EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: JGMongAL on September 09, 2016, 03:36:10 pm
-
I have been tasked with repairing the control board for one of my employers Lincoln Electric Precision TIG 185. We found the fault and have ordered a replacement encoder. As you can see from the picture it is encased in a thick layer of Epoxy. I am wondering what type of epoxy this is so I can order some to hold the new part once it is in place.
Also if anyone has any better ideas for removing the old epoxy than a small hammer and chisel or a Dremel please give me a shout.
-
Always try low impact methods first.
That is to say, warm the epoxy and try digging into it with a sharp tipped tool. If no progress,try a higher temperature. At 150C surface temperature, if no success, resort to abrasion with a ball mill in a Dremel or similar tool.
New epoxy can be a common general purpose two part epoxy. I suggest the 5 minute type two part epoxy such as that made by Loctite.
Fraser
-
Lincoln loves slathering their boards in all sorts of goop (to prevent repair). Years ago they would put lots of silicone RTV all over the board, very annoying. (I like Miller products better, but do own a Lincoln MIG welder :) ). Lincoln also like selling the same model number with loads of different revision numbers making it hard to order parts. OK, rant off.
You can use a soldering iron with an old tip to get rid of most of it, then you are down to the methods described above. At least with the silicone you could pick most of it off the board and then use ether to dissolve the rest. Myself I would use a good quality silicone RTV on the new encoder.
-
That looks like it might be a right shit of a job depending on how hard the epoxy is.
There is every likelyhood you will damage the PCB and your employer must be made aware of this.
I would first try a narrow carving chisel attempting to peel it off in layers but if it's too hard for that a Dremel or die grinder might be the best weapon. Don't attempt to remove all the way to PCB level initially, leave a thin layer rather than risk damaging tracks. When it's all thin you should be able to see the areas where there's tracks and work around them leaving them entirely unless the encoder can't be extracted.
How much it has flowed under the encoder will be the problem but you may be able to cut that now thin layer with a snap off blade knife. Again, evade areas with tracks. Getting each soldered pin clean will be the next trick especially if it's been assembled with that Pb free muck. If you're lucky the holes for the pins are oversized and the pins are well centered, even so you'll need some luck to accomplish a repair like this as undoubtedly it's been assembled without any thought of such a repair. :-- :-- :--
Take your time and be bloody careful, I don't envy you.
-
Try to stick it in up side down to a glass with Acetone, so that only the encoder and the epoxy will touch the Acetone.
Leave it in there for a few hours and the top layer of the epoxy will peel away easily. Repeat this until all epoxy is gone.
I have been successful with this in some cases but not all types of epoxy are reacting to Acetone.
-
If heat & acetone don't work, I'd try methylene chloride (aka Dichloromethane, MEC, DCM, di-chlo, or good old fashioned paint stripper) - evil stuff, will attack many plastics & rubbers, but it's good for softening most hard-cure epoxy & poly resins & won't attack most metals
Roughen up the surface to break the skin, carefully paint it on (I'd use MEC-based gel paint stripper & build a little bund with plasticine around the job to keep it off the rest of the board), leave it for 10 minutes, & scrape away the softened goop. Repeat 2-3 times if necessary. Don't go too far or you'll likely strip the soldermask & possibly start attacking the board material.
-
I'ld go the mechanical route: Since the encoder is anyway going to be removed, cut the shaft and shaft sleeve with a saw or a small grinding disc, then clamp the wole PCB in a (CNC) mill. Take the PCB surface as a reference and mill everything in the encoder area down in small increments (to keep cutting forces low) to a few tenth of a millimeter above the PCB reference. Once there, the remians shouldn't affect installation of the new encoder. Risk of damage would be minimal this way. Hand-held tools / grinders may produce so much reaction force that it slams the tool to locations where it causes damage, so a rigid system as described above is always preferable.
Good luck,
Thomas
-
I'd try acetone first for sure. Don't submerge the board though. What you want to do is make a dam around the component with silly putty or something similar and make a little puddle of acetone covering the epoxy. The top layer will soften after several hours and you will be able to peel it away with a small flat screwdriver or something similar. A few applications/repetitions should do it. If the acetone doesn't soften it, then go the mechanical route - but be sure to wear a respirator or use a vacuum to capture the dust.
-
I'd try and unsolder the encoder (assuming it is through hole) and see if the epoxy is strongly bonded to the PCB or not. In my experience epoxy resin doesn't always bond well so a little prying & wiggeling may be enough the break it off cleanly.
-
Heat it to 100+ degrees Celsius, something like hot air gun or preheater are the best. Many epoxies become soft enough that you can pierce and remove them with a sharp tool.
-
dont mess with Methyl Keytone (MEK) unless you have a vented area and safety gear.
-
dont mess with Methyl Keytone (MEK) unless you have a vented area and safety gear.
Methyl ethyl ketone (MEK) doesn't do anything to cured epoxy anyway (well, it probably does something, but you might be better off waiting for the phoenix bird to fly down and sharpen its beak against it for a few millenia instead), although it's good for thinning & cleaning up before cure. Of the ketones, DIK/DIBK (Diisobutyl ketone) will I think soften most cured epoxies, but afaik it's even nastier than MEK.
In the other hand, MEC (Methylene chloride aka etc. as I wrote above) is the active ingredient of most cured epoxy solvents e.g. "Attack Epoxy Solvent", "Agent MS-111", etc., is much less hazardous than the ketones, and is fairly easy to get (old fashioned paint stripper).
-
dont mess with Methyl Keytone (MEK) unless you have a vented area and safety gear.
Methyl ethyl ketone (MEK) doesn't do anything to cured epoxy anyway (well, it probably does something, but you might be better off waiting for the phoenix bird to fly down and sharpen its beak against it for a few millenia instead), although it's good for thinning & cleaning up before cure. Of the ketones, DIK/DIBK (Diisobutyl ketone) will I think soften most cured epoxies, but afaik it's even nastier than MEK.
In the other hand, MEC (Methylene chloride aka etc. as I wrote above) is the active ingredient of most cured epoxy solvents e.g. "Attack Epoxy Solvent", "Agent MS-111", etc., is much less hazardous than the ketones, and is fairly easy to get (old fashioned paint stripper).
I use DCM (dichloromethane aka methylene chloride) all the time in the lab. It's pretty much interchangeable with chloroform. I will give one warning, if you get it on your fingers, it can cause a burning sensation. This is especially true if the soft area between the fingers. It burns like really concentrated icy hot and it will go through nitrile and latex gloves like nothing. That being said it causes no actually damage, just can be incredibly uncomfortable.
Sent from my LGLS992 using Tapatalk
-
Hi.
I realize this is an old post, but hopefully someone will read it. I need to replace the decoder on this board as well. Removing the decoder was not difficult, however I cannot see any part number on the removed component. Has anyone actually done this and have the decoder make and part number?
Thanks very much!
-
Some very bad safety advice in this thread. M.E.K. has a PEL (a permissible exposure limit during a work shift, set by OSHA) of 200 ppm, but the PEL for dichloromethane is just 25 ppm. Chlorocarbons are also carcinogens.
-
Hi.
I realize this is an old post, but hopefully someone will read it. I need to replace the decoder on this board as well. Removing the decoder was not difficult, however I cannot see any part number on the removed component. Has anyone actually done this and have the decoder make and part number?
Thanks very much!
Welcome to the forum.
You mean encoder don't you ?
Even though it's been a while since the OP posted you could send him a PM as it will go to his email address.....but IIRC there's been a change recently in the forum settings so you need to get five posts under your belt before forum's PM's become available to you. Just a reply in this thread saying 'Thanks' will get you to two posts and then you'll only need to do three more. :)
-
Hi Tautech.
Thank you very much for responding. I appreciate your advice.
And yes, I meant encoder. Stage fright........ :)
-
Hi Tautech.
Thank you very much for responding. I appreciate your advice.
And yes, I meant encoder. Stage fright........ :)
Let's try and help you through that. :)
Is it a detented (stepped) encoder ?
-
Hi Tautech.
I would say yes, as it has 24 distinct "stops, or bumps, or points of physical resistance" in a single rotation. It has 3 pins. I looked it up on digikey and seemed to narrow it down to either Bournes or TT Electronics. There are a few models offered from either manufacture. The spec sheet on a TT Electronics encoder says "2 Bit gray code" whereas the Bournes data sheet that I looked at has "2 Bit quadrature code". I can look up to see what the difference is between these two types and how they function, but the fact is that I don't know what is actually required for this circuit (which has no schematic) to operate correctly. That's why I am hoping someone can tell me "I ordered this part number and our welder functions properly". It's funny that this problem appears to be common for this type of welder, from what I have read.
Thank you for looking at my post!
-
Hi Tautech.
I would say yes, as it has 24 distinct "stops, or bumps, or points of physical resistance" in a single rotation. It has 3 pins. I looked it up on digikey and seemed to narrow it down to either Bournes or TT Electronics. There are a few models offered from either manufacture. The spec sheet on a TT Electronics encoder says "2 Bit gray code" whereas the Bournes data sheet that I looked at has "2 Bit quadrature code". I can look up to see what the difference is between these two types and how they function, but the fact is that I don't know what is actually required for this circuit (which has no schematic) to operate correctly. That's why I am hoping someone can tell me "I ordered this part number and our welder functions properly". It's funny that this problem appears to be common for this type of welder, from what I have read.
Thank you for looking at my post!
I'd say if you get one in the same form factor, shaft size and pin-out you'd be fairly safe as it's not likely to be anything too special or a proprietary part. Check which type is the cheaper and that'll probably be the right one as manufacturers like BOM cost to be low as possible.
-
Hi.
I will do some more looking and then might end up trying it as per your advice.
Thank you again!
-
Sorry, but this makes 5 posts and now I'll attempt to send him a PM. Hopefully he has some info on what he did.
-
Sorry, but this makes 5 posts and now I'll attempt to send him a PM. Hopefully he has some info on what he did.
Hi Scott, the user details for the original poster show that he was last active on this forum in May 2017. So I'm afraid the chances of getting a reply to the PM are slim. What additional info do you need, beyond the other posters' suggestions?
-
Hello all. sorry for the extended absence. logged in here on a lark and saw Scotts pm. though it has been a long time I thought I would check in and see if He had found the part number. it is listed in the Service manual for that particular welder available from Lincoln. (M17789-4) by the manual. I did not plan to be away from here so long but health issues had my sidelined for awhile...almost permanently.