LCB Heat treatment 1

[asal9122]

Hey everyone

We have a problem about heat treatment of LCB which we fail in impact result. First I should say that I have read every single related thread in this site.

We have THREE different chemical composition as :
C: 0.22
Si: 0.26
Mn: 0.81
Cr: 0.16
Ni: 0.35
Mo: 0.08
Cu: 0.05
P: 0.01
S: 0.02
Al: 0.046
Sn:0.009
Ti:0.063
-------------------
C: 0.23
Si: 0.26
Mn: 0.90
Cr: 0.14
Ni: 0.32
Mo: 0.092
Cu: 0.04
P: 0.01
S: 0.016
Al: 0.061
Sn: 0.008
Ti: 0.048
-------------------
C: 0.22
Si: 0.3
Mn: 0.99
Cr: 0.14
Ni: 0.34
Mo: 0.09
Cu: 0.04
P: 0.01
S: 0.014
Al: 0.046
Sn: 0.007
Ti: 0.059
we did these cycle for heat treatment and here is the results: (we need impact result of average 18 J)

1. 940 C - 2h - Air Cool; 898 C- 2h - Water quench - 676 6 h quench up to 300C and then normalize. (Partial similar to what suggested here

http://www.eng-tips.com/viewthread.cfm?qid=363891)

Impact result: 4 J !!!
UTS: 736!! mpa

2. 940 C - 2h - Air Cool; 898 C- 2h - Water quench - 676 4h quench up to 300C and then normalize
Impact result: 5 J !!!
UTS: 643!! mpa


Soak at 650 C 1 hour; 930 C water quench 2 hour; 590-600 2 hour Air cool
impact result : 17 J
UTS: 627

and this is a 16" valve of LCB.

we didn't reach impact result from what suggested in this forum and other threads. Any idea to get impact result of 18 J ? is soaking time appropriate for a 16" valve?
and please be aware that we have almost no chance to change the chemical composition and should focus on heat treatment.

thank you in advance.

 

[bobjustbob]

What is the section size of the test bar you are using?
What is the section size of the valve? You say it is a 16" valve but, depending on the valve type that could mean different things.
What is the purpose of the first normalize (940°C)?
Has your furnace been surveyed?
I suspect that your 2 hours at the austenizing temperature is too short. Even for small valves we would go a minimum of 3hours. For sections above 3" we would start to add time 1/2 to 1 hr per inch depending on the part. However, with the high tensiles that you are getting I don't think this is the root of your problem.

The chemistry looks fine. I normally would want a bit less carbon and more Mn, but you should be able to get the properties with what you are showing.

Your UTS is extremely high. I would not expect you to get good impact with a 106ksi ultimate tensile strength with this chemistry. 106ksi indicates to me that it was tempered closer to 1100°F. That makes me suspect your temper furnace is not operating properly. 1250°F is about correct, we would temper at 1280°F but 30°F is not causing this issue. What was the hardness of your test bar? You should be in the 140-192HBN range.

In my experience, your heat treatment is overly complicated. LCB does not need much as long as you get a good quench. A simple 1650°F with a well agitated water quench followed by a 1280°F temper should do the job assuming your heat treat furnaces are operating correctly.

 

[asal9122]

Thank you for your answer. we surely recheck our furnace tempreature

our test sample is a Y block however, maximum section size in this valve is 11 cm.

first normalize was offered with some people here to maintain a finer grain size.

do you just normalize after temper? Is'n that make a Temper Embrittlement??

For time calculation we use 1 our for first inch and 15 minute for every more. do you suggest to use 30 minute??

 

[arunmrao]

I use 1 hour per inch + 1 hour .

"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[bobjustbob]

I don't know what you mean by "Y" block. Is it the standard ASTM A370 test bar size or are you doing a thicker coupon?

I am not familiar with the Normalize treatment that you use. Normalize has always meant am austenatize followed by and air cool. What you are showing is more of a double temper. I have never experienced temper embrittlement on LCC/LCB material. The standard process was to air cool following temper. I have tried quenching after temper and never saw any noticeable difference in impact properties. I tried performing a normalize before quenching but again did not show significant help with LCC/LCB. Before I left the foundry I attended a SFSA presentation by a professor from Missouri S&T that discussed prior normalizing treatment on high alloy material that showed little to no grain refinement unless a temper was performed between the normalize and the quench. The paper was not finished but it mirrored testing we did on some 4320 material. I never did any prior normalize on LCC/LCB materal, or even higher strength versions, as we never needed it.

One hour per inch is the safest bet. Depending on the paper you read, some recommend 1/2 hour per inch. We would give it 3 hours up to 4" and a half hour per inch after that. Now that is at temperature, we would give an hour for everything to get to temp before counting the cycle time. Even if your thermocouple is showing that you are at temp the mass of material is still coming up. But again, with the ultimate tensile strength that you are reporting I doubt it is a time at temperature issue, especially since you are tempering for 6 hours.

Usually I start by looking at the quench when I see low impact properties. However, that is assuming normal tensile strength. In your situation something in the temper must be wrong in order for you to get the kind of tensile strengths that you are reporting. Even if you had a bad quench, with the chemistry and temperatures shown you should be closer to 80ksi UTS, than 100ksi.

 

[weldstan]

Normalizing after tempering will defeat any metallurgical properties gained by tempering so don't do it. Normalize then temper. Have you checked the grain size of each material in each heat treated condition.
Do not normalize after quench and tempering. Again you will defeat the effects of Q&T.

