Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / Technical Discussions / Can you critique the weld quality of these welds?
- - By MrDisibility 2 Date 01-11-2008 23:09
First, please forgive the "noobness" of my questions. I am very new to weld process controls, and am currently awaiting my copies of Ed Craig's materials.

For those unfamiliar, I am working with stainless piping and 308L Si .035" wire on OTC robots. We are running tests in an effort to convert from He tri-mix over to an Ar/CO2 2-3%. These are images of one of our test pieces:

HE on right, Ar on left.

I am at a loss for what WS or Welding Speed actually is, but we changed from 20 to 34 ipm. I want to think this is travel speed. There is a debate here that it might be feed rate...but I thought feed rate was adjusted through amperage. Any insight?

Closer shot of the He weld



Closer shot of the Ar weld



The He weld looks tighter and smoother at the edges than the Ar weld. Can you help explain this? I have my own thoughts, but am not ready to confidently speculate.

Also, both welds have a similar color, which what looks like  flaking or scaling. The color is decent, but I am not crazy about this property. It happens during solidification. It seems to me like the outer shell of the weld solidifies first, and because of contact with the atmosphere there is an oxidation layer that forms. As the molten weld cools it shrinks and this change in surface area breaks the oxidation layer, revealing a shiny weld substrate. Am I off base? How can I minimize this property without resorting to a passivization process?

Lastly, what impact would using O2 have on our processes, versus CO2? The bulk system here uses O2, and it would be far easier to calibrate that to a 2% mix with Ar rather than switching to CO2.

Marcus
Parent - - By Superflux (****) Date 01-12-2008 06:14
Well, here's my speculation.... Looks like you're depositing too much on the cover pass, ie. excess reinforcement which is ( I "assume") causing the weld to glow, exposed to the atmosphere and hence the oxide crusty stuff. If wasting that much time and filler metal isn't a concern, you might try a trailing gas for protection. I would also be concerned about the slow cooling during the austinizing temp range on SS. How many passes/ layers is this? If it is glowing on Root and filler beads, there's bound to be some contamination issues here. Also, is pure argon an option with this process??? The heavier argon ( without any co2)might give a better inert blanket.
Parent - By MrDisibility (*) Date 01-14-2008 22:34
This is only 1 pass. 16g material.
Parent - - By Sean (**) Date 01-14-2008 23:37
Marcus,

With GMAW or FCAW typically amperage is related to your wire feed speed.  The ratio will vary from wire to wire.  Travel speed is the length of weld deposited per unit of time (typically ipm or mm/s).

I agree that it looks like you have excessive reinforcement.  From the pictures it also looks like there could be a misalignment???  Can you take a macrosection of the weld to check?  It would be useful to have some scale for sizes...

As for the discoloration it is a result of oxidation and the heat created by welding.  It looks like you have quite the amount of heat input (essentially a constant x amps x volts / travel speed) into your joint.  What's your gas flow?  As for using O2 instead of CO2 it is possible although I believe it should be less than 2% CO2 (someone please correct me if I'm wrong).  The CO2 breaks down in the arc and the C reacts with your alloys and the O2 also reacts with your deoxidizers and of course your alloying agents. 

What are you rotating the tube or the robot?
Parent - - By MrDisibility (*) Date 01-14-2008 23:48
Sorry for not putting in more detail...

Yes, I am familiar with travel speed. I believe we are running this around 235 in/min. It is the welding speed that I am not familiar with. It is a little confusing. I think we need to contact the manufacturer for clarification on this one.

These are lap joints, so that is probably causing the misalignment issue (as such, it is not really an issue). A slip fit OD/ID pipe. These are 2.5" tubes 409SS.

I believe the flow meter was showing 30 cfh on the Argon scale.

We intended to use 2% CO2 but had 3% on hand for testing. Will 1% make that big of a difference?
Parent - - By Sean (**) Date 01-15-2008 01:37
A couple of observations based on your comments and re-examining the pictures. 

IMHO The joint looks like you are overwelding and looking at your joint it may be that your work angle on the robot may be off too - a cross section macro would confirm these hypotheses.  This may be putting in a lot of heat into the joint.  Parameters may only be a part of the issue.  Try increasing your gas flow to 35 cfh, increasing your travel speed and check your work angle. 

Have you spoken to the wire supplier for support?  They may be able to send in their applications person to help you out.  I know that when I worked for a wire and gas supplier I spent a lot of time helping folks out with these types of projects.  Alternatively, you may want to contact a local welding engineer to help you out.
Parent - - By Superflux (****) Date 01-15-2008 05:07
OK, 2 more cents worth. I think you just plain over welding it! If you can't back off the settings, can you drop down a size on your filler wire diameter? I can't help but think, that with a smaller bead, the scaling/oxide problem would disappear.
Parent - - By Kix (****) Date 01-15-2008 14:13 Edited 01-15-2008 14:18
Yes the one percent does make a slight difference in color.  I tested with the 97/3 and the 98/2 and never got any color with the 98/3.  I did like the arc characteristics and flow of the puddle with the 98/3 better then the 98/2 though.  98/3 is going to be a slightly hotter gas then the 98/2.  What size wire dia are you running?  Maybe a smaller dia wire will give you the arc characteristics you desire and a less convex bead.  Are you going for full penn on these joints?  If you are getting penn through the backside with no purge on it then those contaminants will come to the surface and give you a case of the nasty on bead appearance.
Parent - - By MrDisibility (*) Date 01-15-2008 21:13
We are using .035 wire now. I was hoping to go up to .045 and speed up the travel if that is possible.

