|
|
Hey guys, I know this is probably this last thing a lot of you would expect, but i dont have to high of self esteem or think that i know every thing (lol) to ask for some help. Im currently the weld engineer at M.I.G. in lexington TN, and Im trying to get some other opinions or tips from people who know welding. Here is the situation i am in. I would say that 90% of our weld defects that gets rejected at our customers are burnthroughs and porosity. At our facility we mainly do GMAW robotic welding with, ABB, Motoman, Panasonic, Fronius, and Fanuc robots. We use Impact Engineering Arc Agents which i set limits for every weld in the plant. If you are not familar with an Arc Agent, it motors the welds current readings and voltage readings. It also detects missings welds and gas flow. It can detect some porosity and burnthrough, but its not one of its high points. It is a really good product as long as you keep limits in range, and not wide open. Im looking for a product or ways to detect porosity and burn through. I am currently working with Impact Engineering to come up with a system where lasers scan the weld to detect porosity and burn through, but from what ive seen the laser is going to be to big to fit into the tight spaces that we weld in. I know i could probably put the laser in a check station after the machine has welded the parts to scan every weld, but with as many machines as we run that will be extremely expensive to make happen, and i do not think management would go for it because of this. Do any of you guys know any tips, or any product that does well at picking these defects up? I have personally made this my goal to try to detect and catch these defects, and I am open to try any process, or idea you have. Thanks Guys for your help.BTW, I didnt come here just to ask for help and never be heard from again. Ive been a long time reader of this site, and it has really helped with out with some problems. Today i finally took the plunge of becoming a member of the site.
Reply:Hire skilled certified welders and ditch the robots! I hate being bi-polar it's awsomeMy Heroes Have Always Been Cowboys
Reply:Talk to a Lincoln Electric rep about their production monitoring software. It is intended to be used with their power supplies, but can use continuously measured arc characteristics to effectively detect defects, rather than relying on visual confirmation.
Reply:A little more detail would help. What mrterial? What filler? What joint configuration?Where are you located?Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:Originally Posted by weldermikeHire skilled certified welders and ditch the robots!
Reply:Originally Posted by weldermikeHire skilled certified welders and ditch the robots!
Reply:Since you are using .035 wire, have you considered using Hobart MetalloyVantage .035 diameter metal cored wire? Wetting and low spatter are said to be less, with less slag inclusions, and burn through(I would think on thin sections). It is manufacturer recommended for robotics applications, and has parameters for your 90/10 mix.http://www.hobartbrothers.com/upload...loyVantage.pdfThere used to be a J.W. Harris product designed specifically for sheet metal called 20 gauge perfect circle. It was a .030 diameter wire and could be used on 20 gauge steel and not burn through which is why I thought of this Hobart wire.http://www.thefabricator.com/article...tal-cored-wirehttp://www.thefabricator.com/article...elding-costs--Last edited by shovelon; 12-03-2012 at 08:21 PM.Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:Originally Posted by mattadams6552lol. Robots show up for work everyday. This is a small town we really dont have a big population, especially welders. I agree people are better than robots, but robots are repeteable, produce more welds, and save money with the volume of welds the plant has.Supe, The Arc agent monitors the weld for: weld duration. gas cfh, current, voltage, and it does a lot more, but that is just what we look for.Shovelon, we are located in lexington tennessee. We have over 50 weld cells with multiple joints. Mainly T joints, but we do have some butt welds. We use ESAB .035 wire (not sure of the exact type, i can get that information tomorrow) with .030 tips. That may sound weird, but it works great. Gas is 90 Ar / 10 Co2. We usually do tips changes every hour to hour and 15minutes of arc time. The metal we weld is usually 1/16 thick, some parts are 1/8 or a little thicker Its just regular steel.All welds are SUPPOSE to 100% visually inspect by the operators, but everynow and then we still get bad parts out of the door. Im trying my best to reduce the chances of this happening..
Reply:There is not much responsablity placed on the operators. they pretty much let them get away with murder with a slap on the wrist. Then thats when i have to get involved and make the machines idiot proof. I may be wrong, but i dont think the wire is causing the burnthrough. We have quiet a bit of part variation ( we get held to a really high standard by our customers, but we dont make our venders do the same for us), and ive noticed when our tip change screen comes up on the robot sometimes the operators clear it out without changing the tip. I cant really do anything about that unless i see it with my own eyes so im pretty much screwed. Maybe one day things will sorta get more strict around here, but it pisses me off because im the one getting hammered when things like this get out the door. A few years ago it used to be completely different. The operators got better training, and operators got worse punishment for letting bad parts get out the door. We do not have hardley any trouble with spatter, we have some of the best robot technicians out there. When our welds are optimized spatter is not a problem. Thanks for the suggestions so far.Last edited by mattadams6552; 12-03-2012 at 09:40 PM.
Reply:I forgot to mention, I know what causes the porosity, and the burnthrough. Im just seeing if there is any system out there that might catch the defects that is better that what i already have. I should have mentioned this in my first post, and i just realized it reading the responses.
Reply:It sounds as if your trying to close the barn door after the cow has gotten out. Since you know what causes the porosity and burn through ... why don't you apply your resources to eliminating or minimizing porosity and burn through then you wouldn't have to put so much emphasis on detecting these defects?HoboLincoln SA200's... at least 15 - 20. They come and go. Growing partial to the "Short Hoods" in my old age. Last count on Short Hoods was 13 in possession.
Reply:Porosity and burn through rarely happens. Its not a problem that happens alot. 9 times out of 10 its because the nozzle is dirty. The other times its air leaks on the fixture that causes it. Burnthrough usually happens when the trim edge is out of spec, or when the tip isnt changed at its limit. The tips get a groove wore in it and it causes the welds to be lower on the material than normal.We usually run anywhere from 1000 to 2000 parts per day on 90% of all the jobs we have. I dont know the exact number, but we have only had 10 - 15 weld related rejections that came back on us. I dont know if i made it sound like it happens all the time, but it is more than normal for us this year. We usually run 5 - 10 weld rejections per year. Im just trying to find something out there that might catch this. I dont want anyone on here to think that im rejecting what you are suggesting. I just think you might not understand my post. Our process has proven to work, but i just want to improve on what we have. No weld engineer at MIG has ever found a sure fire way to catch porosity and burn through and i want to be the one that does through laser imaging or another process.
Reply:As a former manufacturing engineer, it seems like rather than inventing technology that does not exist, that it might be better to focus on really making people change the tips. After all, it is better to prevent defects that detect them.Perhaps you could make it easier to change the tip than to cheat the system. Create a machine that they put the old tip into, which logs the time that the tip was put in. Keep the old tips in a clear plastic tube so it is easy to see if the operator is cheating by putting new tips into the old tip slot. Or make the operator hand in their old tips at the end of the shift. Have them put them in an envelope that they sign. If you open the envelope, and there is only one tip, or brand new tips, you could go after that person.Obviously you would have to have some management support to make this happen. Tell them that the alternative is to give you a million dollars for a laser quality system. ;-)RichardSculptures in copper and other metalshttp://www.fergusonsculpture.comSyncrowave 200 Millermatic 211Readywelder spoolgunHypertherm 600 plasma cutterThermal Arc GMS300 Victor OA torchHomemade Blacksmith propane forge
Reply:I see your position now.I have never heard of a porosity or burn thru hole detector. Don't know if there is such a thing. I agree with HoboHilton and Rafergusen. Perhaps a better system of tip replacement, like a tray with new tips replaced with old ones and time logged. Surely the color would be self evident.I would be curious if you find a flaw detection system. Perhaps someone will chime in yet.Neat stuff you guys make there by the way.Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:Originally Posted by mattadams6552There is not much responsablity placed on the operators. they pretty much let them get away with murder with a slap on the wrist. Then thats when i have to get involved and make the machines idiot proof.
Reply:Thanks guys. I was sorta worried i would piss you all off by trying to explain my situation so much. If you guys havent ever used a weld monitoring system its really neat and it does pick up a lot of defects. Ill try to load some screen shots on here tomorrow. I didnt think there was anything out there, i have been searching the web high and low for almost a month. It is wierd though, im helping a company and doing research to come up with a laser scanner to detect these problems, but i cannot use them on hardley any of our machines because the laser is simply to big to fit in the tight spaces we have to weld in. Did any of you guys go to the fab tech show last month in vegas? I was suppose to do trials with the laser and go to the show with the impact engineering crew there and help represent the product because i was doing the real life testing. Things didnt work out though, we were to busy and things got held up on there end also. i feel like it will be a good product if you have the room though.Back on topic lol. Im liking everyone suggestions. I agree we are having a problem with supervisors being friends with operators, or just putting to much trust into the operators. Our systems are beginning to become relaxed, and we are going to have to do something about that. I will come up with something and when i do i will let yall know what i did.Shovelon, thanks for the compliment. I think we do really good for a smaller factory out in the middle of now where. We get a lot of awards every year for our quality and its helped us stay busy when other people were slow. Im trying to stay ahead of the technology to do everything i can to help keep our quality to what or better than we are expected.
Reply:Originally Posted by mattadams6552Did any of you guys go to the fab tech show last month in vegas? I was suppose to do trials with the laser and go to the show with the impact engineering crew there and help represent the product because i was doing the real life testing. Things didnt work out though, we were to busy and things got held up on there end also. i feel like it will be a good product if you have the room though.Shovelon, thanks for the compliment. I think we do really good for a smaller factory out in the middle of now where. We get a lot of awards every year for our quality and its helped us stay busy when other people were slow. Im trying to stay ahead of the technology to do everything i can to help keep our quality to what or better than we are expected.
Reply:We are not doing any aluminum robotic welding yet. We have been bidding on a job for a few years in mexico that is aluminum. Id love to get that brought up here, it would be a really good learning experience for us.
Reply:This problem isn't even an engineering problem, it's an organizational problem. Document the problem and send it up the chain of command, let them decide if they want to address and correct it."The trouble with quotes on the Internet is that you can never know if they are genuine." - Abraham Lincoln
Reply:Originally Posted by JanErikThis problem isn't even an engineering problem, it's an organizational problem. Document the problem and send it up the chain of command, let them decide if they want to address and correct it.
Reply:Thanks Guys.
Reply:mattadams6552,You've described a management problem not a welding QA problem, if a company is run by people who assign one man a job without the authority to fix that problem? well its pretty obvious there's a management problem not so much with welding.As to hardware- not software- is there an automatic cup brush? I'm thinking of a hollow brush that would allow the arm to push the gas cup into/onto a brush that allowed the tip to fit into a hollow, but the outer cylindrical sides of to clean the inside of the cup or spatter? If the machine load cycle allowed time, this subroutine could be added to 3x , 5x , 10x cycles so the gas cup is power brushed and so is the tip every so-many passes?The same might be said for a tip change fixture? I'm suggestion something like a tool change on NC routers one to clean the cup the other to change tips? Maybe both in one fixture?I'm not familiar with robotic welding but NC routers and mills have similar feature to what I'm describing, adapting some of those tool change fixtures to a welding cell looks like it would go along way toward your QA goals?Also, if you can't influence operators (?) you may be able to install software lock-outs? After so many cycles or weld starts- the arm/cell HAS to go to tip change or gas cup brushing? Cheers,Kevin Morin