 

[Dalinus]

Is it even possible to get some good impact energy with so high titanium content? There must be so many that nice pink sharp edged titanium nitrides that miscrostructure must look like kaleidoscope.

 

[asal9122]

@arunmrao : In this case for example for a valve of 36" which have maximum about 8 inch thickness you estimate 8 hours??!!!!
do you use this for temper too?!

@bobjustbob: from Y block I mean a specimen with dimension around 10*25*20 cm
thank you for your explanation. they were very helpful.
I drag your attention to third heat treatment in which we could reach average 17 J.( there was 3 test : 14,16,22). what do you suggest to make it better to 18 J? changing soaking time to 3 h? or is it just result of carbon and titanium content??? any idea?

@weldstan : I dont use normalize after temper!!! first I normalize and then quench- temper and air cool.

@Dalinus: I read in a paper that titanium around 0.05 could result a good grain size. but unfortunately we could not control it perfectly.But again I drag your attention to third heat treatment in which we reach 17J with same composition.

 

[bobjustbob]

What melt practice are you using?
What is your nitrogen content?
Your test block is pretty large, where are you taking the sample from in the test coupon?
What was the hardness of your test coupon and part after temper?

4"x8"x10" is a pretty hefty coupon to through quench well. I would have sacrificed some carbon and added Mn and Mo if I was working with such a large coupon but we are passed that point now.

Again, from the information that you shared, I would first look at your temper furnace temperature uniformity. You need to be down in the 75-80ksi UTS range first, 140-192hbn. Also, with such a large specimen I would austenatize for 4 hours at temperature. I would not normalize prior to quench nor double temper like you are doing. It seems that you are trying to bandaid a more fundamental issue.

 

[weldstan]

asal9122,
Good but that id not what you stated in heat treatment Nos 1 & 2, "quench up to 300C then normalize". I assume that you meant temper at 300C. If so. your tempering temp was much too low and as a result the high tensile properties were achieved but at a sacrifice of toughness. The aluminum content is indicative of fine grain melting but is somewhat high, Quenching from the normalizing temperature as was done in No.3 further refines the grain and the higher tempering temp decreased tensile properties and increased absorbed energy. Why you soaked the part at 650C for 1 hour is unknown but it would reduce some internal stress.

The S content is a bit high. I would recommend that it be kept below 0.010%.

 

[HBMetals]

Our engineer once did this heat treatment experiment on LCB casting specimens and I just find some documentation and try to translate the literature partially: three LCB specimens(A, B, C), chemical composition analysed by spectrum apparatus - ok,

Specimen A: Normalized + Tempered. Normalizing: 940±10°C, holding, air cooled; Tempering: 650±10°C, holding, air cooled.
Specimen B: Quenched + Tempered. Quenching: 910°C±10°C, holding, water cooled; Tempering: 650±10°C, holding, air cooled.
Specimen C: Normalized + Quenched + Tempered. 940±10°C, holding, air cooled; 910°C±10°C, holding, water cooled; 650±10°C, holding, air cooled.

Mechanical properties after heat treatment: T.S, Y.S, Elongation, Reduction of Area, Min. & Ave. Energy Absorption in impact test @-50°C.

Specimen A: 551MPa, 368MPa, 25%, 63%, 34.5J, 42.5J.
Specimen B: 639MPa, 484MPa, 24%, 57%, 39.5J, 42.2J.
Specimen C: 627MPa, 474MPa, 26%, 63%, 67J, 74J.

The microstructure of the specimen before/after tempering are provided in the attachment.

Engineer's conclusion: the N+Q+T is the optimum heat treatment process to be adopted. Normalizing can improve grain size which further results in compact tempered martensite(see the attached microstructure photo). Good strength and toughness can be obtained.

 

[asal9122]

@bobjustbob: Thank you so much. your opinion was really helpful.I think we should increase our time.

@weldstan : my mistake! there is not normalize it should be air cool. thank you.

@HBMetals : That was reaaally helpful!!! we did some like third heat treatment and I wonder why doesn't it worked for us. according to @Bobjustbob I think we should work on our time.

 

[RAJMETALLURGY]

@ HBmetals: can u please share the chemical composition.

 

[asal9122]

@HBMetals : please share us chemical composition and your specimen dimension and holding time

 

[HBMetals]

Sorry for the late reply. I just come back from vacation. One of my fellow engineer says it seems that your chemical composition is inapproriate. As a rule of thumb, we should adhere to "Low-Carbon High-Manganese" principle in the production of LCB casting. An actual chemical composition is listed as: C=0.164%, Si=0.420%, Mn=1.076%, Ni=0.321%. The carbon content of your specimen is too high. The optimum carbon content shall be between 0.14% and 0.18% according to the experiences. Manganese shall be between 1.0% and 1.2% which dramatically improve the hardenability of the steel. Although, A352 prescribes that Mn shall be ≤1.0%. It also states "For each reduction of 0.01 % below the specified maximum carbon content, an increase of 0.04 % manganese above the specified maximum will be permitted up to
a maximum of 1.10 % for LCA, 1.28 % for LCB, and 1.40 % for LCC". Besides, the addition of rare earth ferrosilicon during deoxidation is also important. It is proved that "0.1% addition of rare earth element may increase low-temp toughness by 50%".

Titanium ^ Nickel ^ Alloy Steel