Hardness Limit of P9 heater tubes

Date 11-21-2010 14:07

I've never heard of replacing a tube due to hardness increase during operation, unless this is for H2S service. In that case NACE MR0175 should be consulted. We have T9 tubes in power boiler service that have been running for over 40 years and only replace if there is a tube leak. Usually the tube leaks are due to sootblower erosion or overheating, not high hardness.

By ashfaqanwer (*)

Date 11-21-2010 14:13

Thanks Sims for your reply.
Actually, the hardness we have noted comes out to be 450 BHN which seems quite problematic.

By MBSims

Date 11-21-2010 16:44

450 BHN does seem a bit high. Any idea of what the cause is? Quenching during operation? Is this hardness at the HAZ or away from the weld?

By ashfaqanwer (*)

Date 11-21-2010 16:52

These hardness readings are of the base metal, the actual tube material.
I think this would be due to operational upsets especially during start-up and shutdown of heater.

By js55

Date 11-22-2010 13:38

Marty's right. Boiler tubes don't usually get changed unless they're leaking. If the material is seeing high hardness I would think it wouldn't stay that way once the boiler is fired back up. Marty's also right in that, the only thing I can think of that would cause a hardness like that is quenching. Might help to do photo mics first and see what it looks like. If the material is a chrome moly the microstructure will have to be martensite to be that hard. What do you suppose is gonna happen to the interstitial carbon when that puppy fires back up again. I'm guessin HCP goes to BCC. Just a guess though.

By MBSims

Date 11-22-2010 14:23

A good suggestion, tube samples should be removed and metallurgically examined to determine the microstructure and whether the hardness is a surface effect, and perhaps some tensile and bend specimens to determine if ductility is acceptable. I would not expect a change from HCP to BCC to occur during operation because the lower transformation temperature would need to be exceeded. 9Cr is very hardenable though, and rapid cooling could cause the hardness to increase.

Ashfaq - What kind of heater is this?

By ashfaqanwer (*)

Date 11-22-2010 14:47

It's a fired heater, externally fired.

By Jim Hughes (***)

Date 11-23-2010 19:06 Edited 11-23-2010 19:09

What are you using to fire the heater? Is it an off gas from something? I have seen 1 1/4 and 2 1/4 Cr tubes increase in hardness due to nitriding caused by the process. In this case it was NH3 and C3 that caused the nitriding. Also what hardness measuring tools are you using?

Jim

By js55

Date 11-22-2010 19:32

WOW. I had my head up my butt on the transformation temp. Good point. And your right about surface hardness. Quite common with the oxidation involved. Better to be sure to have a through thickness issue before scrapping.

By G.S.Crisi (****)

Date 11-22-2010 20:42

ashfaq is talking about an oil refinery fired heater, quite different than a boiler. If a leak develops in a boiler tube some water or steam will spill out. If a leak develops in a refinery heater, an oil product will spill out and the whole heater will catch on fire.
My suggestion may seem unusual. If you measured the tube surface hardness is because you shut the heater down, waited until it was cool and someone got into the heater through the manhole to take that measurement.
This is my suggestion: fire the heater if it isn't already in operation, get it up to nearly 600 ºC, maintain the temperature for half an hour and switch the burners down. Close all the points where air could squeeze in, including the burners. Fired heaters use to have a butterfly valve into the stack, so close that valve.
Let the heater cool down naturally, it'll take several hours to happen. When the temperature allows it, have someone get into the heater and take new hardness measurements.
I wouldn't be surprised if the BHN has lowered to acceptable levels.
Sounds an unusual procedure? Marmaduke Surfaceblow used to solve the most complicated engineering problems by means of unusual procedures. And Marmaduke Surfaceblow was the greatest engineer human kind has ever known (after Leonardo da Vinci, of course)
In any case, before scrapping the tubes, as 3.2 recommends, wouldn't it be worthwhile to test this suggestion?
Giovanni S. Crisi
Sao Paulo - Brazil

By 3.2 Inspector (***)

Date 11-23-2010 02:23

Interesting theory, and thats all it is I am afraid - just a theory.

3.2

By js55

Date 11-23-2010 13:57

Maybe just a theory, but replacement without attempting to determine cause is perpetuating a costly ignorance.

By 3.2 Inspector (***)

Date 11-24-2010 01:22

I agree.

