Welding Tips and Tricks

Not that way!

I was practicing TIG tonight and got flashed bad.
Working on just running straight beads, no filler on 1/4” mild steel. Completely clean. First couple went fine then suddenly blam, absolute brilliant flash partway through the next bead. I didn’t dip the tungsten.
Primeweld 225
130 amps
#8 cup
Stubby kit with gas lens
18 cfm
2% lanthanated

Completely overwhelmed my Esab Savage set at 11. Now there’s a big spot in my vision.

Any idea what might have caused the flash?


Sent from my iPhone using Tapatalk

I hate getting flashed! By welding arcs....

My guess would be it is 100% a malfunction of the hood, and nothing to do with the welder or torch setup. Maybe check your sensors to make sure they are clean. Also may not hurt to remove the battery for a bit and then reinstall it to hopefully force the logic to cycle.

Spartan wrote:I hate getting flashed! By welding arcs....

My guess would be it is 100% a malfunction of the hood, and nothing to do with the welder or torch setup. Maybe check your sensors to make sure they are clean. Also may not hurt to remove the battery for a bit and then reinstall it to hopefully force the logic to cycle.

Thanks
I’ll do that. It’s a new helmet so I will check carefully.


Sent from my iPhone using Tapatalk

I may have had the sensitivity set too low.


Sent from my iPhone using Tapatalk

Auto helmets sometimes don't like tig. The shadows thrown out play curry with the sensors. Speedglas used to have a hidden arc mode, but now it's just better sensitivity.

I left the helmet in the sun all day to ensure it was fully charged.
Bright sun out side. Put on the helmet, sensitivity is set to max, look up at the sun and nothing happened. Curiously it did go dark for a second as I raised my head but I couldn't get it to do it again.
Unless anyone has any suggestions it's going to go back as defective.

The sun doesn't usually cause helmets to darken. Do some proper welds and see how it works. Don't worry about the flash, it's just visible light. Harmful rays can't go through the plastic. Obviously don't keep using it if it flashes again though

IMHO, getting flashed is just a part of life with auto dark helmets. The funny part is, I build all sorts of visible and non-visible light sensors for automation, do all manner of micro controller design work, and there's simply no reason for this stuff to happen. I don't know who does these designs on helmets that can't see a G(*&$% D(*&) welding arc, I mean holy (&*$%, if you can't set a threshold on that, seriously, look into a new line of work.

I think 3/4 of my hatred of the things is the way the piss poor design infuriates me. The other 1/4 is how the batteries which need replaced ALL the time, are buried inside the sensor that's inside 3 layers of framework and require the removal of every part of the unit save the headgear. Why not have batteries in a compartment external to the shade mechanism? Seriously, 25 years of making these things and nobody thought of that yet?

"Yesterday, I couldn't spell eggineer, and today I are one".

That used to be a joke when I was at Purdue, now it is way too true!

weldin mike 27 wrote:The sun doesn't usually cause helmets to darken. Do some proper welds and see how it works. Don't worry about the flash, it's just visible light. Harmful rays can't go through the plastic. Obviously don't keep using it if it flashes again though

Interesting. I thought for sure it would darken. I'll give it a go this morning.

BugHunter wrote:IMHO, getting flashed is just a part of life with auto dark helmets. The funny part is, I build all sorts of visible and non-visible light sensors for automation, do all manner of micro controller design work, and there's simply no reason for this stuff to happen. I don't know who does these designs on helmets that can't see a G(*&$% D(*&) welding arc, I mean holy (&*$%, if you can't set a threshold on that, seriously, look into a new line of work.

I think 3/4 of my hatred of the things is the way the piss poor design infuriates me. The other 1/4 is how the batteries which need replaced ALL the time, are buried inside the sensor that's inside 3 layers of framework and require the removal of every part of the unit save the headgear. Why not have batteries in a compartment external to the shade mechanism? Seriously, 25 years of making these things and nobody thought of that yet?

Ok that definitely made me laugh! Thanks

I hear you on the battery access. My old Esab had non-replaceable batteries. Well non-replaceable by me anyway.

I ran a few beads this morning and all seems to be well. Seems it was operator error with the sensitivity too low.
Thanks to all who helped me sort this out!
D

Spartan wrote:I hate getting flashed! By welding arcs....

My guess would be it is 100% a malfunction of the hood, and nothing to do with the welder or torch setup. Maybe check your sensors to make sure they are clean. Also may not hurt to remove the battery for a bit and then reinstall it to hopefully force the logic to cycle.

Yup

All shields were not created equal. Speedglas is the absolute best. Consider that that was all they did before 3m bought them. As with Lincoln engine drives and miller tigs, the central product is usually the best

VA-Sawyer wrote:"Yesterday, I couldn't spell eggineer, and today I are one".

That used to be a joke when I was at Purdue, now it is way too true!

That's cute... I'll assume (possibly incorrectly) that was a jab at me.

In point of fact, my spelling is usually pretty good as long as I'm not using voice to text.

How about I add a little context:

Here's an optical encoder. For the record, I use these components in devices I build.

Encoder datasheet: https://docs.broadcom.com/docs/AV02-1867EN
Codewheel datasheet: https://docs.broadcom.com/doc/AV02-1450EN

If you read those datasheets, those encoders will operate at 30,000 RPMs 2048 pulse per rev. (In fact, other higher res models will do the same rpms @ 32768ppr... That's over a million individual transitions per second, even on the slow one.
Side note: The waveform output from these things on a scope is nothing short of beautiful...

So, these folks can make a sensor that will see an infinitesimal amount of light, looking through microscopic holes in a metal codewheel, and not miss one (with two different sensors with positions offset from each other, neither missing...), but folks making welding helmet shades can't reliably tell when the sun is out or not...

Just to add to the intrigue, my circa 2004 Esab AD helmet darkens when I look at the sun. So Maybe the AD tech has progressed to the point where it's more 'discerning'?
"I know you're looking at the sun, so I'm not going to darken. Get back to work you slacker!"

Ones I've owned have darkened even when aimed at florescent lights. That's of course when it was working. Pretty much any light source should make it trigger. I'm just surprised at how insensitive they typically are.

Components I've used for making my own discrete sensors are definitely limited to a particular range of wavelength, be it with visible or non-visible light. What I've never seen is a photo-detector that didn't give a response when bombarded with light in the way a welding arc would.

I've used bright lights to help see the weld area, but I don't want the lens to darken before I initiate the arc, that's where the x mode on miller hoods shine. I think other manufacturers have something similar

LtBadd wrote:I've used bright lights to help see the weld area, but I don't want the lens to darken before I initiate the arc, that's where the x mode on miller hoods shine. I think other manufacturers have something similar

I’ve definitely noticed that if I glance up and catch any light from my LED fixtures it goes dark. When lighting the work I have to do is in a way that the light is not pointing at me at all.


Sent from my iPhone using Tapatalk

My shop is 24 x 32 x 10ft with 12 4ft 2 tube LEDs and 8 adjustable 3 petal LEDs all 5500 - 6000K so it's bright. I initially just aimed the helmet at the ceiling and adjusted the sensitivity just to the point where the auto darkening didn't activate and that was it. Never had a flash yet. I have also welded outside in the sun with no issue. I haven't tried this one pointed at the sun to see if it would react but my old helmet wouldn't. Found that out trying to watch the last eclipse when I had to dig out an old fixed helmet with a pair of Ray Bans. Present helmet is a Vulcan Arcsafe.

homeboy wrote:My shop is 24 x 32 x 10ft with 12 4ft 2 tube LEDs and 8 adjustable 3 petal LEDs all 5500 - 6000K so it's bright. I initially just aimed the helmet at the ceiling and adjusted the sensitivity just to the point where the auto darkening didn't activate and that was it. Never had a flash yet. I have also welded outside in the sun with no issue. I haven't tried this one pointed at the sun to see if it would react but my old helmet wouldn't. Found that out trying to watch the last eclipse when I had to dig out an old fixed helmet with a pair of Ray Bans. Present helmet is a Vulcan Arcsafe.

That’s interesting. I’ll see if that helps.
My shop also has 5500K LEDs and yes they are very bright.


Sent from my iPhone using Tapatalk

Bughunter,
Not aimed at you, but it was because of your post. I agree 100% with what you was saying about poor designs.
I run a repair shop where I work on everything from small diesels, down to little 2 strokes. I also do some custom fabrication. I am not an engineer, but I put more effort into the stuff I design, than seems to go into the stuff I work on. Cheap steel fasteners in high corrosion areas. Cheap plastic that gets very brittle in less than 3 years. Some stuff appears to be designed to fail.
I suggested a while back that they use a RF link between the helmet and foot pedal. That way the lens could darken just before the arc starts. That would eliminate the short duration flash that occurs during each arc start. You don't notice it, but it does happen, and can cause fatigue after a long day of welding. They can even adjust darkness with welding current.
Use a standard RF link such as Bluetooth, and any brand welder would work with any brand helmet.

Gotcha, sorry for the misunderstanding. I wasn't sure how to interpret that. I've heard the "well if you think you could do better, why don't you?", argument so many times I almost expect it.

I often wonder if they don't design the sensors for picking up on infrared and then neglect to place the sensor somewhere that they are not hidden by the visible-light-passing plastic lenses. Or they make the protective lenses out of a material that do not pass the spectrum of light the sensor is looking for. The Plastics are actually rather unintuitive. Plastic which will pass visible light typically will not pass infrared, and vice versa. For instance, if you ever look at a motion sensor, the plastic over top of it is totally opaque to your eyes. If you ever build a sensor using one it will say right in the datasheet that you need to use something like polypethylene , the material used to make a milk jug. You may not be able to see through it but the sensor can. If you use a clear lens or plexiglass or something like that, forget it. You may as well have put a metal shield in front of it. The opacity varies with thickness but no matter how thin the wrong material is, it's the wrong material.

I've never really investigated it, but it seems like they might be using incompatible sensors and plastic shields, assuming that the light is going to be so powerful that no matter what they use, enough light will come through it. Then in reality a little dirt gets on the lens and all of a sudden the sensor doesn't hit the threshold to trigger the shade.

A better analogy might be, think about the last time you used the remote for your television and it didn't work. We might be talking about six or seven orders of magnitude less power, but we expect it to sense that all the time. LOL

I usually find it works best to adjust your helmet to where it's just past that point to where the brightest lights in the shop don't trigger it but not any further than that. Otherwise if you're like me you'll get annoyed by your helmet constantly dimming, crank the sensitivity down and then go light up on some thin stainless and realize it's not going to trigger.

On the bright side, getting momentarily flashed through a helmet is no big deal, it'll block all UV and IR regardless of the shade or if it dims at all and those are the big dangers to your eyes.

Miller helmets actually do have a workaround for this problem, and I've seldom dealt with this since I've started using a digital elite, they have a setting called X-mode where it'll electromagnetically sense the arc rather than relying on light at all. If you're working around other welders it's not always the best option as with too high a sensitivity it'll trigger from everyone around you and too low might leave you staring right into your buddies arc because he is a foot too far from you. But for any other situation it's amazing.

I've been giving a bit more thought to the 'perfect' welding helmet. With a RF link between the welder and helmet, there are a few other things we can do. How about a Heads-up display that tells you the current. Add red, yellow, and green LEDs to indicate welding voltage as it relates to arc length. We can even change the shade in sync with pulse welding at low to medium frequencies. No flicker.

If you really want to go overboard....... a small camera mounted on the torch. High Res display in the helmet. 5X and 10X Zoom when desired.
IR sensors to monitor temp of surrounding metal. A warning system to let you know when the 'bottom is about to fall out'.
How about displaying guide lines for desired puddle size.