Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / General Welding Discussion / 4140 shaft repair for HS rotating element
- - By engineer1984 (*) Date 12-22-2016 01:39
Hi all,

We have a small gouge in a customer's shaft due to improper seal installation in the field.

The shaft was sent to me to see if it could be repaired.  I gave it a quick glance (I was in a hurry that day) and wrote 'prepare weld area, weld repair (TIG weld + temper bead, etc, etc).  Of course, of all the things I did that day, the one I spent the least amount of time on is the one that now is getting everybody in the world to look at my repair recommendations with a microscope! 

Another engineer brought up the concern of residual stress.  He thinks we need to do a proper PWHT.  We'll hang the shaft vertically if we do a PWHT to minimize any chances of damaging a perfectly new shaft.  (This is common procedure when we do this, but no one likes to do this)

The gouge is about the area of your pinky nail (maybe) and the depth of about .25" or less.  The shaft is 4" in diameter.  Is this residual stress argument overkill?  I'm not thrilled in subjecting a finished shaft (.0005" tolerance on many of the turns) to PWHT if I don't have to. 

Cheers,
Parent - - By 803056 (*****) Date 12-23-2016 02:21
Weld it and check it for straightness and balance. It it flies, all is good. If it is out of balance or doesn't check out when you check the run out, PWHT can be your backup position.

Al
Parent - By engineer1984 (*) Date 12-23-2016 05:08
That was my idea too.
Parent - - By Tyrone (***) Date 12-23-2016 12:07
Hey Al,
If the run out doesn't check out, will PWHT bring it back to alignment?

Tyrone
Parent - - By 803056 (*****) Date 12-23-2016 14:28
That's a good question. The elastic portion of the distortion should be recoverable, any plastic strain would have to be corrected after PWHT. However, considering the amount of weld that is needed (base on the information provided), there should be minimal plastic strain.

Actually, I would not expect much in the way of elastic strain if the amount of weld is as minimal as suggested in the original post. I would be tempted to weld the gouge, rotate the shaft and place a small amount of weld on the opposite side of the shaft to balance the residual forces. The shaft is going to be remachined anyway, so the weld added (to balance the RS) should not have an adverse affect as long as the proper preheat and slow cooling is observed.

Merry Christmas.

Al
Parent - By engineer1984 (*) Date 12-24-2016 03:48
Thanks guys!

I was actually throwing around the idea of welding the other side of the shaft, but was also maybe thinking that was dumb : )

The other engineer thinks that the left over residual stress may show up during operation due to 'vibration stress relieving'. This shaft will be spinning around 10,000 RPM and let's say vibrates around .0005" - .001" per cycle.
- - By 803056 (*****) Date 12-24-2016 15:35 Edited 12-24-2016 18:51
I don't believe it was mentioned what the shaft was made from. I've seen many shafts made out of 41XX and free machining steels. Both add a level of complication; the 41XX are subject to high hardness if not preheated sufficiently and the free machining steels, well lets just say, don't weld them.

The service conditions in this case also add a level of complexity as well. 10 000 rpm requires very close balancing to minimize fatigue. If there is any danger of property loss or injury it may be best to simply replace the shaft with a new one. I encounter this several weeks ago. The small gouge was on the order of 1/15 inch deep and an 1 1/2 inches long. The ramifications of a failure far outweighed the cost of replacing the shaft. Yes, the cost of material and remachining was on the order of $4000, but the cost of in-service failure could be millions. Replacement was the better option.

Merry Christmas.

Al
Parent - - By welderbrent (*****) Date 12-24-2016 17:20
The grade is in the title Al...4140.

Merry Christmas to you and all others here.

He Is In Control, Have a Great Day,  Brent
Parent - - By 803056 (*****) Date 12-24-2016 18:50 Edited 12-25-2016 14:40
I thought I saw it somewhere, but it wasn't in the post, so I figured it was one of the other inquiries. Oops!

Balanced welding (both sides), and considering the potential for high hardness; suggest high preheat, and as a minimum, a tempering operation up around 800 degrees F once all the welding is completed. It isn't high enough to minimize the residual stress, but it will reduce the hardness in the HAZ. Stress relief is recommended with slow cooling.

With the complications of may be easier to make a new shaft depending on the diameter, length, machining requirements etc. If the shaft is forged, it is a cost driver that would push us back to welding.

There is nothing that is cut and dry. One must consider all avenues and factors that influence the final decision.

Merry Christmas you all! 

Al
Parent - - By engineer1984 (*) Date 01-04-2017 22:13
Thank all of you for your thoughts.

The shaft is a forged shaft and (unfortunately) has the coupling hub as part of the shaft.  This makes for an expensive shaft replacement.  It also means that we don't have an inventory shaft forging on hand. 