I assume your references to 98/3 are supposed to read 97/3...right?
Parent - By Kix (****) Date 01-16-2008 13:01
Yes I meant to say 97/3.  ZCAT has a very good option for you with the downhill progression technique.  This will give you a good looking bead and maybe even some color with the 98/2.  I've noticed we allways get color on our downhill fillets with the 98/2 (argon/co2).
Parent - - By MrDisibility (*) Date 01-15-2008 21:14
Hey guys, we are weaving these too. To help get coverage over inconsistent cuts as they may come up.
Parent - - By ZCat (***) Date 01-16-2008 04:51
If you go up to .045 you are gonna need more heat and it's probably gonna get even blacker. You might try turning the gas up some and fiddle with your welding head placement. Is the gun directly on top?
I would try it in a downhill type configuration, if the pipe is turning away from you, have the gun a little way back towards you from top dead center. That will cut down on the heat some. 
Parent - By MrDisibility (*) Date 01-16-2008 18:18
Thanks guys, we are going to try cranking up the speed and gas a little!

We start at the top and move down the circumference, as the pipe rotates. The part is robotically welded.

Marcus
Parent - - By MrDisibility (*) Date 01-16-2008 18:28 Edited 01-16-2008 18:32
Regarding the need for more heat with larger wire, I found an article on thefabricator.com: http://www.thefabricator.com/Consumables/Consumables_Article.cfm?ID=903

Productivity Increase With Shielding Gas, /Wire Change. A major automotive parts manufacturer evaluated several metal-cored wire/shielding gas combinations to produce a critical flexible coupling for automotive exhaust systems. The required circumferential weld joined a mixture of 409, 304, 321, and INCONEL® alloy base materials. The major objective was to reduce defects to improve part acceptance rates and welding speeds. Cost reduction was essential with improved quality.

A 0.045-inch-diameter 18 chromium/columbium metal-cored wire used with several argon/CO2 blends was compared to the currently used 0.035-in.-diameter 308 low carbon, high silicon solid wire with argon/oxygen shielding. Current production rates were measured at three units per minute. With the metal-cored wire-argon/CO2 combination, the production rate increased to more than five units per minute.

Metal-cored wire, with its higher deposition rate for the same current level, provided excellent wetting characteristics and a broad arc shape, resulting in a flat bead shape with acceptable penetration. Little or no postweld machining of overweld was required to meet the original equipment manufacturer's (OEM's) specifications.

Now, they don't say that the overall appearance improved, but that it did meet OEM specs. This is the route I was thinking of taking once we get the gas change figured out.

Marcus
Parent - - By Sean (**) Date 01-16-2008 19:22
Once you finishing your existing project (i.e. changing parameters, shielding gas, technique, etc.) you may want to explore material preparation before you look at changing wire sizes.  More consistent joints & fitups will reduce your variation and allow you to travel even faster with your existing setup.  Also you may already have the best wire size for your circumstances.  Note that they say, "Little or no postweld machining of overweld was required to meet the OEM specs...".  They may have larger deposition rates with a bigger wire, but if they are overwelding more than they were before are they really saving anything?  It just says that the welds meet OEM specs...
Parent - - By MrDisibility (*) Date 01-17-2008 19:13
Thanks Sean. This project is indeed intended to standardize our existing processes in order to isolate material fitment issues that come up. Right now we have three variables. Programming, Fixturing, and Material fitup. We are tackling hiccups with them all at once.

Thanks for the help everyone!
Parent - - By aevald (*****) Date 01-17-2008 19:53
Hello MrDisibility, I have mainly been reading and observing on this one, but I can make a couple of comments. I have observed a fair number of robotic applications in different fabrication and machining operations. First and foremost, don't make multiples of changes at the same time. You might not be able to readily determine which changes caused which results. When you have thousands or hundreds of thousands of dollars invested in a robotic system don't go low-ball on the fixturing and carefully consider it's design and function.
     Case in point: I was visiting a fabricator in the Bellingham WA. area, he had a very state-of-the-art robotic welding cell and had built some in-house fixturing to accomodate their particular application. He ended up having some repeatability issues, coupled with some accuracy problems as well. Short story: fixtures weren't constructed heavily enough, they would allow for deflections as the part was manipulated, the fixtures were also heating up and expanding, contracting, and bowing, causing positioning and accuracy to vary. Consider whether fixture cooling should be a requirement of your application.
     If your systems are older, consider the mechanical condition and stability of the equipment. Lash and play as a result of wear, lack of proper adjustment where take-up adjustments can be made, can all contribute to problems with performance from your equipment.
     You may already be well aware of the few items I have mentioned here so forgive me if this has already been carefully considered. Just remember to formulate a punchlist of some form or fashion to be sure that you have covered all of your bases and considered everything, even the small and inconsequential types of items. Good luck and best regards, aevald
Parent - By MrDisibility (*) Date 01-18-2008 17:25
Your advice is accurate and appreciated!

Our changes were made one at a time among 4 different variables. By carefully making changes and recording the results we were able to come up with our processes easily. Although it did take more time than we would have liked.

Indeed our fixtures are heavy and strong, and some here feel that the parts may be constrained a little too much. Although most of them were designed for manual welding processes so we had to implement a fixture review process in order to adapt them to robot capabilities.

Calibration is an item that struck me on Ed's site as well. We are currently building an SOP manual to help address this, and other issues concerning accuracy of our settings.
Up Topic Welding Industry / Technical Discussions / Can you critique the weld quality of these welds?

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill