Resin Leaks between tank and frame  

Page 1 / 9
  RSS
Xulkal
(@xulkal)
Trusted Member

Hey Folks,

Anyone else have a problem with resin very slowly leaking out between the tank and frame?

  • Using Siraya Tech Fast resin
  • Ensured all the frame screws are snug and secure
  • Verified there is no punctures in the FEP anywhere
  • Replaced the FEP 3 times, just to be sure

No luck and still get slow leaks that need to be cleaned up off of the edge of the moving LCD platform way too frequently. Best I can tell is that the leak is coming from the holes where the tank bolts to the LCD platform, and it seems like the frame metal is too thin and flimsy to get a good seal in that area. Looking for help here as I want to enjoy resin printing more, but this leak has been keeping me cleaning up more than just printing.

Posted : 01/10/2020 3:28 am
Michael
(@michael-19)
Eminent Member

Try tightening the screws holding the FEP to the resin tank tighter than you feel comfortable, but not so tight that you strip the thread.

What you should be seeing is that the FEP is pulled down into the cavity beneath the frame metal, so that the FEP stretches and takes the shape of the cavity.

Yes, the frame metal is flimsy, but its acting like a great big washer for all the screws. The screws are so close together that screwing them down tight should give an even seal to stop liquid leaking. You may even like to take the resin tank out of the printer and fill it up with water? Make sure its not leaking before putting it back in.

I've had a couple of occasions where resin would get under the FEP, but that would be fixed by replacing the FEP, so there must have been pinpricks in the FEP too small for me to see?

Another reason for seeing resin below the FEP may be that you transfer some resin there accidentally when cleaning the resin tank? I use a fresh sheet of paper towel to wipe the underside of the FEP and then am careful not to touch the underside of the FEP again.

This post was modified 1 month ago by Michael
Posted : 18/10/2020 2:35 am
Xulkal
(@xulkal)
Trusted Member

@mbarr

I've tried tightening to various levels starting light and working my way to crazy hard. Did the water test, leaking slowly. Visually inspected the tank for defects, nothing visibly wrong. Everything points to the current tank/frame design being poor and prone to failures. It's like it's missing a necessary o-ring. I'm pretty much out of ideas on how to get my SL1 back up and running reliably without the dread of a slow leek.

Posted : 20/10/2020 8:29 pm
Michael
(@michael-19)
Eminent Member

Sounds like tank, frame metal and FEP should be replaced under warranty. Shouldn't be leaking like that.

Posted : 22/10/2020 9:35 am
gnat
 gnat
(@gnat)
Prominent Member

Just had this happen to me. I absolutely know that resin was not spilled outside the tank in any quantity, but when I went to empty the tank yesterday due to a failed print I found my LCD covered in resin (including some cured areas). I cleaned the tank and can find no sign of a puncture or tear. I filled it with water (well past the 100% mark) and placed it on a paper towel most of yesterday and over night, this morning the towel was still dry. Obviously that is a static test and does not account for the movements during the printing process, but prior to putting it down I did poke at the underside a bit to try to simulate the layer change force.

The FEP is fairly scratched up (though still crystal clear) so I'll change it anyway to be safe as I really don't want to intentionally court having to waste half a day cleaning up the printer and all the cleaning materials, but I am stuck wondering how the resin got out...

MMU tips and troubleshooting...
Posted : 26/10/2020 3:11 pm
gnat
 gnat
(@gnat)
Prominent Member

Haven't changed the FEP on the tank yet as I had a second tank. Found a print failure this morning with part of the print attached to the FEP. Finally got around to dealing with it and when I pulled the tank to empty it I found a nice mess underneath.

Both tanks had their FEPs installed by Prusa. Both tanks passed water leak tests prior to going on the printer. The first tank passed water tests after getting cleaned up. I know for certain that no resin was spilled. WTH is going on? I wasted most of Saturday cleaning up the printer and now I've got to do it again a few days later?

MMU tips and troubleshooting...
Posted : 29/10/2020 6:55 pm
vintagepc
(@vintagepc)
Noble Member

Silly question... but is the resin somehow overflowing the tank when the platform is lowered into it, and then seeping under? Is the tank maybe tilting slightly too much during the separation step?

Have you tried with water in the tank gently pushing up on the FEP film from the bottom? This should show in short order if the pulling action of the lifting build plate is opening a small hole that is otherwise not present. 

Posted : 29/10/2020 8:55 pm
gnat
 gnat
(@gnat)
Prominent Member
Posted by: @vintagepc

Silly question... but is the resin somehow overflowing the tank when the platform is lowered into it, and then seeping under? Is the tank maybe tilting slightly too much during the separation step?

It's scary how closely we think alike 😉

I did think of that mainly because it parks the tank in the down position after a print. If I refill to 100% at that point it is then well above the 100% mark when it raises the tank back up to start printing. So when I switched tanks I also started raising the tank to the level position before adding resin to the 100% mark. I have also watched closely for any sign of splashing and seen none (e.g. no residual resin down the sides of the tank.

The first leak was a royal mess by the time I found it (it didn't drip into the printer that I could see, but I did have to unbolt the LCD so I could lift it out and clean the frame) so trying to identify a source was fruitless (I also assumed that I had punctured the FEP). In this second case all the resin was located within the bounds of the print area on the LCD (e.g. definitely not ran down the side and seeped under). After wiping the tank off I don't see any indication that it worked itself from inside out (e.g. bad FEP seal) either.

Once I clean the second tank up I'll be able to find out for sure, but I highly doubt a puncture as the only thing that has touched it is resin, water, and a cotton pad. So unless the print process itself is tearing/puncturing it I don't see how a puncture is possible.

Have you tried with water in the tank gently pushing up on the FEP film from the bottom? This should show in short order if the pulling action of the lifting build plate is opening a small hole that is otherwise not present. 

Yeah I did try that with the first tank. Not knowing how much force is really being applied during the lifting process, however, it's possible I didn't replicate it effectively.

MMU tips and troubleshooting...
Posted : 29/10/2020 9:11 pm
vintagepc liked
gnat
 gnat
(@gnat)
Prominent Member

Support's response was that FEPs are only guaranteed for 10 prints (seriously?) and (summarizing) I need to change the FEP more often. For the record, the first tank may have hit the 10 print mark, but the second tank was at 5 tops.

He also seemed surprised that I just left the tank in place between prints and topped off the resin as needed.

So I am supposed to change the FEPs, then remove the tank every couple of prints, and change the FEPs on a number of prints basis. He is also sending me some FEPs.

I'll play along, but given that there is no sign of spillage or a puncture I have to wonder if that will have any effect...

In talking to him, however, I did notice a pattern. With the first failure I had a print fail in a "printed, but not correctly" way. I ran my finger and a plastic spatula against the FEP and didn't feel anything so I tried the print again. It failed, but in a worse manner that left part of the print stuck to the FEP. When I removed the tank to clean that up is when I found the leak. For the second tank I had a print fail yesterday where everything came out of the resin, but there was warpage and the parts had pulled away from supports in some places. I figured that the orientation was the problem (they weren't parallel with the platform, but not far off) so I resliced, tried again, and ended up with the failure that required me to empty the tank.

So I'll keep some metrics for awhile and see if that really is a pattern or not where it soft fails before a hard fail and leak.

MMU tips and troubleshooting...
Posted : 29/10/2020 11:50 pm
vintagepc
(@vintagepc)
Noble Member

Interesting idea - maybe try a "print" with coloured water instead of resin?

Could help narrow down whether it's the stickage/peeling or something else

... It occurs to me that if the print sticks really well, the FEP might "snap" loose when peeling prints, and maybe flick some resin out of the tank. But then you should see it in more places than just under the tank...

 

 

Posted : 30/10/2020 2:14 pm
gnat
 gnat
(@gnat)
Prominent Member
Posted by: @vintagepc

Interesting idea - maybe try a "print" with coloured water instead of resin?

Could help narrow down whether it's the stickage/peeling or something else

Interesting idea. I'm not keen to "waste" prints now that I know the FEP has such a supposedly short lifespan. The auto leveling isn't working for me so he wanted me to video the process of going through the calibration and then the resulting failed prints as well as how I manually level the platform, but I'm not keen to mess with what's working and waste prints. In addition to the FEP, the LCD has a lifespan too.

... It occurs to me that if the print sticks really well, the FEP might "snap" loose when peeling prints, and maybe flick some resin out of the tank. But then you should see it in more places than just under the tank...

While possible in theory, I would think it would have to be extremely low (well below the min level) before that would be possible. I would expect the volume and mass of the resin in the tank to dampen such forces to the point that escape velocity wouldn't be possible. Regardless, there are no signs of splashing.

The support person did mention that he has heard of people using nFEP on other printers and supposedly getting longer life spans from it. I looked it up and it claims to not stick to the cured resin as strongly, but it is double the cost of what Prusa wants for their FEP sheets and you still have to cut and punch out the holes.

MMU tips and troubleshooting...
Posted : 30/10/2020 2:59 pm
vintagepc
(@vintagepc)
Noble Member

Yeah, the consumables (FEP, display, resin, LED) is what's really put me off of SLA in addition to the resin health concerns. Price at one point too, but there are now decent sub-500 offerings out there. 

If the water test also leaks you might be able to see where the leak starts if you put a thin sheet of paper towel or newsprint paper between the vat and the display; and then check it occasionally for wetness - should be much easier since the water is clear and you don't have to look through murky or opaque resin. 

Posted : 30/10/2020 5:03 pm
gnat
 gnat
(@gnat)
Prominent Member

So I replaced the FEPs on both tanks last night. The second tank I didn't bother cleaning first, just wiped the worst of the residual resin out of the way and then cleaned it up while the FEP was off.

The first one, however, was cleaned up on Saturday and then spent Sat night and parts of Sunday full of clear water. Each time I emptied it I saw no sign of residual resin (e.g. no oily sheen on the surface) so I had no reason to suspect any resin left running around, but it turns out I was wrong.

When I removed the retaining plate from the tank I found a fair bit of resin underneath it and even more in the recessed area under the FEP. I did notice while cleaning it that there were small amounts of resin between the bottom lip of the tank and FEP which I had to really work to get out. With the retaining plate in place, however, it didn't appear that the resin worked itself up higher.

Given this I'm suspecting that maybe that is what is happening. It's not a perfect seal between tank and FEP so the resin works it's way under and then wicks up under the retaining plate. It then finds exits from the FEP either around the edge or through the screw holes and can then get down to the LCD itself.

There is a fair bit of room between the sides of the plate and the inner sides of the tank. Would there be a negative effect (other than the mess to deal with when changing the FEP) to putting a small bead of silicon sealer inside the tank around the edge?

MMU tips and troubleshooting...
Posted : 30/10/2020 5:27 pm
Xulkal
(@xulkal)
Trusted Member
Posted by: @gnat Given this I'm suspecting that maybe that is what is happening. It's not a perfect seal between tank and FEP so the resin works it's way under and then wicks up under the retaining plate. It then finds exits from the FEP either around the edge or through the screw holes and can then get down to the LCD itself.

There is a fair bit of room between the sides of the plate and the inner sides of the tank. Would there be a negative effect (other than the mess to deal with when changing the FEP) to putting a small bead of silicon sealer inside the tank around the edge?

In my own water tests, this is exactly what is happening. As for the silicon sealer, I've had the same idea, but am concerned about how, not smooth that is going to be and thus any impact on the final result that will have. It's almost like this design is missing an o-ring & channel milled into the tank right by the edge of where the frame sits to get a good water tight seal.

Everything I can think of around this problem still comes back to the initial design of this tank/frame setup as being, well crap and prone to failures like this. Not really what I was expecting coming from Prusa.

I even went as far as having a CNC shop mill me a 3mm frame bracket out of 304 stainless steel as flat as they could to remove the possibility that it's just the sheet metal frame flexing. The result was a big money sink as it was still leaking even with that new frame bracket.

This post was modified 4 weeks ago 3 times by Xulkal
Posted : 30/10/2020 5:32 pm
gnat
 gnat
(@gnat)
Prominent Member
Posted by: @vintagepc

Yeah, the consumables (FEP, display, resin, LED) is what's really put me off of SLA in addition to the resin health concerns. Price at one point too, but there are now decent sub-500 offerings out there. 

I knew about (and didn't like) the low LCD lifetime, but I was prepared for that. The LEDs are another known thing, but should have a decent life. The FEP was the surprise and honestly I may well have rethought it if I understood that changing the FEP was so frequent. Still better than the Form 3 that has a limited (4l I think) life span for their tanks and you have to replace the entire tank for what you can buy a cheap resin printer for...

If the water test also leaks you might be able to see where the leak starts if you put a thin sheet of paper towel or newsprint paper between the vat and the display; and then check it occasionally for wetness - should be much easier since the water is clear and you don't have to look through murky or opaque resin. 

I wonder how difficult it would be to do a custom firmware that would not fire the LEDs but go through the rest of the motions. That should not impact the life expectancy of the any of the parts and would mean that it could be watched with the hood up for a better view.

If the pulling of the resin being pulled off the FEP is part of the issue, however, that would not be replicated though.

MMU tips and troubleshooting...
Posted : 30/10/2020 5:42 pm
vintagepc
(@vintagepc)
Noble Member

I wonder if you could just disconnect the power to the LED.

Even if the leak does not happen, it will help narrow down the problem area - though water is far less viscous so should wick more easily through tiny openings over time if it is purely motion related.

 

Posted : 30/10/2020 6:00 pm
Xulkal
(@xulkal)
Trusted Member

So here's exactly why this tank/frame/fep design is a problem The attached image is from my most recent brand new Prusa replacement FEP and water test that failed. As you can see in the attached image, the FEP stretches, we know this as this is what ensures a tight flat surface to print on. Well, the pre punched screw holes also stretch, and in some cases enough to reach the edge of the frame and provide the perfect spot for water to leak through. This FEP was installed using a traditional cross tightening pattern as mentioned in the FEP installation documents. There is 100% absolutely nothing we as consumers can do to prevent this, it's fully up to shear dumb luck / material consistency at the punch holes as to if they stretch enough during installation to leak or not.

So, 100% this is a bad tank/frame/fep design and needs to be reworked for a SL1s or SL2 model, otherwise this will be a constant issue for users.

Posted : 30/10/2020 6:00 pm
gnat
 gnat
(@gnat)
Prominent Member
Posted by: @vintagepc

I wonder if you could just disconnect the power to the LED.

You don't want to get me started on my experience with UV power and Prusa support...

Short answer, no you can not do that as it measures the voltages on the LED array (guessing to see if it needs to be replaced).

MMU tips and troubleshooting...
Posted : 30/10/2020 6:59 pm
gnat
 gnat
(@gnat)
Prominent Member

@ryan-c29

I didn't notice when I removed them last night, but that is pretty much exactly what both my FEPs look like. Every hole except the 4 corners are elongated. Messy as it may be, it really makes me suspect that a silicone bead might be the way to go.

MMU tips and troubleshooting...
Posted : 30/10/2020 7:04 pm
Xulkal
(@xulkal)
Trusted Member
Posted by: @gnat

@ryan-c29

I didn't notice when I removed them last night, but that is pretty much exactly what both my FEPs look like. Every hole except the 4 corners are elongated. Messy as it may be, it really makes me suspect that a silicone bead might be the way to go.

I'll let you be the first to test that =p I've got an email into support about this design failure. I've gone through 8 new FEPs at this point and kinda ticked off. Best success I had was by using a FEP from another company that was oversized and had no holes punched. I punched the corners first, tightened those down to stretch it over the tank, then punched the rest of the holes after stretching, no leaks from that method.

This post was modified 4 weeks ago by Xulkal
Posted : 30/10/2020 7:09 pm
Page 1 / 9
Share:

Please Login or Register