Access Panel Seal

[DanH]

Sometimes I have an idea which actually works out well. Amazing, but true.

Subject is the hatch kit for the forward top skin of a 7, 9, or 10. The OP-43 installation drawing suggests forming a proseal gasket by using a piece of plastic wrap. I tried it and wasn't satisfied; too many surface irregularities because the plastic tends to wrinkle when trapped between the hatch and the support ring.

So I stripped it out and started over. This time, I sprayed three coats of ordinary PVA mold release on the back side of each hatch panel. Cure the PVA overnight, then mix proseal, spread it generously on the support ring, and screw down the hatch. Let it cure two or three days, then skim the squeeze out from the upper surface with a razor blade. When you remove the screws, the hatch will pop right out, leaving the PVA adhered to the proseal like a sheet of plastic. It won't peel off...but PVA is water soluble. Take the skin panel to the nearest water hose and it will rinse right off.

The nice thing is how easy it is to correct if not perfect. Got an air bubble cavity or other surface imperfection? Shoot another round of PVA on the hatch, slop more proseal on the clean dry previous, and screw it down again. The new proseal will adhere to the previous application just fine.

I buffed the assembled cover plate and skin with a red scotchbrite pad, which leaves a filler in the gap. A quick razor trim will clean up the ID of the ring.
.

 

[terrye]

Great idea as usual Dan. But what if the forward skin panel is already riveted on and voltage regulators, wiring etc. are mounted on the ribs below the access panel. How to remove the PVA then?

 

[skelrad]

When making a seal like this, do you screw the panel down tight, or do you leave it backed off a tiny bit so that the gasket is just a hair thicker and has to compress a little in its final state when the acces panel is screwed down tight?

 

[rvbuilder2002]

Sometimes I have an idea which actually works out well. Amazing, but true.

Subject is the hatch kit for the forward top skin of a 7, 9, or 10. The OP-43 installation drawing suggests forming a proseal gasket by using a piece of plastic wrap. I tried it and wasn't satisfied; too many surface irregularities because the plastic tends to wrinkle when trapped between the hatch and the support ring.

So I stripped it out and started over. This time, I sprayed three coats of ordinary PVA mold release on the back side of each hatch panel. Cure the PVA overnight, then mix proseal, spread it generously on the support ring, and screw down the hatch. Let it cure two or three days. When you remove the screws, the hatch will pop right out, leaving the PVA adhered to the proseal like a sheet of plastic. It won't peel off...but PVA is water soluble. Take the skin panel to the nearest water hose and it will rinse right off.

The nice thing is how easy it is to correct if not perfect. Got an air bubble cavity or other surface imperfection? Shoot another round of PVA on the hatch, slop more proseal on the clean dry previous, and screw it down again. The new proseal will adhere to the previous application just fine.

I skimmed the squeeze out with a razor and buffed off with a red scotchbrite pad. It left a nice seam filler around the edges. A quick razor trim will clean up the ID of the ring.
.

Great idea
It was actually considered as a method to put in the instructions but it was abandoned because it seems PVA is only readily available in 1 gal jugs and it’s not likely many people would have it on hand.

 

[DanH]

... if the forward skin panel is already riveted on and voltage regulators, wiring etc. are mounted on the ribs below the access panel. How to remove the PVA then?

A sponge and a bucket to rinse it out should get 'er done.

When making a seal like this, do you screw the panel down tight, or do you leave it backed off a tiny bit so that the gasket is just a hair thicker and has to compress a little in its final state when the acces panel is screwed down tight?

I screwed it down tight. It's a good question. I'll test the resulting seal tomorrow with running water.

It was actually considered as a method to put in the instructions but it was abandoned because it seems PVA is only readily available in 1 gal jugs and it’s not likely many people would have it on hand.

...in an aluminum shop

https://www.fibreglast.com/product/PVA_Release_Film_13

 

[cvairwerks]

Known as “Form in place” gaskets in the F-16 world. Thickness is controlled by using a washer at each fastener location. Superglue them on prior to spraying any type of mold release on the removable panel. Typically washer thickness is .032” or .063”. Trim as necessary after curing. Repairs can be done by cleaning damaged area and adding new sealant. Use a teflon type tape on the door instead of mold release for fast cleanup. Lightly sand after curing to blend out any extra contour.

 

[Desert Rat]

A sponge and a bucket to rinse it out should get 'er done.



I screwed it down tight. It's a good question. I'll test the resulting seal tomorrow with running water.



...in an aluminum shop

https://www.fibreglast.com/product/PVA_Release_Film_13

Dan- I too am unimpressed with all the wrinkles left by the Saran Wrap process.

Do you know the difference between the PVA #13 in your link and PVA #10? I ask because my typical go to supplier for stuff like this is US composites in Florida and they sell PVA 10 for a whopping $5.90 a pint or if you want to splurge, you can get a quart for $8.60

 

[mburch]

It was actually considered as a method to put in the instructions but it was abandoned because it seems PVA is only readily available in 1 gal jugs and it’s not likely many people would have it on hand.

For those of us lucky enough to have a TAP Plastics store nearby, they'll sell you individual pints of PVA.

16-ounce containers are also plentiful on Amazon: https://www.amazon.com/dp/B078XH3NYR/

 

[rvbuilder2002]

...in an aluminum shop

https://www.fibreglast.com/product/PVA_Release_Film_13

Yea
I used PVA, quite a bit actually, for fiberglass prototyping and production mold work. The place I bought supplies only had it in 1 gallon jugs. I never thought to check anywhere else for a smaller quantity package.

 

[rapid_ascent]

They have it at ACS too. Maybe not the best deal in the world but its there.

https://www.aircraftspruce.com/catalog/cmpages/moldpva.php?clickkey=33895

I may have to give this a try. I've holding off on making my cutouts.

Another idea I had is you can now have elastomeric materials 3D printed. I'm not sure of the cost of these and it would probably take a lot more work to figure out the exact dimensions. Maybe someone with 3D printing knowledge could comment on this approach.

 

[DanH]

Known as “Form in place” gaskets in the F-16 world. Thickness is controlled by using a washer at each fastener location. Superglue them on prior to spraying any type of mold release on the removable panel. Typically washer thickness is .032” or .063”. Trim as necessary after curing. Repairs can be done by cleaning damaged area and adding new sealant. Use a teflon type tape on the door instead of mold release for fast cleanup. Lightly sand after curing to blend out any extra contour.

Thickness control with a washer might be possible on the RV if the builder installed a spacer ring between the skin and the nutplate ring so the finished hatch plate would be flush with the skin.

Tape won't work here due to bridging at the screw dimples. The PVA approach allows the proseal to form down in the nutplate countersinks and even in the screw threads.

Do you know the difference between the PVA #13 in your link and PVA #10?

I doubt there's much practical difference, but you'll need to try them to see.

Partall brand PVA is available in three levels of solids. A high solids PVA might only require one coat and thus be faster for large mold work. Never tried any myself.

Handy stuff, and lasts a long time on the shop shelf. Shoot it with a cheapo touch up gun.

 

[DanH]

Yea
I used PVA, quite a bit actually, for fiberglass prototyping and production mold work. The place I bought supplies only had it in 1 gallon jugs. I never thought to check anywhere else for a smaller quantity package.

With as much proto experience as you have, I assumed you bought yours in 5 gallon buckets

 

[Reflex]

Love the idea. My idea was to use plumber's putty. It's clean, fast, inexpensive, readily available, and easy to clean up. I like Dan's idea better.

I'm a bit slow as I learn slowly and retain little. So, I'd like to be able to remove the gasket in case of error in making it, wear, leaks, or a redo if I don' t get it quite right.

Would spraying the ring itself allow the formed gasket to be easily removable?

Fred

 

[DanH]

Would spraying the ring itself allow the formed gasket to be easily removable?

