plastic pressure vessel

[davelin]

Hi,
I've been given a "legacy project" which consists of designing a pressure vessel, meant to hold 1.5 bar absolute pressure of N2 or 0.5 absolute air pressure.

The are 3 design constraints that make this task problematic:
1) The vessel must be made of plastic or glass because of RF issues.
2) The vessel is meant to hold a round PCB, which has a 250 mm diameter. which is an issue because of (3)
3) The vessel will be placed in an oven with an internal dimension of 278 mm. the oven will reach a max of 60 deg C.

This means, in order both to fully encase the PCB and also fit in the oven, the ID has to be 250, with an OD of around 275 mm. which gives it a wall thickness of about 12.5 mm.

my second option is to clamp the pcb between 2 cylinderical , and equalize the pressure between them (see pic). doing this allows my minimal ID to reach 180 mm (WT of up to 35 mm). and the PCB actually has through holes meant for bolts.

after all this, my questions are:
*do you think this is feasible? I'm wary of using plastic for pressure holding applications, especially with heat involved.
*If so, Which material would you use? I was thinking polycarbonate, since it tends to stretch rather than shatter when failing.
*Which WT do you think would be good? I've done the calculations, and obviously 35 mm WT is over designed. But in this case I would like to make this as robust as possible.

In general - I've looked through the PVHO safety standards, which speaks more about acrylic viewing windows. But besides that there are very few references on applications like these. I've also found some OTS options, but they don't fit my constraints as they are today.

Thanks in advance.

 

[EdStainless]

At 60C there are many composites that will have good usable strength.
I have seen various fiberglass/epoxy formulas used in this temp range.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Rputvin]

Is the pressure 0.5 barg or a range of -.5 barg to .5 barg - are you designing for internal pressure only?

If you need this only to contain the nitrogen atmosphere at slightly above 1atm while the PCB is in the oven have you considered introducing a nitrogen atmosphere in the oven and avoiding the intermediary vessel?

Are there fittings or other features on this vessel to allow you to add pressurized nitrogen? If so reinforcement of the nozzle attachment might be a concern for plastic.

There's likely a fiberglass reinforced plastic (FRP) material that will meet your needs, or a higher density machined plastic such as UHMWPE or Delrin which have higher strength and temperature ratings ranging well above your 60°C mark. Standard polyethylene (PE) tanks are typically rated for 48-50°C (120°F) continuous service with a few noted here and there that intermittent 60°C (140°F) is acceptable - how long is the PCB in the oven, and how long does the PCB need a nitrogen atmosphere before and after its time in the oven? Is the oven temperature tightly controlled, or are there possibilities that some of the surface temperatures of the vessel will exceed the 60°C mark by being exposed to burners/elements or uneven heating zones?

 

[davelin]

Thanks for your reponse!
Answering your questions in order:
*The pressure range is indeed -0.5 barg to 0.5 barg.
*Unfortunately the oven isn't meant for a pressurized environment.
*Yes, the vessel will have to be fitted with ports to allow the nitrogen to enter, as well as ports for cables. I've also though about this issue, beside placing a steel cover (which is problematic at the moment), is there a known solution for this issue?
*The oven is very tightly temperature controlled so it should be evenly heated, but it is possible to add additional temperature sensors.
*In terms of time- essentially the users want the system to work as long as possible, slightly varying the conditions according to need. if needed, it's possible to limit this time.
*The nitrogen atmosphere just needs to be for the time of the heating (meaning it will be evacuated before or after.

Regarding materials, I hadn't thought of Delrin. Do you think this can fit this particular usage? I have found a report from the Princeton Plasma Physics Laboratory about the usage of a polycarbonate vacuum vessel in some testing- Link. Do you think this would be applicable here?

and regarding the WT, do you think it will be wiser to do the "thinner" full cylinder configuration? or clamp the PCB between 2 "thicker" pressurized vessels (assuming I'm able to properly seal them)?

Thank you

 

[EdStainless]

You stand zero chance of doing this without a glass filled product.
Find someone to make you a custom fiberglass/epoxy tube and be done with it.
They can even extend one end and put threaded ports in it for you.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[Rputvin]

Is this a "legacy" project meaning it had a solution but it needs to be re-engineered because the solution can't be recreated for whatever reason - or is this a project that has had many owners who never could get a solution figured out and now it's been passed to you to allow you to spin your wheels on it for a while before it stagnates, falls off, and gets reassigned when there's another unwitting engineer to pick it up again?

I agree with Ed - go for a reinforced material if you're going for internal and external pressure, as well as long times exposed to heat. My experience with UHMW and Delrin at higher temps is using blocks for isolation and replaceable slider pads, not as a thin-walled containment vessel.

With regards to wall thickness - once you have a material selected you probably want to minimize the wall as much as possible since you're sticking an insulator between your heat source and the part needing heat. Has there been any analysis on this, or is this still R&D and that's why "as long as possible, slightly varying conditions" is the answer?

I still think you'd be better off with something like a small loop of piping that has a heater and a fan to circulate nitrogen at a controlled temperature, rather than trying to engineer a plastic vessel to fit in an oven and let it set there all day. There's a reason ovens and furnaces are built to create a vacuum or inert gas atmospheres internally, placing a part in a vessel to place in an oven is taking a step in the wrong direction.

 

[davelin]

I agree with everything you guys are saying, and I've brought this up many times.
When I say "legacy", I mean it is a project that started way before I arrived and now I'm stuck with the constraints that prior (and some still in the company) engineers left for me. This include the dimensions of the PCB, the oven and the pressure (and heating) conditions.
I've actually already found an off the shelf solution but for smaller dimensions that what we have. But since my company really wants initial results for this system, I'm stuck with these constraints until we can show something.

I would be happy to go to reinforced materials, but have no experience with this. Most companies I've found make very large scale vessels. Do you have I direction I can go to in order to find someone who can manufacture this?

 

[EdStainless]

Most of the people that make FRP pipe will custom wind a piece of material for you.
You will pay for the mandrel and the part, but it will only be a fraction of the cost of buying a real furnace that can handle moderate pressure and vacuum.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed