Why does a weld continue to crack if re-welded ?

Thread Tools
 
Search this Thread
 
Old Feb 27, 2008 | 08:51 PM
  #1  
RCautoworks's Avatar
Thread Starter
Honda-Tech Member
 
Joined: Nov 2002
Posts: 17,227
Likes: 2
From: Illinois
Default Why does a weld continue to crack if re-welded ?

I'm curious, as of late I have seen one manifold that a customer brought in. It was a cc-fab manifold and it had multiply cracks. So I cleaned it up and took care of the manifold, but the manifold came back and the same cracks reappeared. So I'm curious on why something like this happens. Is it because of bad fitment, over heating the metal ( from the original welds ) . And what is the way to fix this issue, cut the weld our totally and lay a brand new weld down ?
Reply
Old Feb 27, 2008 | 09:11 PM
  #2  
AspectIndustries's Avatar
Honda-Tech Member
 
Joined: Sep 2003
Posts: 3,166
Likes: 0
From: Boost is good
Default Re: Why does a weld continue to crack if re-welded ? (RCautoworks)

What normally works for me is.

Use a Grinder/Cutting wheel and "rebevel" where its cracked (cut down). and Re weld as its a brand new mocked manifold? Works everytime for me? Welding over a crack doesn't solve anything.

Dun lie, your own manifold is cracking! I WAS JUST KIDDING! guys...
Reply
Old Feb 27, 2008 | 09:20 PM
  #3  
RCautoworks's Avatar
Thread Starter
Honda-Tech Member
 
Joined: Nov 2002
Posts: 17,227
Likes: 2
From: Illinois
Default Re: Why does a weld continue to crack if re-welded ? (AspectIndustries)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by AspectIndustries &raquo;</TD></TR><TR><TD CLASS="quote">
Dun lie, your own manifold is cracking! I WAS JUST KIDDING! guys... </TD></TR></TABLE>

LoL , sadly enough its ccfab and he is no where to be found to help out his older customers

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by AspectIndustries &raquo;</TD></TR><TR><TD CLASS="quote"> Welding over a crack doesn't solve anything.</TD></TR></TABLE>

Why though ?
Reply
Old Feb 27, 2008 | 10:06 PM
  #4  
k24em2's Avatar
Honda-Tech Member
 
Joined: Apr 2005
Posts: 1,895
Likes: 0
From: Lincoln, Ne, USA
Default Re: Why does a weld continue to crack if re-welded ? (RCautoworks)

Do you drill the ends of the crack
Reply
Old Feb 27, 2008 | 11:13 PM
  #5  
RC000E's Avatar
Honda-Tech Member
 
Joined: Jul 2004
Posts: 5,463
Likes: 2
From: I'm everywhere Focker
Default Re: Why does a weld continue to crack if re-welded ? (k24em2)

Yeah, don't forget that part. The ends of the Cessna's wings that I used to fly had holes galore...lol. I actually welded an aluminum softball bat once. I drilled and welded, and it didn't recrack. I went to do that to the second bat and once I welded I heard this crazy hissing noise...come to find out aluminum bats have some inflated balloon inside...crazy!
Reply
Old Feb 28, 2008 | 01:44 AM
  #6  
weiRtech's Avatar
OG Fabricator
 
Joined: Feb 2005
Posts: 1,864
Likes: 1
From: Burlington, Ont., Canada
Default

i'd do the same as aspect. you have to treat it like a new weld and grind out/bevel what is there so you are getting some penetration into the base metal.
Reply
Old Feb 28, 2008 | 01:53 AM
  #7  
RS2 FAB's Avatar
 
Joined: Jan 2008
Posts: 515
Likes: 0
Default Re: (weiRtech)

it's becuase the base weld is cracked/no good

all other welds/passes over it would be compromised...........in pipe welding if the orignal e6010 or 5p weld is compromised....... in 1 spot they are required to completley remove, clean and once again bevel the area...........maybe that's what you need to do...remove it completly, clean the **** out of it bevel and go at it again
Reply
Old Feb 28, 2008 | 07:15 AM
  #8  
coll9947's Avatar
 
Joined: Mar 2004
Posts: 34
Likes: 0
From: Boise, ID, USA
Default Re: Why does a weld continue to crack if re-welded ? (RCautoworks)

Contamination. Lack of proper pre- and post-heating.

