Charpy testing CD4MCuN castings 3

[foundry1]

I am looking for the causes of low Charpy test results for ASTM A995 gr 1B (CD4MCuN) castings. All other mechanical & metallurgical results meets requirements but the Charpy results are 50% below the acceptance criteria. I know the problem is in heat treat (SA + WQ) but I am not exactly sure where.

Any help would be greatly appreciated.

foundry1

 

[bruv]

What temperature are you testing at, and what results are you expecting/achieving?

The first thing to look at would be your SA time and temperature, test block section thickness, and transfer time from the H/T oven to the quench. Is your quench tank big enough for the amount of metal being put in it? Is your charpy test temperature correct and not too low?

 

[foundry1]

We are looking for a 3-sample result with a minimum average of of 45J and minimum individual of 36J.
Temperature is -10C.


There were 4 heats sampled. The following chart displays the mechanical results along with the average of each charpy tests.


HRB YIELD TEN. ELONG RA AVG CHPY
REQ'D >28 70 100 16 36-45
HEAT #
BCBA 24 71.4 111 37 55.6 136
BDBA 27 72.6 111 31 49.4 30
BEBA 23 70.3 113 34 50.6 27.7
CCBA 23 72.4 112 32 50.7 24.3

I do not have all the answers to your questions but intend to find that out.

Thank you for your response. I look forward to any further input you might have.

foundry1

 

[metengr]

foundry1;
I would strongly suggest you or have a met lab perform metallographic examination of the suspect CVN impact test failures and evaluate their microstructure. This can be easily done on one of the fractured CVN impact pieces, and use the other piece for SEM evaluation (if rquired). You may find your smoking gun.

 

[GRoberts]

I strongly 2nd metengr's comments. Try the screening test from A923. If you have bad microstructure for heat treatment, that will help show you.

 

[foundry1]

GRoberts,
Thanks for your response. I've discovered that three of these heats had slightly elevated carbons, max C was .047. What bearing could this have on these Charpy results or the microstructure?

Foundry1

 

[metengr]

foundry1;
Not much effect for this difference (0.040% versus 0.047% by mass). What is the nitrogen content, and more important did you have the lab analysis done as recommended above?

 

[EdStainless]

on duplex grades always run the A923 metallography before you waste time doing impacts. I would suggest the method C corrosion test as well, though the test temperature may be a bit of a problem since the grade isn't listed. Start with 20C and see what results you get. You should be able to compare one heat to another this way.
I know a group at ASTM that would really like any data that you generate on this.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[metengr]

foundry1;
Here is a tip, visit the web page below, scroll to the bottom of the page and download the document titled "Practical Guidelines for the Fabrication of Duplex Stainless Steels." It is an excellent technical reference document if you will be dealing with duplex stainless steels on a regular basis, and is free.

http://www.imoa.info/moly_uses/moly_grade_stainless_steels/duplex_stainless_steel.html

 

[foundry1]

Thanks for your responses. The material specification is A995 Grade 1B. Nitrogen content was from .23 to .25% The lab work was done by a Bodycote lab in California. The metallurgy results for the four heats read as follows:

"Ferite-point count conforms, Exhibits 45.27%* ferrite.
Etch/Mag: ASTM #80/100,200,500X"
"Microstructure conforms, Free from intermetallic phases and precipitates. Etch/Mag: ASTM #80/100,200,500x"

*These ferrite percentages range from 45.27 to 46.77% for the four heats.

My customer has had to reject these castings based on the Charpy results. My concern is that if I remake them I will follow exactly the same procedures and practices that produced these rejected castings. I think the lab made a mistake in the Charpy tests but have no way to prove it and no way to guarantee that they will not do it again.

Any Recommendations?

Foundry1

 

[metengr]

Do you have enough material to re-run CVN impacts using another met lab?

 

[foundry1]

NO

 

[arunmrao]

foundry 1,Interesting problem and many valuable suggestions. I have no experience with this alloy chemistry ,but as a general practice I have some comments to make.

It is sad that you do not have additional test piece to offer. The only recourse is now to produce a heat (hope it is economical and sensible) undr exactly similar conditions and have some additional test bars processed.

It is not recommended to draw samples from castings as they may not be sound.

I always make a separate test check and then offer test bars to the customer. this is expensive,but you are confident and can foresee,if there are any issues involved before making an offer

You mentioned that all tests conform to the standards,but the impact results are suspect . Next time arrange a witness test and also involve the customer in this program. This will strengthen his confidence in your casting and any minor deviations can be condoned.

It is my experience that carbon plays an important role in imact test values. 0.07% is a big margin,I would be more comfortable with lower carbon.

Can you also check the route that you have issued for making the charge. Are there any issues of tramp elements showing up?

I have not failed. I've just found 10,000 ways that won't work." — Thomas Edison
_____________________________________

 

[bruv]

Do you have any spare castings that you can remove from the production cycle to cut up and use as test pieces? As things stand at the moment, the product is unacceptable to the customer. Using actual castings for the test should give you a better indication of the properties of the castings themselves. Also, if they do pass, a) the result will tell you that there may be something wrong with the processing of your test blocks and b) you have only lost a few castings instead of an entire heat.

 

[foundry1]

My customer is going to buy the castings but cannot use them as upgrades for their current order. I am going to duplicate this order to meet their upgrade requirement but pour plenty of test blocks with each heat so this doesn't happen again. Any suggestions on a good reliable ISO qualified lab to do back-up testing?

Foundry1