Sure, but you don't want to. It will be thinner than tissue in places.

Fully adhered polysulfide sealant is easily removed.

 

[dstates]

I buffed the assembled cover plate and skin with a red scotchbrite pad, which leaves a filler in the gap.
.

Dan,

I like this idea a lot... can you explain the comment you made a the end of your post about leaving a filler in the gap? I'm not following what or when you did this.

One more question... could the PVA be applied with a fine misting spray bottle?

Thanks!

 

[gbitzer]

I have 3 of these forward hatch openings on my 6. I can't imagine trying to work on the avionics without those hatches on RVs with a slider canopy. Leaks have been a challenge when flying through heavy rain. I'll give this a try this out once I determine how to protect that painted exterior skin. Thanks for sharing DanH.

 

[RONSIM]

Off and on often, so I used silicon grease

Scuba type, and it worked fine.

 

[mburch]

can you explain the comment you made a the end of your post about leaving a filler in the gap? I'm not following what or when you did this.

I'm not Dan, but I took this to mean that he sanded all the squeeze-out on the exterior side down level with the skin, leaving a tiny ring of cured proseal in the gap between the skin and access panel.

 

[georgemohr]

What do you all think about using mould release wax here?

 

[dmattmul]

Normally used

What do you all think about using mould release wax here?

Used with PVA before the PVA is applied or at least I do. Just helps the release better. Works by itself also. (Mold release)

Next time will try SIKAFLEX.

Anyone tried these? (Sealed nutplates)

 

[DanH]

Dan... can you explain the comment you made a the end of your post about leaving a filler in the gap? I'm not following what or when you did this.

See photo below (Matt nailed it).

One more question... could the PVA be applied with a fine misting spray bottle?

I dunno. Try it!

Scuba type, and it worked fine.

I think silicone grease is great idea. My little water test last night was inconclusive; the panel isn't riveted yet, so it has a lot of little dimpled holes in it

What do you all think about using mould release wax here?

Not needed. PVA easily releases from clean aluminum. When you pop the panel out, the PVA film will remain in the opening.

 

[Snowflake]

When making a seal like this, do you screw the panel down tight, or do you leave it backed off a tiny bit so that the gasket is just a hair thicker and has to compress a little in its final state when the acces panel is screwed down tight?

Underrated comment right here. Gaskets like this require compression to seal, unlike the Proseal joints in your fuel tank that are bonded to each surface to provide sealing. After you tighten the screws down as this gasket is formed, the sealant will weep and relieve the pressure. That leaves a gasket that fits perfectly, but you can't compress it later on because you already tightened the screws during molding.

I suspect you'll find that the fit is wonderful right now, but any slight movement or flex will cause it to allow water in over time.

 

[DanH]

I suspect you'll find that the fit is wonderful right now, but any slight movement or flex will cause it to allow water in over time.

I agree, which is why the silicone grease seems like a good idea.

Plan B, not quite the same thing, but close; I have two EDIS modules and two Megajolt controllers on a side panel below the forward baggage floor of my -8. The baggage door is notoriously leaky, so the electronics have a 4 mil plastic sheet "rain shield" positioned above them. Think of it as a umbrella.

 

[ColoradoSolar]

I 3d printed mine out of TPU. It was a pain to get them off of the build plate but they seem to have come out ok. I am still building so I am not sure how well they will hold up. I can publish the files if someone is interested.

 

[Lucmac]

I 3d printed mine out of TPU. It was a pain to get them off of the build plate but they seem to have come out ok. I am still building so I am not sure how well they will hold up. I can publish the files if someone is interested.

I’m interested! Thanks.

 

[Draker]

Wow, these are great looking. Makes me feel embarrassed of my ugly but by-the-plans proseal gaskets. Doesn't seem worth it to re-do but very nice!

 

[Snowflake]

I agree, which is why the silicone grease seems like a good idea.

It's a start, but even the silicone grease will eventually weep out under pressure and leave you with a great fit but not a great seal.