Drilling the ends of the crack and completely cleaning out the weld area and re-beveling is the way to go, as suggested by the other posters.

Mild steel melts at 2800F or so, but between room temp and melting point the metal changes its crystalline structure several times. The most common structure is Martensite, which has a huge range of about 1300F to 1800F, and is a very brittle crystal structure within the steel. (Body-centered tetragonal fyi) Making good welds isn't just about hand control and knowing how to set up the machine, it's also about heat control and input. Controlling heat is the real challenge in welding, and is almost always the cause of failures, because the crystalline structures that make up steel and its alloys become compromised by the heat.

Pre-heating and post-heating for different alloys to specific temperatures with specific cooling rates is crucial to prevent the formation of Martensite within the base and weld metal.

If you have ever done a weld, quenched it in water right away, then broken the weld in half you might notice the "shiny" crystal-like appearance of the weld surfaces. That is Martensite rearing its ugly head.

In certain situations where hardness is desired over ductility, Martensite is ok, but in steels with a higher alloy content, martensite can cause the alloy metals to become seperated from the steel molecules and congregate in clumps, thereby making the alloy basically worthless in that crucial heat affected zone around the weld, where you need the metal to be strongest. Remember, most welds made by competent certified welders don't fail, the base metal fails in the heat affected zone because the alloy is broken down by the heat and the crystalline structure is not uniform.

This is especially common in chrome moly where people weld it without post heating it. All the chromium and molybdenum that makes chrome moly so strong (and expensive) gets seperated from the steel, like your chocolate milk powder slowly sinking to the bottom of your glass if you don't drink it right away. When put under stress, chrome moly structures that aren't post-heated to VERY high temperatures can appear totally safe and solid, but in fact have under-bead cracking that will only make itself known after catastrophic failure.

Then you factor in the warping due to the extreme post-heating and suddenly chrome-moly isn't that appealing. For reference, most 7000hp top fuel cars only use chrome-moly on the back half of the chassis. For 95% of applications it is used for, it's overkill. I know this post isn't about chrome-moly, but it's a great example of how important heat is to welding.

Every alloy is different in how you use pre- or post-heating to affect the formation and shift of crystalline structures.

Martensite = hard but brittle, prone to heat-related failure.
Reply
Old Feb 28, 2008 | 08:39 AM
  #9  
manifoldmiketyson's Avatar
Honda-Tech Member
 
Joined: Feb 2004
Posts: 438
Likes: 0
From: Tampa, FL
Default Re: Why does a weld continue to crack if re-welded ? (coll9947)

probably the carbon (from exhaust) on the inside of the piece contaminating the welding causing a "higher carbon" weld making it more brittle.

and the stuff mentioned above

it will only get worse the more times to try to weld.

i was reading up last night about 309 rod and its uses and found that the proper way to weld ss to carbon steel is weld a buffer on the carbon steel with 309 and then machine the suface. after the surface is machine weld the ss to the carbon steel in the buffer zone. this allows minimal transfer of the carbon steel into the ss which takes away from the strength of the welded joint.

also, most times pre-heating stainless can be harmful instead of helpful.
Reply
Old Feb 28, 2008 | 08:58 AM
  #10  
cptengineer's Avatar
Honda-Tech Member
 
Joined: Sep 2005
Posts: 380
Likes: 0
From: chino, ca, us
Default Re: Why does a weld continue to crack if re-welded ? (RCautoworks)

this usually occurs due to the re-crystalization of the material upon welding, on cooling, new grain sizes and grain boundries occur. Often the crack growth is promoted to to flaws in the material structure and when subjected to repeated tension and compresion cycles, failure will often occur at the newer gain boundry caused by the heat effected zone of your weld. I am not a welder- I would love too, this is just from a materials engineering view point.
Reply
Old Feb 29, 2008 | 10:39 AM
  #11  
ManBearPig4silly's Avatar
Honda-Tech Member
 
Joined: Oct 2004
Posts: 1,972
Likes: 0
From: sheridan, wyoming, usa
Default

the cracks usually extend beyond what you can see on the surface. If it's the weld from the flange to the runner your best bet is to cut the runners past the HAZ and put in a splice piece of fresh SS. The mild steel flange is much more resistant to being welded multiple times. Other than that there isnt much you can do that will last.
Reply
Old Feb 29, 2008 | 11:29 AM
  #12  
.RTErnie's Avatar
Honda-Tech Member
 
Joined: May 2002
Posts: 3,599
Likes: 0
From: BeLlInGhAm, Wa
Default Re: (ManBearPig4silly)

Its from weld porosity. There isn't full penetration so there is air trapped between his welds... when the manifold comes up to temp the air within in the weld expands and propagates a crack. The manifold will continue to crack until the end of time unless the whole weld is ground out. Trust me... I know

In one of Carrol Smith's "x to win" books he shows a side cut of a crack and shows how the material around the affected zone was also cracked. So its more than likely that you arent "welding" the whole cracked region.
Reply
Old Feb 29, 2008 | 03:42 PM
  #13  
G2turbo_terror's Avatar
Honda-Tech Member
 
Joined: Jan 2002
Posts: 638
Likes: 0
From: Minneapolis, MN, usa
Default Re: Why does a weld continue to crack if re-welded ? (manifoldmiketyson)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by manifoldmiketyson &raquo;</TD></TR><TR><TD CLASS="quote">
i was reading up last night about 309 rod and its uses and found that the proper way to weld ss to carbon steel is weld a buffer on the carbon steel with 309 and then machine the suface. after the surface is machine weld the ss to the carbon steel in the buffer zone. this allows minimal transfer of the carbon steel into the ss which takes away from the strength of the welded joint.

also, most times pre-heating stainless can be harmful instead of helpful.</TD></TR></TABLE>

Where did you read this? Very interesting.
Reply
Old Mar 2, 2008 | 09:24 AM
  #14  
manifoldmiketyson's Avatar
Honda-Tech Member
 
Joined: Feb 2004
Posts: 438
Likes: 0
From: Tampa, FL
Default Re: Why does a weld continue to crack if re-welded ? (G2turbo_terror)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by G2turbo_terror &raquo;</TD></TR><TR><TD CLASS="quote">

Where did you read this? Very interesting.</TD></TR></TABLE>

http://www.ssina.com/view_a_file/weldingbook.pdf

page 11. the term is "buttering"
Reply
Old Mar 2, 2008 | 10:46 AM
  #15  
rorik's Avatar
Honda-Tech Member
 
Joined: Dec 2003
Posts: 920
Likes: 0
From: bumfuck egypt, SD, US
Default Re: Why does a weld continue to crack if re-welded ? (manifoldmiketyson)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by manifoldmiketyson &raquo;</TD></TR><TR><TD CLASS="quote">

http://www.ssina.com/view_a_file/weldingbook.pdf

page 11. the term is "buttering"</TD></TR></TABLE>

Good link. I read most of it. Does anyone actually anneal their manifolds at 1800-2150 F, and then "rapidly cool" them? Definitely seems beneficial, but I don't have a kiln to do this in..
Reply
Old Mar 2, 2008 | 10:49 AM
  #16  
rorik's Avatar
Honda-Tech Member
 
Joined: Dec 2003
Posts: 920
Likes: 0
From: bumfuck egypt, SD, US
Default Re: Why does a weld continue to crack if re-welded ? (coll9947)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by coll9947 &raquo;</TD></TR><TR><TD CLASS="quote">Contamination. Lack of proper pre- and post-heating.

Drilling the ends of the crack and completely cleaning out the weld area and re-beveling is the way to go, as suggested by the other posters.

Mild steel melts at 2800F or so, but between room temp and melting point the metal changes its crystalline structure several times. The most common structure is Martensite, which has a huge range of about 1300F to 1800F, and is a very brittle crystal structure within the steel. (Body-centered tetragonal fyi) Making good welds isn't just about hand control and knowing how to set up the machine, it's also about heat control and input. Controlling heat is the real challenge in welding, and is almost always the cause of failures, because the crystalline structures that make up steel and its alloys become compromised by the heat.

