What causes hardness to fall off rapidly? I have a heat treated hot rolled steel part that failed in service which was examined and revealed a hardness drop of 10pts within 1/2 inch of the failed zone. A metallurgist examined the grain structure and concluded fine grain throughout. Does this mean the heat treat process is not to blame?
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Sharp Hardness Variation
- Thread starter driveline
- Start date
Not sure, but several fabrication processes can lead to a sudden dropoff in material properties.
A hot bending process using an induction heating machine together with a water quench on the adjacent unbent pipe section can lead to a transition zone that may have been heated above the first critical temp . The correct post bending procedure would be to N+T the entire piece, but if they skipped that final step, then the crystal structure of the transition zone may be incorrect and a premature failure is to be expected. A micro hardness traverse may show where the incorrect crystal structure regions are.
A similar problem occurs if an incorrect interpass temp is used during welding. If the interpass temp is too high (perhaps due to using too high a welding amperage so as to allow thicker weld electrodes and fewer passses) then the intercritical region oof teh HAZ would become overtempered and have incorrect crystal structure over a region greater than one wall thickness.
Another common error is to assume equal furance temperatures throughout a heat treatment furnace. There can be high temperature gradients in a furnace, and some parts fo teh piece might not have been heat treated correctly.
A hot bending process using an induction heating machine together with a water quench on the adjacent unbent pipe section can lead to a transition zone that may have been heated above the first critical temp . The correct post bending procedure would be to N+T the entire piece, but if they skipped that final step, then the crystal structure of the transition zone may be incorrect and a premature failure is to be expected. A micro hardness traverse may show where the incorrect crystal structure regions are.
A similar problem occurs if an incorrect interpass temp is used during welding. If the interpass temp is too high (perhaps due to using too high a welding amperage so as to allow thicker weld electrodes and fewer passses) then the intercritical region oof teh HAZ would become overtempered and have incorrect crystal structure over a region greater than one wall thickness.
Another common error is to assume equal furance temperatures throughout a heat treatment furnace. There can be high temperature gradients in a furnace, and some parts fo teh piece might not have been heat treated correctly.
More information needed,
1)How was the part heat treated? If it was was surface hardened it would make a difference.
2)10 points reduction of hardness on what scale? 10 points Vickers 500g Micro-hardness no big deal , 10 points Rockwell C big issue.
3)What material?
1)How was the part heat treated? If it was was surface hardened it would make a difference.
2)10 points reduction of hardness on what scale? 10 points Vickers 500g Micro-hardness no big deal , 10 points Rockwell C big issue.
3)What material?
- Thread starter
- #4
davefitz,
How can the furnace process be monitored to prevent this? Our QC guy has been pushing the use of cpk for some time.
Carburize,
(1) This part was not surface hardened. I believe they use an austemper process, they claim it's proprietary as the parts are commonly 1/2 thick or greater. (2) Oh yeah, forgot to mention the variation is on the Rockwell C scale! (3)Spring steel
How can the furnace process be monitored to prevent this? Our QC guy has been pushing the use of cpk for some time.
Carburize,
(1) This part was not surface hardened. I believe they use an austemper process, they claim it's proprietary as the parts are commonly 1/2 thick or greater. (2) Oh yeah, forgot to mention the variation is on the Rockwell C scale! (3)Spring steel
Before I would go blaming the heat treater, I would have a detailed metallurgical analysis performed of the spring. You indicated some examination was performed - what specifically was the cause of failure (fatigue, corrosion, etc)? You mention a fine grain structure - what exactly was the microstructure?
- Thread starter
- #6
metengr,
The ductile failure was due to a sudden overload which would probably have failed the part anyway, I know this because a second part failed that was made to spec. The metallurgist said the grain structure was fine bainite throughout. You say "detailed metallurgical analysis" could you be more specific? I have had chemical, hardness, microhardness and charpy performed.
The ductile failure was due to a sudden overload which would probably have failed the part anyway, I know this because a second part failed that was made to spec. The metallurgist said the grain structure was fine bainite throughout. You say "detailed metallurgical analysis" could you be more specific? I have had chemical, hardness, microhardness and charpy performed.
driveline;
You have covered the "essentials" in terms of a metallurgical analysis. Assuming the chemistry is within specification, the heat treatment appears correct (based on your statement that the part contained a uniform bainitic structure), a variation in hardness was observed. Going back to your original post - is this a change in microhardness?
I would double check the microhardness readings and make sure you have performed the proper conversions to HRc scale. As carburize noted, a variation in microhardness is to be expected.
You have covered the "essentials" in terms of a metallurgical analysis. Assuming the chemistry is within specification, the heat treatment appears correct (based on your statement that the part contained a uniform bainitic structure), a variation in hardness was observed. Going back to your original post - is this a change in microhardness?
I would double check the microhardness readings and make sure you have performed the proper conversions to HRc scale. As carburize noted, a variation in microhardness is to be expected.
- Thread starter
- #8
metengr,
I double checked the microhardness readings by asking for a direct Rockwell reading. I know that they can be suspect, case in point I had two labs and our in-house QC check this and all three were different. But the average was off by 10pts. One lab actually measured 16pts low. Can alloy segregation do something like this?
davefitz,
I failed to mention that the area is near a bend in the part.
I double checked the microhardness readings by asking for a direct Rockwell reading. I know that they can be suspect, case in point I had two labs and our in-house QC check this and all three were different. But the average was off by 10pts. One lab actually measured 16pts low. Can alloy segregation do something like this?
davefitz,
I failed to mention that the area is near a bend in the part.
If it is adjacent to a bend, then the lack of a post bending N+T may be the culprit, as explained above.
Regarding furnace improvements, I understand that the furnace temperaure gradient problem had been solved in tehpast by supply of a hot gas recirculation fan. It should also be solved I I think) by increasing the time spent in the furnace, as the high radiation heat transfer coefficint at 1900 F should equalize the metal temperature over time.
Regarding furnace improvements, I understand that the furnace temperaure gradient problem had been solved in tehpast by supply of a hot gas recirculation fan. It should also be solved I I think) by increasing the time spent in the furnace, as the high radiation heat transfer coefficint at 1900 F should equalize the metal temperature over time.
driveline,
As stanweld confirmed, alloy segregation can influence hardness gradients. There is an excellent article from Corus about distortion that shows an example of segregation and it's effect on hardness. Use the following link for more information:
As stanweld confirmed, alloy segregation can influence hardness gradients. There is an excellent article from Corus about distortion that shows an example of segregation and it's effect on hardness. Use the following link for more information:
Driveline, there could a number of different reasons why you observed a hardness variation in your component. The best way to determine the root cause is to follow the advice given above regarding a thorough metallurgical analysis. One steel manufacturer that provides such services can be found at
Click on the Tech Desk tab, and you will have access to the metallurgical helpline. Information that needs to be provided for anyone to properly analyze your problem are the following:
1.) Material grade (more specific than "spring steel")
2.) Heat treatment details including austenitizing time and temperature as well as furnace type used, quenching medium, tempering temperature and number of tempers, etc.
3.) Part geometry and dimensions
4.) Application and mode of failure
5.) Were any finishing operations performed on the component after the heat treatment process was completed?
6.) Method of hardness testing
The grade of steel that you are using, the thickness of the part, and the method of quenching are all factors that can affect the resulting hardness. The grade of steel is important because some grades cannot be hardened throughout their thickness no matter how rapidly they are quenched, while others may be through hardened in thick sections by simply quenching in still air. The thicknes of the part is important because in order to through harden it the heat must be removed at a sufficiently rapid rate, and this rate is grade dependent. If the rate is too slow (a slack quench) then the part may not meet the hardness requirements. If heat is not drawn from the part in as uniform a manner as possible, then hardness variations can result. If the length of quench time is insufficient, then a relatively large amount of retained austenite can result in hardness variations as well as distortion issues. The furnace type can have a measurable effect on the resulting hardness profile. Salt bath furnaces usually produce the best overall heat teat response. If you are using an austempering process, are you quenching the parts in a salt bath or are you uising a different quenching medium?
The method of hardness testing is also important. Are you testing for Rockwell C directly, or are you using a different hardness testing machine, such as Vickers, and converting those numbers to RC? Is the hardness testing machine properly calibrated, and is the person performing the test properly qualified? I have witnessed examples where a testing machine that was calibrated on a daily basis was used for an extended period of time with a broken diamond indenter. I had to bring it to the attention of the lab supervisor that something was seriously wrong with this machine. The lab techs just continued using it without realizing they were using a bad setup. If you have not done so, you may want to take some hardness readings yourself to confirm the findings that were posted.
If you would like to consider an independent source for comparison purposes, the following link will rpovide you with a commercial source for austempering parts:
Good luck.
Maui
Click on the Tech Desk tab, and you will have access to the metallurgical helpline. Information that needs to be provided for anyone to properly analyze your problem are the following:
1.) Material grade (more specific than "spring steel")
2.) Heat treatment details including austenitizing time and temperature as well as furnace type used, quenching medium, tempering temperature and number of tempers, etc.
3.) Part geometry and dimensions
4.) Application and mode of failure
5.) Were any finishing operations performed on the component after the heat treatment process was completed?
6.) Method of hardness testing
The grade of steel that you are using, the thickness of the part, and the method of quenching are all factors that can affect the resulting hardness. The grade of steel is important because some grades cannot be hardened throughout their thickness no matter how rapidly they are quenched, while others may be through hardened in thick sections by simply quenching in still air. The thicknes of the part is important because in order to through harden it the heat must be removed at a sufficiently rapid rate, and this rate is grade dependent. If the rate is too slow (a slack quench) then the part may not meet the hardness requirements. If heat is not drawn from the part in as uniform a manner as possible, then hardness variations can result. If the length of quench time is insufficient, then a relatively large amount of retained austenite can result in hardness variations as well as distortion issues. The furnace type can have a measurable effect on the resulting hardness profile. Salt bath furnaces usually produce the best overall heat teat response. If you are using an austempering process, are you quenching the parts in a salt bath or are you uising a different quenching medium?
The method of hardness testing is also important. Are you testing for Rockwell C directly, or are you using a different hardness testing machine, such as Vickers, and converting those numbers to RC? Is the hardness testing machine properly calibrated, and is the person performing the test properly qualified? I have witnessed examples where a testing machine that was calibrated on a daily basis was used for an extended period of time with a broken diamond indenter. I had to bring it to the attention of the lab supervisor that something was seriously wrong with this machine. The lab techs just continued using it without realizing they were using a bad setup. If you have not done so, you may want to take some hardness readings yourself to confirm the findings that were posted.
If you would like to consider an independent source for comparison purposes, the following link will rpovide you with a commercial source for austempering parts:
Good luck.
Maui
- Status
Similar threads
- Locked
- Question