I only dwell on this as I made a similar mistake recently at my day job, when trying to seal a bolted HDPE enclosure with RTV... Completely forgetting that (a) nothing sticks to HDPE, and (b) you can't fix (a) without putting pressure on a gasket... The RTV looked great when done, but pinhole leaks appeared immediately when the tank was filled... And it was only a 1 gallon tank.

Plan B, not quite the same thing, but close; I have two EDIS modules and two Megajolt controllers on a side panel below the forward baggage floor of my -8. The baggage door is notoriously leaky, so the electronics have a 4 mil plastic sheet "rain shield" positioned above them. Think of it as a umbrella.

That's a good belt-and-suspenders approach... No harm in a backup that doesn't weigh much.

 

[rapid_ascent]

Eric,

I'd be interested in the files. I might give it a shot. It's experimental aircraft after all.

 

[ColoradoSolar]

Eric,

I'd be interested in the files. I might give it a shot. It's experimental aircraft after all.

I will get the files ready in the next few days

 

[Fred.Stucklen]

I also pro-sealed the doubler to the skin so that water couldn't sneak under it..

I 3d printed mine out of TPU. It was a pain to get them off of the build plate but they seem to have come out ok. I am still building so I am not sure how well they will hold up. I can publish the files if someone is interested.

 

[ColoradoSolar]

I also pro-sealed the doubler to the skin so that water couldn't sneak under it..

I was going to do that but I forgot when I actually riveted the doublers on. Don’t feel like drilling out all those rivets

 

[Lucmac]

I will get the files ready in the next few days

Hi Eric. Did you post the files somewhere?

 

[Snowflake]

I was going to do that but I forgot when I actually riveted the doublers on. Don’t feel like drilling out all those rivets

I will say from experience that the pain of drilling those rivets out, in such an easily accessible location, will be far outweighed by the pain of replacing avionics that get wet later on... And when it happens you'll be drilling the rivets out to proseal it so it doesn't happen again.

 

[ColoradoSolar]

Hi Eric. Did you post the files somewhere?

Sorry Thanksgiving got in the way. Here is the link to it on thingiverse. It contains an STL file that one can send directly to the slicer and the Fusion 360 in case someone wants to modify anything.

 

[ColoradoSolar]

I will say from experience that the pain of drilling those rivets out, in such an easily accessible location, will be far outweighed by the pain of replacing avionics that get wet later on... And when it happens you'll be drilling the rivets out to proseal it so it doesn't happen again.

Yea you're right. I will put it on my ever growing todo list

 

[gyoung]

Sorry Thanksgiving got in the way. Here is the link to it on thingiverse. It contains an STL file that one can send directly to the slicer and the Fusion 360 in case someone wants to modify anything.

Thanks for posting the files. I have no experience with TPU and see that the Ninjatek brand you used is 4x more $ than most other brands. Some reviews claim it's the most flexible which I can see being desirable for a thin gasket. I've wanted to try out TPU anyway so don't mind the extra cost if it's worth it. Any thoughts on brands or suggestions on printing TPU? Thanks.

 

[dmn056]

For form-in-place seals, a Tinnermans washer seated between the panel and the flange to which it is attached will provide about the right spacing to allow the seal to be compressed later. The washer needs to be coated with your mold release agent before installation.

Also, nearly any spray film lubricant that doesn't include silicone will work as a release agent.

 

[ColoradoSolar]

Thanks for posting the files. I have no experience with TPU and see that the Ninjatek brand you used is 4x more $ than most other brands. Some reviews claim it's the most flexible which I can see being desirable for a thin gasket. I've wanted to try out TPU anyway so don't mind the extra cost if it's worth it. Any thoughts on brands or suggestions on printing TPU? Thanks.