Pre-heating and post-heating for different alloys to specific temperatures with specific cooling rates is crucial to prevent the formation of Martensite within the base and weld metal.

If you have ever done a weld, quenched it in water right away, then broken the weld in half you might notice the "shiny" crystal-like appearance of the weld surfaces. That is Martensite rearing its ugly head.

In certain situations where hardness is desired over ductility, Martensite is ok, but in steels with a higher alloy content, martensite can cause the alloy metals to become seperated from the steel molecules and congregate in clumps, thereby making the alloy basically worthless in that crucial heat affected zone around the weld, where you need the metal to be strongest. Remember, most welds made by competent certified welders don't fail, the base metal fails in the heat affected zone because the alloy is broken down by the heat and the crystalline structure is not uniform.

This is especially common in chrome moly where people weld it without post heating it. All the chromium and molybdenum that makes chrome moly so strong (and expensive) gets seperated from the steel, like your chocolate milk powder slowly sinking to the bottom of your glass if you don't drink it right away. When put under stress, chrome moly structures that aren't post-heated to VERY high temperatures can appear totally safe and solid, but in fact have under-bead cracking that will only make itself known after catastrophic failure.

Then you factor in the warping due to the extreme post-heating and suddenly chrome-moly isn't that appealing. For reference, most 7000hp top fuel cars only use chrome-moly on the back half of the chassis. For 95% of applications it is used for, it's overkill. I know this post isn't about chrome-moly, but it's a great example of how important heat is to welding.

Every alloy is different in how you use pre- or post-heating to affect the formation and shift of crystalline structures.

Martensite = hard but brittle, prone to heat-related failure.</TD></TR></TABLE>

According to the link posted above, martensite is not formed with austenitic stainless, the type we use for manifolds. "In general, pre heating is not helful becasue no structural changes, such as martensite formation, occur in the weld or the heat affected zones"
Reply
Old Mar 2, 2008 | 11:07 AM
  #17  
KENetics1's Avatar
Honda-Tech Member
iTrader: (3)
 
Joined: Mar 2003
Posts: 1,808
Likes: 0
From: WI, United States
Default Re: Why does a weld continue to crack if re-welded ? (manifoldmiketyson)

I've heard of the term "buttering" before in a welding class, the instructor said it was good for welding on cast material. Forgot if it was cast-iron or cast-steel.
Reply
Old Mar 2, 2008 | 02:17 PM
  #18  
manifoldmiketyson's Avatar
Honda-Tech Member
 
Joined: Feb 2004
Posts: 438
Likes: 0
From: Tampa, FL
Default Re: Why does a weld continue to crack if re-welded ? (KENetics1)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by KENetics1 &raquo;</TD></TR><TR><TD CLASS="quote">I've heard of the term "buttering" before in a welding class, the instructor said it was good for welding on cast material. Forgot if it was cast-iron or cast-steel.</TD></TR></TABLE>

i could see it working best on cast because of all the contamination in most cast pieces. carbon steel tube, plate, sheet and bar is typically a lot cleaner.
Reply
Old Mar 2, 2008 | 03:36 PM
  #19  
damnraz's Avatar
Honda-Tech Member
 
Joined: Jun 2004
Posts: 426
Likes: 0
From: Kent, WA, USA
Default Re: Why does a weld continue to crack if re-welded ? (RCautoworks)

I don't know if this would apply to stainless but on other types of material you can get rid of those oxygen bubbles by fusing over it again slowly. You'd be doing it right if you see the sparks while going over the weld. in Aerospace mfg we X-ray our welds and this method seems the way to go to get rid of porosity.
Reply
Old Mar 4, 2008 | 11:12 AM
  #20  
G2turbo_terror's Avatar
Honda-Tech Member
 
Joined: Jan 2002
Posts: 638
Likes: 0
From: Minneapolis, MN, usa
Default Re: Why does a weld continue to crack if re-welded ? (manifoldmiketyson)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by manifoldmiketyson &raquo;</TD></TR><TR><TD CLASS="quote">

http://www.ssina.com/view_a_file/weldingbook.pdf

page 11. the term is "buttering"</TD></TR></TABLE>

Good read, thank you.