So far, the repair:

Pre-machine shaft for HVOF repair to see which locations need weld repair. The biggest area (and maybe only area) is the one described above.
NDT
Weld repair as necessary.  (400 F pre heat)
Weld 180 degrees from repair for largest repair
NDT
Runout check
PWHT vertically
NDT
Runout check
HVOF repair to bring up to spec.
NDT / Dimensionals.

Besides making a new shaft, the other option might be:  machine down gouged area and install sleeve.  However, there are some technical problems with this solution.  I'm looking into it.

I wish we could use the gouge 'as is' since it is on a chamfer and not a sealing area, however the o-ring slides up on this chamfer.  I'm concerned with cutting the o-ring. 

Thanks again for all the thoughts and I hope everyone enjoyed their break.

Cheers,

Andy
Parent - - By Tyrone (***) Date 01-05-2017 12:32
Let us know how it turns out!
Tyrone
Parent - - By jwright650 (*****) Date 01-09-2017 12:21
Let us know how it turns out!
Tyrone

...pun intended?....:cool:
Parent - - By Tyrone (***) Date 01-09-2017 12:40
hehe...not intended...
Parent - - By engineer1984 (*) Date 01-09-2017 20:58
Well... I kinda got burned on this.  The continual lesson that we have to learn once a year or so:  Always ask if the QA documentation fully explains the problem.

These pinky tip sized gouges are around the entire shaft.  I was under the impression that there was only one large gouge.  Something wasn't lining up with my communication with this site in Canada (we work in Houston).  I finally asked for more photos of the shaft damage. 

Too bad because we already pushed the weld repair with the customer and the customer even traveled to the Canada site.  No wonder they were pushing back. 

So now we're looking into the sleeve repair option and renigging on the weld repair.

I guess I don't understand why we (engineers) have to constantly ask the QA team for basic information or have to verify data, but I guess that's why we get paid more than the QA team..

Ugh.
Parent - By jwright650 (*****) Date 01-10-2017 12:14
Well, it's hard to make an informed decision (to repair by welding or not to repair by welding) when you didn't have all of the facts.

As an inspector who is reporting on field work, I can see both sides of this situation.
Parent - - By Tyrone (***) Date 01-10-2017 13:35
Communication.....
We do things a little different here.  Whenever QA writes up a non-conformance, a Powerpoint is attached that clearly states the issue, with pictures, the root cause (if known), corrective action and the proposed disposition.  All other functional areas gets to review, put in their 2 cents, approve/reject.

This process seems to work here.  Maybe you could suggest it at your place.

Tyrone
Parent - - By engineer1984 (*) Date 01-11-2017 04:16
Your QA proposes a solution?

That's a neat idea.  I'd be happy if our team would do half of what you described.
Parent - - By Tyrone (***) Date 01-11-2017 12:39
The solution can come from anyone.  Since QA has to write the document,  they ask for everyone's input for a solution.  Everyone pretty much has a heads-up before seeing the document.  Then it's just a formality to accept the solution.

Might have to pay your QA more to take this responsibility.  :wink:

Tyrone
Parent - By engineer1984 (*) Date 01-13-2017 18:00
Haha, if only I was in charge...  I'd solve all the world problems.

: )
Parent - By SHebel Date 02-14-2017 18:35
I was looking at your post and I know my timing is off, but hopefully it's useful; for future reference if nothing else.

I know your co-worker mentioned that some stress may be taken out during usage because of vibrations (and it's definitely possible), but there's no way to know if random vibrations will take all the stress out before the shaft warps, nor if it'll even make it back to installation before it starts distorting. I was curious if you looked into vibrational stress relief equipment at all... There's one by the name of Meta-Lax, which can take out the thermal and residual stresses to prevent distortion, and it can also be used during welding to increase weld penetration and give a smoother, slower cooling weld puddle. Either or both use of it might be helpful for you, but the stress relief in this case probably more so.

The main con I've heard about Meta-Lax is that vibrations don't change the physical properties of the metal, such as harden, soften, straighten, etc., just remove stress. Of course, in the case of a shaft like this it's the biggest pro too, since you wouldn't need to worry about distortion in the process of stress relieving it.

You can check it out at www.meta-lax.com, depending on what other jobs you do it might be worth looking into.

Know it's a bit late, so I hope I'm not beating a dead horse... but since I can't find any real "final resolution" post, maybe I'm not too late, just late. Either way, I'm interested in hearing what finally happened with the repair.

Steve
- - By 803056 (*****) Date 01-10-2017 14:11
One word - photograph.

Al
Parent - - By mcostello (**) Date 01-14-2017 03:59
A photograph is worth 10 points of blood pressure, a year off a life, and about 4 handfulls of hair if any left by then.
Parent - - By 803056 (*****) Date 01-14-2017 05:45
:yell:
Parent - By fschweighardt (***) Date 01-16-2017 23:35
HVOF may not be the ideal method for building up that much material.
Up Topic Welding Industry / General Welding Discussion / 4140 shaft repair for HS rotating element

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill