Auto Invision II - Porosity on start-up

[TimW.]

I have a Miller Auto Invision II MIG welder connected to a Fanuc Arc Mate 100i robot. The torch is cooled with a Miller Coolmate 4. We are using MC-706 Metal-Shield Flux-Core wire at 25VDC with a flow rate of Argon at 40 cubic feet per hour as measured with ball gauge. The wire speed is 300 IPM, and the travel rate is 27 IPM.

This is a proven process, and we are using the same programs and consumables now as we have for about 15 years.

We are now getting porosity in the first 5 inches of the weld. We have been through the gas thoroughly with only a few question marks remaining there. I have started looking into the electrical circuit.

The display shows 25 VDC, and about 24.8 VDC is measured at the output of the welder. At the AMD-4 wire feed on the Robot Arm, the voltage is measured at about 1 VDC less or 23.8. The electrical connections have been cleaned, and new cables are on order. It is set for 245 Amp, so the cable resistance would be around 8 milliohm to produce a 1 VDC drop.

Here are my questions. On start-up, the voltage at the wire feeder will swing in 100’s of millivolts through the trouble area while the voltage at the output of the welder swings in the 10’s of millivolts. This swing gradually slows and the porosity disappears. This could be due to the cable resistance (V=IxR) but I am not sure. We are welding in a V groove, so it is difficult to tell if an erratic arc is jumping from one side of the groove to the other, or if we have an electronic issue where it is having a difficult time getting initial control of the arc. How can I tell the difference between the two?

So, we weld an oval shape; pause for 10 seconds, then resume on the other side of the ring that is being welded. There is NO porosity in the second weld so I don’t really think room air is getting “venturi-ed” into the groove. I am planning a test where the weld starts 180 degrees from where it does now to see if the porosity moves, or if it stays in the same spot.

There is a qualified service center nearby but they want me to send the power supply in. Would they be able to qualify / calibrate it without a load?

Also, which board controls the arc? Is it the Welding interface board? I am starting to run out of ideas so I am ready to get something else on order.

Thank you in advance,
Tim W.

[H80N]

What is your gas PreFlow setting...??

[TimW.]

I am not familiar with the Gas PreFlow Setting. Is that related to LED #5 on the customer interface board?
That one is called "Input Signal from Robot for Shielding Gas". Is that parameter in the Fanuc Control?

We did a test where shielding gas bypassed the solenoid valve and was directly connected. We manually
turned the gas on and then initiated the weld cycle.

Thank you for your reply
Tim

[H80N]

I am not familiar with the Gas PreFlow Setting. Is that related to LED #5 on the customer interface board?
That one is called "Input Signal from Robot for Shielding Gas". Is that parameter in the Fanuc Control?

We did a test where shielding gas bypassed the solenoid valve and was directly connected. We manually
turned the gas on and then initiated the weld cycle.

Thank you for your reply
Tim

Does the gas flow for few seconds before the weld cycle starts...?? at the same time??... or does it begin to flow after the arc starts..??

[H80N]

Here is an Auto Invision II

manual for overview (probably NOT the correct one)

https://www.millerwelds.com/files/ow...01396A_MIL.pdf

See Section 5-1. Weld Cycle For Pulse Welding & 5-2. Setting Preflow Sequence Display Page 40 of the PDF

We need your serial number to get the correct one

[TimW.]

Hi H80N,

The serial number is LB079050 and the manual looks like it is the right one. Thank you. I can see the Preflow setting described on page 34 and I will look into that now. We did bypass the shielding gas controls and manually turned the gas on before starting the weld cycle. We still had porosity. I will check it again and let you know what happens.

I am setting up a welding test now. We are machining special pieces so we do not have to test on a production piece. This will take some time to get all ready to do. We should be able to do 16 tests when finished.

If all else fails, what do you think about sending the Invision II to the service center listed on this website?

Thank you
Tim

[H80N]

Hi H80N,

The serial number is LB079050 and the manual looks like it is the right one. Thank you. I can see the Preflow setting described on page 34 and I will look into that now. We did bypass the shielding gas controls and manually turned the gas on before starting the weld cycle. We still had porosity. I will check it again and let you know what happens.