Now, how important would it be to "machine" the surface of the buttered weld VS. grinding/personal judgment. I'm wondering if a pipe welded (both sides) to a buttered head flange could crack if there was air trapped in a uneven cavity.

I'm prolly splitting hairs but I'm just thinking this through.
Reply
Old Mar 4, 2008 | 04:47 PM
  #21  
manifoldmiketyson's Avatar
Honda-Tech Member
 
Joined: Feb 2004
Posts: 438
Likes: 0
From: Tampa, FL
Default Re: Why does a weld continue to crack if re-welded ? (G2turbo_terror)

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by G2turbo_terror &raquo;</TD></TR><TR><TD CLASS="quote">

Good read, thank you.

Now, how important would it be to "machine" the surface of the buttered weld VS. grinding/personal judgment. I'm wondering if a pipe welded (both sides) to a buttered head flange could crack if there was air trapped in a uneven cavity.

I'm prolly splitting hairs but I'm just thinking this through. </TD></TR></TABLE>

if your getting pockets, your going to know during welding. the machining aspect of the process is to bring the tolerances back to the original spec. if your were not going to do this, you would have to accomodate for a bead thickness below the pipe or whatever you were planning on welding.

not to mention, your buttered surface may be twice the width of the finish weld. without machining it might look retarded.
Reply
Old Mar 4, 2008 | 07:11 PM
  #22  
G2turbo_terror's Avatar
Honda-Tech Member
 
Joined: Jan 2002
Posts: 638
Likes: 0
From: Minneapolis, MN, usa
Default Re: Why does a weld continue to crack if re-welded ? (manifoldmiketyson)

Okay. Thanks for clearing that up for me.

I'm making myself a new manifold this summer so I know what I'm gonna try.
Reply
Old Mar 7, 2008 | 10:08 AM
  #23  
JCushing's Avatar
Honda-Tech Member
 
Joined: Nov 2001
Posts: 3,138
Likes: 0
From: Suck it Trebek
Default Re: Why does a weld continue to crack if re-welded ? (RCautoworks)

[QUOTE=RCautoworks]

LoL , sadly enough its ccfab and he is no where to be found to help out his older customers
[QUOTE]

no way he offered a lifetime warranty remember
Reply
Old Mar 7, 2008 | 10:12 AM
  #24  
RCautoworks's Avatar
Thread Starter
Honda-Tech Member
 
Joined: Nov 2002
Posts: 17,227
Likes: 2
From: Illinois
Default Re: Why does a weld continue to crack if re-welded ? (JCushing)

[QUOTE=JCushing][QUOTE=RCautoworks]

LoL , sadly enough its ccfab and he is no where to be found to help out his older customers
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">

no way he offered a lifetime warranty remember</TD></TR></TABLE>

LoL he is no where to be found when his customers needed him.
Reply
Old Mar 8, 2008 | 10:52 AM
  #25  
TurboSI56's Avatar
Honda-Tech Member
 
Joined: Jun 2006
Posts: 224
Likes: 0
Default Re: Why does a weld continue to crack if re-welded ? (RCautoworks)

I know this is off-topic but I've seen welds crack on log style manifolds between the 2-3 runner on the inside of the joint (between the two tee's facing the head flange) and have found that cutting the flange between the two cylinders helped and eliminated the cracking issue.
Reply
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
Karsten
Honda Accord & Crosstour (2003 - 2012)
0
Jul 19, 2018 12:06 PM
tallblondkid
Welding / Fabrication
8
Apr 17, 2013 02:39 AM
tallblondkid
Forced Induction
1
Apr 14, 2013 02:45 PM
94VTEC2TONE
Classifieds: Forced Induction
2
Jan 14, 2010 10:50 AM
project dc2
Forced Induction
36
Mar 29, 2004 10:05 AM




All times are GMT -8. The time now is 05:43 PM.