Reply:This is such an excellent thread. Lots to think about. Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:Originally Posted by Kevin Morinmattadams6552,You've described a management problem not a welding QA problem, if a company is run by people who assign one man a job without the authority to fix that problem? well its pretty obvious there's a management problem not so much with welding.As to hardware- not software- is there an automatic cup brush? I'm thinking of a hollow brush that would allow the arm to push the gas cup into/onto a brush that allowed the tip to fit into a hollow, but the outer cylindrical sides of to clean the inside of the cup or spatter? If the machine load cycle allowed time, this subroutine could be added to 3x , 5x , 10x cycles so the gas cup is power brushed and so is the tip every so-many passes?The same might be said for a tip change fixture? I'm suggestion something like a tool change on NC routers one to clean the cup the other to change tips? Maybe both in one fixture?I'm not familiar with robotic welding but NC routers and mills have similar feature to what I'm describing, adapting some of those tool change fixtures to a welding cell looks like it would go along way toward your QA goals?Also, if you can't influence operators (?) you may be able to install software lock-outs? After so many cycles or weld starts- the arm/cell HAS to go to tip change or gas cup brushing? Cheers,Kevin Morin
Reply:Maybe you need to tell management they're steppin' over dollars to pick up dimes. MikeOl' Stonebreaker "Experience is the name everyone gives to their mistakes"Hobart G-213 portableMiller 175 migMiller thunderbolt ac/dc stick Victor O/A setupMakita chop sawI have. I made the arguement that if we had an air leek on a fixture, the extra 10 to 15 CFH (35-40 CFH total) would help to reduce the chance of porosity getting out the door. Their response was, we need to get the air leeks fixed lol. I have the manufacturing engineers backing me up, but my plant manager is the one who resists it. Maintenance really helps me out with the diffusors though. If they see a diffuser is getting wore, even though it will gauge, they usually destroy it. and replace it with a new one.
Reply:Originally Posted by mattadams6552Also the cleaning station is programed into the robot, so it goes to the cleaning station after so many cycles, but there are still times when the operator needs may need to clean the nozzle. We have been on a money saving kick here latley and this has prompted some of our problems i honestly believe. Management decided to let our diffusors be run longer, made the diffusor gauges easier for a bad or almost wore out diffusor be used longer. I have been in charge of lowering the gas CFH at the plant. We run every robot at 25-30 CFH. This makes any air leak a major problem. I have fought the lowering of the gas flow, but it didnt go anywhere. Its it really easy to produce porosity.
Reply:I see you are using ESAB wire, If the spool is not bright and shinny copper color but a dull copper color it will feed eratic and give problems.This is not knocking ESAB as i use their wire but get a spool like that now and then.
Reply:Lars66. I will keep an eye on the color of the wire in the future. We get the barrels of wire.Shovelon. I have never used that hose before. Actually ive never heard of it until now. We just use regular gas hose you can get at the hardware store. I make sure not to use air hose though lol. I have done a study on the build up of gas in the line you are talking about. I have always used the least amount of gas line that i have to use. I am going to research the hose, and visit the link you provided. Thanks alot for the help!!
Reply:Originally Posted by mattadams6552Lars66. I will keep an eye on the color of the wire in the future. We get the barrels of wire.Shovelon. I have never used that hose before. Actually ive never heard of it until now. We just use regular gas hose you can get at the hardware store. I make sure not to use air hose though lol. I have done a study on the build up of gas in the line you are talking about. I have always used the least amount of gas line that i have to use. I am going to research the hose, and visit the link you provided. Thanks alot for the help!!
Reply:Shovelon, i got a question for you. how long of hose do you normally use? Im just wondering because on all our machines now with our standard hose we use 10ft at the most from the regulator to the welder. the hose passes through a flow meter that monitors gas flow through out the weld. Im not seeing the spike in gas at the start of the welds. Ive been extremely busy the last two days at work, and i didnt have time to get a picture of our weld monitor, but it shows the gas flow on it. Im just wondering if we can benefit from this product.
Reply:It reminds me of the stories told to me by welders who worked the shipyards during WW2. Performance was measured by the amount of rod the welder consumed. It is easy to figure out that people would take rod from the tool room then throw it off the dock. On paper they were really doing a good job. The next response was to measure the length of weld each man put down. That was easily solved by stealing 1/4 inch steel rods from the machine shop, dropping them down in the joint and welding over. The real solution was proper inspection. Even today it is more of a people problem than an engineering problem. I am wondering if some sort of inductive pickup following along behind could sense porosity?
Reply:Originally Posted by mattadams6552Shovelon, i got a question for you. how long of hose do you normally use? Im just wondering because on all our machines now with our standard hose we use 10ft at the most from the regulator to the welder. the hose passes through a flow meter that monitors gas flow through out the weld. Im not seeing the spike in gas at the start of the welds. Ive been extremely busy the last two days at work, and i didnt have time to get a picture of our weld monitor, but it shows the gas flow on it. Im just wondering if we can benefit from this product.
Reply:I sent the gas saver link to my maintenance supervisor. They have already used the gas saver lines back about 5 to 8 years ago is what he said. He said they did work, but we found a way to keep the spike down with out it. Thats why im not seeing the spike now with our gas flow sensors.
Reply:Originally Posted by mattadams6552I sent the gas saver link to my maintenance supervisor. They have already used the gas saver lines back about 5 to 8 years ago is what he said. He said they did work, but we found a way to keep the spike down with out it. Thats why im not seeing the spike now with our gas flow sensors.
Reply:I'm a little confused, You stated your are the "Weld Engineer" and that It would be the" last thing we expect" you asking questions. You also stated you have no formal College education, so are you claiming to be a Welding Engineer or are you using a generic overrated term for set up man. You also countered skilled workers suggestion with Robotics are repeatable , sounds like they repeat their mistakes to me. LOL. Porosity, burn through, the usual suspects, to high a heat, travel speed, gas washing away by drafts, to much , to little gas, contaminents in the system or gas itself? Just a shot in the dark but on your robotic systems I am sure you go through a ton of wire, are you using any kind of lubricators on the wire for the liners, can cause problems sometimes? some guys like them, some say they are a contaminant.Last edited by kolot; 12-05-2012 at 05:21 PM."Si Vis Pacem Para Bellum"Lincoln Idealarc 250 AC/DCMillermatic 251 Syncrowave 300 30A spoolgunLincoln MP210Hypertherm 45(2) LN 25(2) Lincoln Weldanpower 225 CV(4) SA200 1 short hood SA250 SAM 400
Reply:I am M.I.G. lexington's weld engineer. I do not set up robots, i make sure the robots are set up.. I monitor welds, over see our whole weld process, make weld parameter sheets, do cost reduction projects, etc. I dont know what you dont understand about the question. I just figured not many engineers ask questions on here, alot of guys think engineers are stuck up, know everything, and dont need anyones elses opinion. Well im not that type of guy, Im open to any suggestion to make job easier. Actually you sorta sound alot like an engineer to me with your arrogant comments. sorry if you dont like that, but thats just how your first paragraph comes off as. Anyways a robot is just as good as the operator or setup guy. If your operator doesnt change the tip, keep the fixture clean, clean the nozzle, crash the robot, then yes a robot will repeat mistakes..We try to weld hot and fast to decrease cycle time and have more production, but every weld we do goes through weld optimization. We do not use lubricators on our wire. Thanks for the suggestions at the middle and end of your post, you could have left the first paragraph out. It really had nothing to do with the question asked.
Reply:Should have added in the first post normally it is the last 1/3 or so off a spool where the trouble starts but sometimes right off the bat. The local welding supply used to always open the package and look at the wire color before sending it out the door. Really like ESAB wire but if not bright and shiny copper color nothing but a pain in the a$$.
Reply:I agree, we do have trouble with with the wire once it gets down close to the bottom of the barrel.
Reply:Just my 2 centsI would imagine the company is located in Tenessee to save labor costs. Labor costs are further reduced by the use of machine welding. Probably the next step is to locate in China. Or trained monkeys perhaps?Crap in...........Crap out. Chasing the cheap solution usually winds up costing money in the long run. I have no sympathy for the situation.Maybe the products can be sold to robots, unemployed/underemployed humans tend to reduce consumption. AND robots apparently can't detect flaws, so it's a win-win"Any day above ground is a good day"http://www.farmersamm.com/
Reply:Originally Posted by farmersammJust my 2 centsI would imagine the company is located in Tenessee to save labor costs. Labor costs are further reduced by the use of machine welding. Probably the next step is to locate in China. Or trained monkeys perhaps?Crap in...........Crap out. Chasing the cheap solution usually winds up costing money in the long run. I have no sympathy for the situation.Maybe the products can be sold to robots, unemployed/underemployed humans tend to reduce consumption. AND robots apparently can't detect flaws, so it's a win-win
Reply:Actually we take jobs out of mexico and just recently turkey, and bring them to tn. Im sure the owner is like everyone else he just tries to save when he can. Some day he might take it to another place. i hope not, but im sure it could happen. Im just glad I got the opportunity to get a lot of really good experience with out an education. The education i will be working on shortlyLast edited by mattadams6552; 12-06-2012 at 03:35 PM.
Reply:Farmersamm is just a little grumpy today because he has to go to Tulsa and let someone dig deep into his wallet.
Reply:Originally Posted by mattadams6552I forgot to mention, I know what causes the porosity, and the burnthrough. Im just seeing if there is any system out there that might catch the defects that is better that what i already have. I should have mentioned this in my first post, and i just realized it reading the responses.
Reply:PM Sent, and i look forward to hearing back from you.
Reply:Originally Posted by mattadams6552I am M.I.G. lexington's weld engineer. I do not set up robots, i make sure the robots are set up.. I monitor welds, over see our whole weld process, make weld parameter sheets, do cost reduction projects, etc. I dont know what you dont understand about the question. I just figured not many engineers ask questions on here, alot of guys think engineers are stuck up, know everything, and dont need anyones elses opinion. Well im not that type of guy, Im open to any suggestion to make job easier. Actually you sorta sound alot like an engineer to me with your arrogant comments. sorry if you dont like that, but thats just how your first paragraph comes off as. Anyways a robot is just as good as the operator or setup guy. If your operator doesnt change the tip, keep the fixture clean, clean the nozzle, crash the robot, then yes a robot will repeat mistakes..We try to weld hot and fast to decrease cycle time and have more production, but every weld we do goes through weld optimization. We do not use lubricators on our wire. Thanks for the suggestions at the middle and end of your post, you could have left the first paragraph out. It really had nothing to do with the question asked.
Reply:At our facility, only 2 engineers out of 7 have actual degrees. No i dont have a piece of paper from a school that says im an engineer, but if you would like i can snap a pic of my job description, job title, etc for ya, i mean if its really that big of a deal..... 3 of my really good friends in high school went to school to be engineers, i took the technical college path. I have a machinist cert. and some welding certs. Im engineering, and they are still looking for a job. I got extremely lucky to get the job that i have, with no 4 year degree, but like i said im working on getting the paper so that if something ever happens here i have the schooling to go with the experience.
Reply:Originally Posted by lars66Farmersamm is just a little grumpy today because he has to go to Tulsa and let someone dig deep into his wallet.
Reply:Feedback loop on a modified tight beam automobile type backup sonic sensor pointed at the weld, mounted next to torch. continue travel of arm past end of weld to continue the scan with sonic sensor. If the feedback changes, you have a alert code. Use the right frequency, and it will detect porosity via return frequency change.Last edited by Housedad; 12-07-2012 at 10:57 AM.Victor Oxy weld/cut Setup Meco Midget Torch outfitEverlast 2016 250EX & Power Cool W300Lincoln Power mig 210MP w/amptrol and Spool GunEverlast Powerarc 200 Hypertherm Powermax 45Bandsaw, Lathe, grinders, press, Bridgeport, etc.
Reply:Originally Posted by mla2ofusSo, Kolot, what's your excuse?? It appears someone pissed in your cheerios!! Mike |
|
发表于 2021-8-31 23:31:13