New welder and have some questions

By welder_guy2001 (***)

Date 11-18-2002 05:11

i'm not sure that metal contracts more than it expands. but i do know that after you weld something, the bead contracts w/ the length of the weld as well as w/ the width of the weld. and the weld bead fuses w/ the base metal starting w/ the toes and progressing toward the interior of the bead. since the toes fuse first and the rest of the bead is still shrinking, that pulls on the base metal and causes warpage.

yes, i agree that auto frames are probably mild steel...but i'd check into it before doing anything to one. they probably aren't heat treated, but some of the heavier frames might be an alloy of some kind...i don't know.

By CHGuilford (****)

Date 11-18-2002 14:58

Dirk, the metal doesn't really contract more than the original expansion. What happens is during welding the metal expands and occupies more volume than it did in the cool state, but is restrained by the surrounding cooler metal. That means the hot metal has to go someplace.
Remember that hot metal is of lower strength than the cool base metal. The cooler metal elastically deforms somewhat so it returns to its original state. The hot metal plastically deforms meaning it displaces towards zones of low/no stress and stays there. In other words, the hot metal expands towards the ends of the weld and the exterior surfaces and changes its dimensions. Same volume of metal, just put in different places. Now when the metal cools down, it occupies the same original volume as before but the dimensions have slightly changed.
The most common illustration I can think of is if you could heat up a bar in a vise, without heating the vise, you would find the length and height would increase but the width would decrease. Still the same volume though, if all else is equal.
The actual dynamics that occur in welding are a good bit more complicated. The principal is still the same, only now you have to consider weld volume, total heat input, base metal strength and size, filler metal strength, external restraint, and other things like that.

As far as automotive frames, I don't know what steel types are used, so I can't give you a good answer. I will say that if preheat isn't needed, it won't hurt either. And I would use low hydrogen electrodes if possible. The most important thing I like to remember is what will happen if it comes apart on the highway? Perhaps, others might have more info for you.

CHGuilford

By jwright650 (*****)

Date 11-18-2002 16:13

Generally with mild steel, thickness will dictate the amount of preheat needed to keep the base metal from acting as a heat sink after welding. Car frames are normally not thick enough to worry about preheat. Be careful of the joint geometry and design it for the loads that could be intruduced on the highway. I don't want to be passing you in the other lane and have something go wrong with that welded joint. Full penatration joints should be inspected very carefully by someone qualified to inspect spliced frames.
Be careful,
John Wright

By Dirk O Date 11-19-2002 03:42

So you are saying that the molecules move around to a no stress spot and stay there.So if i measure along a weldment with calipers i will get a measurement saying that certain spot is thicker than the rest of the metal? So when it cools down do the molecules return or do they stay there?
We weld alot of 304 stainless at work and i noticed that shrinks a quite more than mild steel when welded , and i am guessing this is the alloys , how come stainless seems to contract more than mild steel.
Thanks,
Dirk

By RonG

Date 11-19-2002 13:10

Think of dropping a marble in a pool of still water.

The energy moves out ward from the point of impact or heat and as it travels the energy is spread in all directions and dissapated.

If you could freeze every thing (remove the source of heat) and take measurements you would most likely find a reduced area where the marble inpacted with the missing material evenly distributed outward. Its all still there, just looks different.

By CHGuilford (****)

Date 11-19-2002 15:15

That's right, a measurement (in the right places) will show the metal is thicker than before. However the increase is small so I don't think an ordinary caliper will register the difference.
Some of the molecules will remain in their new location. As you know, when all the molecules are hot, they want to spread out. Some get "squeezed" out of shape by outside restraint, like squeezing a handful of modeling clay. But on cooling, the molecules pack together so less volume is required than while hot (but the same volume as before). So some of the molecules return to their original position while some will be in a new position, all according to how the restraining forces "molded" the metal.
To build on Ron's analogy, picture a puddle of mud and drop a rock in it. After the crater is formed, some of the mud will run back into the hole, while some of the mud will stay around the crater. You have the same volume as before, just in a different shape.
Stainless steel has a higher coefficient of thermal expansion that mild steel does. I am thinking about 3 times more but I can't put my hands on my references right now so don't quote that number without checking first. But anyway since, stainless expands more, it will "squeeze" out of shape more. So if mild steel will distort under restraint and heating, stainless will do the same only to a greater degree.
CHGuilford

By jwright650 (*****)

Date 11-19-2002 17:05

Coefficients of expansion for 100 degrees F
Mild steel = .00065
Stainless steel = .00099
Aluminum = .00128