3.2

By G.S.Crisi (****)

Date 11-23-2010 17:47

3.2,
theories must be confirmed by practice, and human kind has always proceeded that way since the stone age.
Want an example? The existance of atoms was first proposed by Greek philosopher Democritus centuries BC.
For about two milleniums the theory was forgotten, until English cientist John Dalton revived it. In 1810 he announced his theory and explained the ten properties of atoms, 4 of which are still valid.
Nowadays nobody doubts of the existence of atoms, even if nobody has been able to actually see them.
Giovanni S. Crisi

By ashfaqanwer (*)

Date 11-23-2010 19:57 Edited 11-23-2010 19:59

Thanks for your value addition.
I will see if this can be done.
In any case, can you please suggest what should be the allowable hardness for P9 heater tubes for continuous operation.

Regards,
Ashfaq

By G.S.Crisi (****)

Date 11-26-2010 18:00

I would say that 240 BHN is a good number to start with. By this I don't mean that this number must be achieved at all costs, but is a number from which you fellows can start a discussion in which you'll argue in favor or against it until you reach an agreement.
Giovanni S. Crisi

By ashfaqanwer (*)

Date 11-26-2010 21:00

Thanks Giovanni for your reply.
We will do so.

By Jim Hughes (***)

Date 11-27-2010 13:54

ashfaqanwer,
the main problem is ASME does not help much with hardness values. B31.3 would tell you 241 HB Max for your material. ASTM will say 250 Max for P91 which is a P9 modified, but does not give a value for P9. Since the tensile is 60K then a min. based on converting BH to tensile would be around 131 HB. As mentioned in an earlier post,

1. What are you firing the heater with?
2. What device are you using to check your harness? Is it a MIC 10, or 20? or are you using a mech. device like a Telabrineller, or Pin-Brineller. The reason I bring up the device is if your using a electronic device there are some varables that can cause it to give you erroneous hardness numbers.

Thanks
Jim

By ashfaqanwer (*)

Date 12-02-2010 05:44

PROCEQ Equotip 2 hardness meter which measures the hardness by REBOUND method. Heater is being fired with 90% pure hydrogen gas.

By Jim Hughes (***)

Date 12-02-2010 13:03

ashfaqanwer,
that was my concern when you first posted your question, was that you were using this type of tester. They work ok in the hands of someone that has a good procedure, but there are so many varriables that can cause the instrument to give you wrong information, such as material cleanliness, prep, technique, calabration, and battery life. Those readings seam very high to me, and because of the high readings I would question your instrument and do some follow up with a mechanical tester.

Thanks
Jim

By ashfaqanwer (*)

Date 12-02-2010 13:20

Thanks Jim,
What else do you recommend to do? If the hardness is like in the range of 350-400 BHN, should replacement of tubes be planned?

By Jim Hughes (***)

Date 12-02-2010 20:38 Edited 12-05-2010 13:19

I have seen these numbers fluctuate between 100 and 250 points between a electronic hardness tester and a mechanical tester. So the first thing I would do is re-test using a telabrineller or pin type brineller. The readings you get will be very close to what the hardness is. If the readings are still in the range that you originaly posted, then bringing in a metallurgist would be my next step. Structural Integrity would be my next phone call. You can google them and find contact info. They are one of the leading experts in this type of investigation. That being said if you retest and your numbers are in the 250 range then your ok. (my opinion)

Thanks
Jim

By MBSims

Date 12-02-2010 23:45

Are you sure the results are in HB and not HL? As I recall, the Equotip testers also used a conversion chart to go from HL to HB, so use of the correct conversion chart for the material type should be checked. Also, I recall there is a correction factor for tests performed with the tester oriented horizontal or overhead vs. down flat. All these should be checked. We prefer the MIC-20 tester for reliable field hardness measurements. Our Equotip tester is too inconsistent.

By G.S.Crisi (****)

Date 12-03-2010 00:17

Marty,
I confess my ignorance. The three scales for metal hardness I've always known are Brinell, Vickers and Rockwell. The latter has three sub-scales, namely A, B and C.
Shore is used for elastomers.
What's then HL?
Giovanni S. Crisi

By MBSims

Date 12-03-2010 02:40 Edited 12-03-2010 02:43

HL is the Leeb or rebound hardness test per ASTM A956. More info here:

http://lark.tu-sofia.bg/~ntt/pesu/readings/hardness.pdf

http://en.wikipedia.org/wiki/Leeb_rebound_hardness_test