I am setting up a welding test now. We are machining special pieces so we do not have to test on a production piece. This will take some time to get all ready to do. We should be able to do 16 tests when finished.

If all else fails, what do you think about sending the Invision II to the service center listed on this website?

Thank you
Tim

would guess it is probably a programing/parameter error... preflow shield gas purge.... unlikely a machine defect

[H80N]

This is the correct manual version...for serial # LB079050

https://www.millerwelds.com/files/ow...96188A_MIL.pdf

.

[TimW.]

would guess it is probably a programing/parameter error... preflow shield gas purge.... unlikely a machine defect

Hi H80N,

The reason we feel it is a machine defect is because this is a proven process; with 10,000's of good parts welded over the years. This affects all product types and is not limited to a bad edit on a single type of part. It affects 100's of programs so it would need to be a global change if it is a programming/parameter error.

The our welders have gone for the day (I am maintenance) so I have not been able to do any testing. We are using Flux core wire (FCAW) and not a GMAW-P pulsed welding process. Our robotic welders are unfamiliar with that parameter, so this will be something new for us. I will try the preflow setting at the first opportunity, but the question remains why now after 15 years? We did tests which included manually purging the shield gas before starting the weld cycle. If the preflow setting works, then it suggests to me that there is still an underlying machine related problem that needs to be fixed.

Thank you
Tim

edit: thank you for finding the correct manual

[TimW.]

Setting-up and getting to the welding test is not progressing as fast as I had hoped. We have several tests to start with including the preflow setting.

One other test I am interested in is to start the weld at a different point in the oval. This should either point to something in the start (erratic arc), or something in the robot arm.

In the meantime, I would like to look just a little bit further down the road if that is possible. The Fanuc control sends commands to the Auto Invision II to adjust voltage and amperage. Given the age of the equipment, I suspect this is done with D to A (Fanuc) and A to D (Miller) chips. Is it possible to run the welder on either a battery box (steady constant voltage) or on a potentiometer if the control is resistive? Has anyone done this?

If this is possible, then I should be able to rule out the command signal from the Fanuc. I would be very interested in other types of tests anyone can think of which may point us in the right direction.

We are using a known process and have been through the shielding gas pretty deep over the past several weeks. We even bypassed the solenoid and manually started the Argon before starting the cycle without changing anything.

Thank you for your thoughts
tim

[cruizer]

Well, its not an Invision problem, its a gun or feeder problem, likely a gun issue, or a gun to feeder issue. How could it possibly be the Invision, as its only the power source. not going to give you Porosity

[TimW.]

Well, its not an Invision problem, its a gun or feeder problem, likely a gun issue, or a gun to feeder issue. How could it possibly be the Invision, as its only the power source. not going to give you Porosity

The gun was replaced and the feeder wheels were turned over to give a fresh surface. The reason we are leaning towards Invision / Fanuc is that it appears to be having a difficult time getting control of the arc. The fanuc is programmed, and it sends control signals to the power source. If the signal from the fanuc is erratic, then the weld will be erratic. If the fanuc signal is good then the Invision might be doing the wrong thing with reference voltage. The fanuc also controls the wire feeder. If there are any simple things that we have not ruled out, then they sure are not coming to my mind.

I plan to post again later but I need to contact someone concerning this problem early this afternoon.

thank you
Tim

[TimW.]

I attached a picture so you can get a better idea. It looks like a gas issue, but we have been through the shielding gas for two weeks and cannot find anything. This test was done with a 1-second preflow, and the program was started 180 degrees from where it normally starts. The porosity still began after start-up. Sometimes it is right away, and other times it is after a couple of good inches. It never shows up in the second weld.

Sometimes it looks like the arc is too hot which suggest a control issue. We have someone from the company that integrated the machine coming on Monday. I think they will most likely find the problem. I will let everyone know what the solution is.

[cruizer]

Hmmm, not a gas issue, that's a heat (high amp) issued, , too much wire feed on the starts. Dial the run in back. You must have a start sequence , dump the start voltage, wire speed, and shorten the start time. Orrrrr, it may quite possibly be the opposite spectrum where the material is cold , and may need a preheat. Porosity is always tough, but in this case not a gas problem. Also found that firms that suddenly switched wires have problems. ESAB appears to be the best wire and rod now a days