I don't have much experience with TPU and NinjaTek was the most flexible I could find when I was looking. I only bought a 500g roll. In my trials I tried a 0.2mm nozzle to get very thin layers but that didn't work because it just jammed. So I ended up printing with a 0.4mm nozzle. The main thing that I found is it sticks very well to a PEI build plate and you need to be careful removing it from the plate or it will stretch. I ruined on print doing that. I am by no means a 3d printer expert so others may chime in with better ways to get TPU not to stick so hard.

 

[rapid_ascent]

Well I cut my access panel cutouts today. It definitely took some time getting the gap the way I wanted it but I think it looks pretty good now. Dimpling was done after fitting BTW.


Dan, I can't really tell from your pic what happens in the area around the screw holes. Is there actually any gasket in that area?


Eric, I'm curious why you decided to make the screw holes so large. With the holes that size it seems you aren't sealing around the screw. Did you try making the holes smaller?

I think I'm going to give Dan's approach a shot. I'll get to buy another chemical I haven't tried before that way.

 

[ColoradoSolar]

Well I cut my access panel cutouts today. It definitely took some time getting the gap the way I wanted it but I think it looks pretty good now. Dimpling was done after fitting BTW.


Dan, I can't really tell from your pic what happens in the area around the screw holes. Is there actually any gasket in that area?


Eric, I'm curious why you decided to make the screw holes so large. With the holes that size it seems you aren't sealing around the screw. Did you try making the holes smaller?

I think I'm going to give Dan's approach a shot. I'll get to buy another chemical I haven't tried before that way.

I did start with the holes smaller but I enlarged them so the dimples would fit better. You can always modify the Fusion 360 file to make the holes smaller if that fits your need better.

 

[Ron B.]

Once the seal is cured, do we paint right over the Proseal that fills the gap or does it need to be removed first. I would thing the paint would chip at the edges whereas if we paint on the soft Proseal. Thanks and Happy Holidays, Ron

 

[mburch]

I followed Dan's technique and my access panel seals turned out really nice:

More pictures here: https://www.rv7blog.com/2024/02/27/access-panel-seals/

 

[Lucmac]

Sorry Thanksgiving got in the way. Here is the link to it on thingiverse. It contains an STL file that one can send directly to the slicer and the Fusion 360 in case someone wants to modify anything.

Did you use 85 or 95 shore hardness TPU. Also, how’s the final thickness worked out? Is the panel flush with the fuse or biased high or low?

 

[TASEsq]

For form-in-place seals, a Tinnermans washer seated between the panel and the flange to which it is attached will provide about the right spacing to allow the seal to be compressed later. The washer needs to be coated with your mold release agent before installation.

Also, nearly any spray film lubricant that doesn't include silicone will work as a release agent.

To allow the sealant some thickness to be compressed later, would it work to just not tighten down the screws all the way when forming the sealant?

 

[ColoradoSolar]

Did you use 85 or 95 shore hardness TPU. Also, how’s the final thickness worked out? Is the panel flush with the fuse or biased high or low?

The TPU filament I used is NinjaFlex and is 85 shore hardness. I don't remember the final thickness, I will have to measure them. But the panels will sit slightly proud because the panels and the skin are the same thickness and the backing plate is riveted to the skin.

 

[dmn056]

To allow the sealant some thickness to be compressed later, would it work to just not tighten down the screws all the way when forming the sealant?

Yes, it would. However, for me it seems easier to get a uniform seal thickness by using the spacers. It also makes it easier to remove the fasteners after sealant curing without tearing the seal around them. This is a yet another YMMV situation.

 

[RVDan]

In my work, we used the Chemseal low adhesion removable panel sealant CS3330B1. PRC has a similar product. You apply when putting the panel on, and when you reopen the panel peel the sealant off and apply new to reseal the panel. No leaks.

 

[DanH]

In my work, we used the Chemseal low adhesion removable panel sealant CS3330B1. PRC has a similar product. You apply when putting the panel on, and when you reopen the panel peel the sealant off and apply new to reseal the panel. No leaks.

Sounds like a winner...

 

[MElstien]

Dang, Now I have to clean mine and re-do it. Thanks Dan