Welding helmet electronic auto darkening filter (ADF)

[dorkshoei]

First off, I'm assuming I'll have more luck finding someone who is knowledgeable about electronics (and also welds) here on eevblog than I will finding someone knowledgeable about electronics on a welding forum.

The auto darkening filter (ADF) in my welding helmet stops working after a couple hours of welding but the following day it works fine again.  Repeatable pattern.    The ADF is basically an electronic LCD based darkening unit.

Normally when the sensitivity is set to max,  it will darken just with the ambient light from my LED shop lights,  so I have to back the sensitivity down to 8.  At this point, if I look up at the LED overhead lights it will darken.  If I look back at the work,  it is clear.  If I strike an arc,  it darkens.   

After an hour or two of welding,  the ADF starts to become intermittent,  then stops entirely.   If I set sensitivity to max and look up at overhead LED lights,  it will not darken at all.   Oddly if I go outside and look up at sun,  it will darken.  But it won't darken for an arc.

If I leave it overnight, the next day it's working fine again for another hour.

It is supposed to flash the screen if the batteries (2x 3024) are bad.  This doesn't happen.  Both batteries test good.  Replacing with new batteries does not fix problem.  I've cleaned and tightened the battery contacts.

There are a few videos on Youtube on "repairing" these but they all seem to address the case where the unit has no user replaceable batteries (the batteries are soldered onto the pcb).     On mine the batteries are field replaceable via access panels.

The one thing that may be a factor is that I can hear something rattling if I shake the ADF.   Possible this has always happened, I can't say.  When it rattles it sounds a bit like the clicking noise the ADF makes when it activates/darkens.
 
I know these things commonly fail after a few years (mine is 5 years old) but the fact that it's repeatedly ok the next day makes me wonder if it may be repairable?  I need something reliable so I'm not opposed to cutting open the ADF to see,  curious is anyone has any ideas on what may be at fault.   I have an SMD soldering setup.

[fzabkar]

I know nothing about these devices, but I would think that you would need to locate the UV sensor(s) and examine the circuit in that area.

[dorkshoei]

I know nothing about these devices, but I would think that you would need to locate the UV sensor(s) and examine the circuit in that area.

Thanks.  I'm hoping someone here is skilled in electronics,  welds and has repaired one of these before.   Big hope I know.

[fzabkar]

I would think that your helmet would have something like these:

https://www.isweek.com/wholesale/uv-sensors_125

[Cerebus]

My guess, and it is a guess, is that it's nothing to do with the electronics and everything to do with heat. You weld, the filter absorbs light, it gets hotter. You weld for a long time and the filter gets significantly warm. The filter 'fails'. You stop welding. The next day you try again, with a cold filter, and it works again.

The filter is a Liquid Crystal Display and the liquid crystal in question is affected by heat as well as electric fields. My weather station is demonstrating this at the moment, as the combination of a lowish battery and uncommonly warm weather for the UK is affecting the display contrast and making it hard to read.

Here's a sheet of liquid crystal demonstrating the effect of heat on the orientation of the crystals:



This is the classic 'mood ring' effect and this example is engineered to show the effect of heat as much as possible using a liquid crystal specifically designed for the effect. In this case just the heat from a hand. Your LCD is behind a highly effective infrared filter, so the effect takes much longer to show up (and of course isn't engineered to deliberately show pretty colours).

Try not to weld with your nose in the weld puddle, lean back a bit, take more breaks, wave your mask around to cool it during your breaks and I suspect that your problem will go away.

There is quite possibly a combination of heat and ageing, in which case cooling off the filter will only offer temporary respite and it's time for a new welding hood.

[dorkshoei]

it's nothing to do with the electronics and everything to do with heat. You weld, the filter absorbs light, it gets hotter. You weld for a long time and the filter gets significantly warm.

For the last few weeks I've just been welding 20ga stainless,  30A max current.  Not that much arc light.

It's an entirely new issue.  Previously I've been able to weld at 150A for several hours.

When it stops  darkening on arc,  it'll still darken if I go outside and look directly at sun.   To me it seems less to do with the LCD and more to do with the light detection as fzabkar suggested.

[Cerebus]

Like I said, just a guess. Yup, 30A ain't much so I'm probably barking up the wrong tree.

As to the Sun, skylight is polarized and the LCD relies on polarization for its darkening effect. A shift in crystal angle is what does half the polarization  and a polarizing filter(s) in line with it does the other half. It's that crystal angle (the 'twist' in twisted nematic fluid, another name for the type of liquid crystals used in LCDs) that is driven by both electric fields and heat. So it is possible that you're seeing darkening with skylight but not with much more intense welding light as a result of the skylight being polarized and the filter only doing some of its normal level of polarization. I took that into account (but didn't explicitly say so) in effectively discarding "the sun" as a counter argument to what I was thinking.

But, as I say, "30A" is a more effective dent in my theory. Anyway, it's an easy hypothesis to prove or disprove, just try cooling your mask off next time it happens. Then we'll know whether I'm an insightful genius or just talking an idiot out of his arse.

By the way, for those who aren't familiar with these beasties, here's a couple of illustrations:

Here's a typical generic autodarkening insert. Note the huge silicon photodiode array that is the light sensor:



The better quality (read "a manufacturer's name that you recognise") tend to have less obvious photodiodes and/or more than one. In this case a 3M filter:


Here's a typical stackup of the filter:

[wizard69]

I'm pretty sure the sensors that that trigger darkening are looking for UV.   

In any event this may seem silly but is the sensor getting dirty and you wipe the hood down every morning.   You probably have already addressed this but I know how dirty my hood gets running flux core.

If I understand your series of posts you indicate that the hood responds to sunlight even after it has started to fail with weld.   This is odd but the sun delivers a fairly broad band intense light source.   It may be an issue where the detector is no longer responding well to UV.

I realize that the Lens assembly provides 100% protection for UV from the weld arc but even so I wouldn't mess with this and would simply go out an buy a new auto dimming module.   That might fly in the face of normal practice on this forum but my eyes are important to me.   The other thing is that these auto darkening units have improved substantially over the years.   I'd just use this as a reason to buy a new high contrast hood.    I know this sound like a cop out for the electronically inclined and in a way it is, however i'm a big fan of throwing out safety devices that have malfunctioned.   At work I would never send a safety relay out for repair nor try to repair one myself.    It is just a mind set I like to maintain.

it's nothing to do with the electronics and everything to do with heat. You weld, the filter absorbs light, it gets hotter. You weld for a long time and the filter gets significantly warm.

For the last few weeks I've just been welding 20ga stainless,  30A max current.  Not that much arc light.

It's an entirely new issue.  Previously I've been able to weld at 150A for several hours.

When it stops  darkening on arc,  it'll still darken if I go outside and look directly at sun.   To me it seems less to do with the LCD and more to do with the light detection as fzabkar suggested.

[Twoflower]

Note in advance: I'm not a welding expert (in fact I never welded). But ain't the 'big array of sensors' just solar cells to power the LCD? At least that's my understanding after a quick goolgle-fu: https://www.optrel.com/en/technologie/blendschutzkassette-adf/funktionsweise-einer-blendschutzkassette-eines-auto-darkening-filters/

[Cerebus]

But ain't the 'big array of sensors' just solar cells to power the LCD?

In the cheap ones it seems to do double duty as both power source and detector. A solar cell is just a whacking great photo diode after all. The better class of product seems to have a solar cell (sometimes none), a primary battery and a couple of smaller photodiodes for sensing (two for redundancy and increasing field of view of the sensors would be my rationale for doing it that way).

[dorkshoei]

But ain't the 'big array of sensors' just solar cells to power the LCD?

Yes.  You're correct.    Cerebus was mistaken regarding that photo from AliExpress.   If you look closely at the image you can see the 2 sensors below the solar cell.

In this picture (of my ADF) the 4 red arrows are the light sensors (cheaper units only have only 2)    The green arrow is the solar cells.  The batteries (typically coin cell)  are backup for when there is insufficient light.   Not all units have solar cells.  Some operate just on batteries.

I tried cleaning them (q-tip and isopropyl).  They were dirty but it made no difference.

I'll buy a new helmet.  Once it arrives I'll crack open the old ADF and see if I can figure out what's wrong,  maybe fix it as a backup.

[dorkshoei]

I wouldn't mess with this and would simply go out an buy a new auto dimming module.

This lid is 5 years old.  No replacement modules available.  Possible I can find something of a similar size but usually a replacement module is about the same cost as the original helmet that contained the module. 

[dorkshoei]

I wouldn't mess with this and would simply go out an buy a new auto dimming module.

This lid is 5 years old.  No replacement modules available.  Possible I can find something of a similar size but usually a replacement module is about the same cost as the original helmet that contained the module.

Interesting.  My unit measures (approx) 132mm x 115mm x10mm

These are similar:
https://www.aliexpress.com/item/32620432927.html 
https://www.aliexpress.com/item/32954110785.html

[Cerebus]

Yes.  You're correct.    Cerebus was mistaken regarding that photo from AliExpress.   If you look closely at the image you can see the 2 sensors below the solar cell.

Well spotted. I didn't spot them and was guessing that any photo I found from an image search that had "aliexpress" slapped across it was an example of the lowest common denominator* so, truthfully, I didn't look too hard for them. I have seen a teardown of an ADF (that I wouldn't touch in a month of Sundays) that only has the silicon photo cell - from the same sort of places that churn out "welding gloves" that you wouldn't risk doing gardening in.

* A little unkind, if you're careful there are some very good vendors on Aliexpress.