Example from AISC ASD 9th edition:
A piece of mild steel is exactly 40ft long at 60 degrees F. Find the length at 90 degrees F., assuming the ends are free to move.
Change in length = .00065 x 30 degrees difference x 40ft. divided by 100 = .0078 ft. of change in length at 90 degrees F.
Try plugging in the numbers given above for stainless steel and aluminum and notice the difference from mild steel.
Restrain the ends and the unit stress really builds up.
John Wright

By CHGuilford (****)

Date 11-19-2002 17:38

Thanks, JW!
CHGuilford

By Dirk O Date 11-19-2002 23:23

So if i clamp tubing to a table to keep it from warping during welding , i will cause more stress in the weldment?
You never stop learning , so explain.
I was wondering , if i go to college for Mechanical engineering, will the degree teach me all of this stuff?
Thanks
Dirk

By pipewelder_1999 (****)

Date 11-20-2002 03:17

Yes. In general,no distortion means residual stress.

Gerald Austin

By RonG

Date 11-20-2002 12:50

Yes! and you will also learn about thngs to undo the stress like PWHT.

By pipewelder_1999 (****)

Date 11-20-2002 19:29

And also techniques like the half bead temper technique and bead placement that reduces residual stress and distortion.

G Austin

By billvanderhoof (****)

Date 11-20-2002 04:54

Another thread on this topic here
http://aws.org/cgi-bin/mwf/topic_show.pl?id=2531#8218

By jwright650 (*****)

Date 11-20-2002 12:13

Dirk,
If it can't move (restrained by clamps), Yes, it will have stress building as you add heat. ie. remember the old pressure cooker your grandma used when you were a kid? Seal up that hole where the jiggler sits and the pressure will continue to rise until something gives(green beans all over the ceiling).
John Wright

By Dirk O Date 11-20-2002 16:23

So i do i keep my weldments from having distortion if i can't restrain, even good tacks can streach to breakage. or will there be less distortion if i tack it with clamps , unclamp and then weld it?
Are there any good books that can help me learn the stresses of welding , and metals and metallurgy that don't require phd to read?
Also how is post weld heat treatment done, and what dose it do, iam guessing it anneals the metalm but i'm not sure.
Dirk

By jwright650 (*****)

Date 11-20-2002 19:11

Dirk,
D1.1 :2002 has a para 5.8 that deals with stress relief. Para. 5.26.2 talks about straightening after welding has distorted the member. Be careful because the temps. given are for mild steels. Other codes have similar procedures.
Also "Flame Straightening Technology" by John P. Stewart and another good read is "Distortion Control" by the same guy.
Note that it is OK to brace the work, or to build fixtures to help minimize distortion, because welding will pull the steel. But, how much stress relieving is done should be in your contract for the work you are doing. Sometimes there is no stress relieving at all. But, the point "We"(others that have commented on this) is, stress will build up if restrained.

By Dirk O Date 11-21-2002 00:40

I cannot find either book by John Stewart . Is the AWS book Residual Stress and Distortion a good book to have on hand? I work in a general fab/sheet metal shop , so we don't have a lot of structural welding or code books lying around.
I know welding steel will pull it , that what the intent of my question was to find out if steel when welded, will contract more than it expands.I am guessing not, since if I weld one side of a t joint it will pull to the side i welded , and won't pull back straight.
Thanks,
Dirk

By jwright650 (*****)

Date 11-21-2002 12:44

Dirk,
The books by John P. Stewart were given to me several years ago, so I'm not sure where they came from. There is an address:
Copywrite 1981 by John P. Stewart, 233 Dupras Ave., LaSalle, Quebec, Canada, H8R 3S4 Tel:(514)366-8410 Fax:(514)366-6358
Tenth Printing Jan. 1996
If you can't find these books, don't sweat it, there are lots of other good books on the subject. Several were listed in other posts. I'm just partial to what I'm familiar with. The guy has written the book in language that I can understand and it just makes sense. Nothing can make up for actual experience in straightening distorted pieces. After you have had to wrestle with a piece to get it to move with heat in the direction you wanted you'll learn what to do and not do. Remember to use your temp. sticks to keep a check on your heat to keep from over heating your piece. For mild steel I use 1100F temp. sticks to help monitor the heat(at 1200F the steel will cripple). Around the place you need to add heat, take the temp. stick and draw a circle around the spot you plan to heat. When the marks you made with the temp. stick melts, you have reached that temp., stop heating and move to another spot. Practice makes better(not perfect because steel is not perfect).
Get out there and practice! If you only get weld on one side, you will have to heat the other side to get it back straight.
John Wright

By pipewelder_1999 (****)

Date 11-20-2002 19:31

Welding Handbook Volume 1 8th/9th edition.

Weldability of Steels by R.D. Stout

and various Lincoln Electric publications by Omer Blodgett

Have a nice day and I hope this helps.

Gerald Austin