistent results, all within 25 psi, and with a standard deviation of 15 psi.

Machine stamped parts had a standard deviation of nearly 60 psi.

 

[btrueblood]

No, that doesn't sound ridiculous. You probably did a much better job of controlling the edge burr with your method than a punch can do. The burr results in trouble with edge fixity (clamping).

FWIW, the neatest idea I ever saw for a precision burst disk involved using a snap-over-center Belleville washer that pierced the disk at a very repeatable/reliable pressure (these are for low-pressure disks that control pressures for vacuum/pneumatic conveying systems that load rail cars). The rocket factory I worked for made our own disks, and we had nothing but trouble at lower pressures, the material can be very inconsistent due to the metallurgical issues that Ed points out (properties of material become much more sensitive to things like grain size when the material thickness heads towards zero; and you need thinner disks for lower pressures). We had a stamping die built to scribe the centers of our disks in a cross-shape pattern, and had a series of weights that were used to load the stamping die, to try and control the depth of the lines. It helped, but I seem to recall a precision of less than 10% below something like 300 or 500 psi.