Physics problem (not homework....work work)

For the discussion of the sciences. Physics problems, chemistry equations, biology weirdness, it all goes here.

Moderators: gmalivuk, Moderators General, Prelates

sgt york
Posts: 331
Joined: Mon Oct 01, 2007 7:21 pm UTC
Location: Chapel Hill, NC
Contact:

Physics problem (not homework....work work)

OK, I have a problem with something we built. I don't get it.

First, look at the picture. There are 3 components; the body, the lid, and the insert. The insert fits into the small hole in the center of the lid. The lid then fits into the body. You can see other lids fit into the body, only one is removed and turned upside down for you to see.

The body is a block with 6 holes bored about 3/4 through it to form wells.

The lid has 3 diameters to it: One is the outer diameter; this is the lip that fits over the well. The next in is the inner diameter (blue); this is supposed to be equal to the diameter of the well in the body (also blue). This part of the lid is tapered; it's about 0.5mm wider at the top than at the bottom, allowing the user to get a nice, tight fit. The last is the insert diameter (red), equal to the diameter of the insert (red). This is tapered in the same way (just less so).

The insert (or our purposes here) is just a cylinder with a lip to prevent it from being put too far in the lid.

When it was constructed, it fit together well. The insert fit quite snugly into the lid and the lid fit snugly into the body. It was a tight fit, but only finger tight; It was easy to set up and break down using only my hands, but would hold tight if moved, shaken, turned over, etc.

This all changed upon autoclaving it. (one of the design requirements was to withstand 121C without deforming irreversibly). Now, the lid is too big to fit into the body, and the hole in the lid is too big to hold the insert. Each expanded by a bit less than 1mm. This was after the whole thing cooled down to room temperature.

This is all made from white acetal (Delrin). The insert is disposable, made of PVC plastic, and not autoclaved. For the curious, it is designed as part of a cell culture device. The tubing you see is different stuff in different places. I don't think it could affect this, though.

I know that the lids should be treated as rings; they will expand when heated in the way I saw here. But shouldn't the body expand the same way (it didn't; it maintained its form better, changed within tolerance)? Also, shouldn't it revert when it cools? Acetal is supposed to hold its shape really well up to temps around 200C.

And no, I'm not an engineer. I'm wingin' this. Totally.
Attachments
widget.jpg (78.26 KiB) Viewed 2474 times

Charlie!
Posts: 2035
Joined: Sat Jan 12, 2008 8:20 pm UTC

Re: Physics problem (not homework....work work)

Hmm, the softening point appears to be not too much higher than autoclave - around 150 C. Was there any pressure on the lid or box that might have deformed it while it was hot?
Some people tell me I laugh too much. To them I say, "ha ha ha!"

sgt york
Posts: 331
Joined: Mon Oct 01, 2007 7:21 pm UTC
Location: Chapel Hill, NC
Contact:

Re: Physics problem (not homework....work work)

Charlie! wrote:Hmm, the softening point appears to be not too much higher than autoclave - around 150 C. Was there any pressure on the lid or box that might have deformed it while it was hot?

No, pretty standard autoclave setup. Sitting in a metal basket, wrapped lightly in foil. That a/c isn't rated for biohazard waste, so it probably can't get up to 150C.

twerpoid
Posts: 13
Joined: Wed Jul 16, 2008 2:35 am UTC

Re: Physics problem (not homework....work work)

Hypothesis1: Change in size post-autoclave is due to a change in moisture content.
Potential ways to test or explore this hypothesis:
- If you store the finished product in a high humidity environment for a while, does it fit better?
- measure exact length of scraps of similar dimensions before and after autoclave,
- or before and after storing in drybag vs humidity chamber

Hypothesis2: Change in size post-autoclave is due to stress-relief or other phase-change in bulk material.
Tests:
- measure exact length of scraps of similar dimensions before and after autoclave, several times
- Pre-autoclave bulk material to its final density/phase before machining to shape

Hypothesis3:
The difference in the change of size between the body and the lid is due to the geometry.
The lid is a ring that expands more easily, the body is more of a solid block.
Some mesh-grid plastic trays in the SC industry have lots of issues with warping,
unless they have enough mesh cuts in specific places.
Tests:
Design the body to a more uniform thickness by milling the back of the body around the holes,
ie make it look like a pressed sheet metal muffin tin, rather than than a slab with holes in it.

Other Workaround ideas
If the parts change shape irreversibly only once,
- then you may be able to get by with pre-autoclave before machining,
- or change size of holes such they work post-autoclave.
Try different materials.
Design different tapers to allow for greater range of dimensional stability.

sgt york
Posts: 331
Joined: Mon Oct 01, 2007 7:21 pm UTC
Location: Chapel Hill, NC
Contact:

Re: Physics problem (not homework....work work)

twerpoid wrote:(a lot of good ideas & thoughts)

Thanks for the ideas!

I'm trying repeated autoclaving first. I took the measurements and popped it in there for a short run; should be done in about 45'. I'll measure it hot, then again after it cools to see how it all changes. Hopefully, it only changes shape once due to stress-relief or phase-change as you suggested. Then, I can just re-mill the lids from pre-autoclaved material to fit the current chamber and go from there.

My reasoning here is that if it's humidity, I'm screwed. It's a tissue culture system, and it has to be used at near 100% humidity, and I live in NC so my lab is pretty humid. #3 would require a complete redesign; there is some plumbing inside (hollow areas connecting the wells).

If it's one of those or I need to change materials, I'll probably just switch sterilization techniques instead. I have chemical sterilizing equipment, it's just been mothballed and I need to get the reagents for it. Autoclave was the first choice because I have three right outside my door.

dic_penderyn
Posts: 35
Joined: Mon Apr 14, 2008 9:29 am UTC
Location: Merthyr Tydfil, Wales

Re: Physics problem (not homework....work work)

It seems to me that you have to build in some tolerances for the setup.
If you require a seal between the lid and body, could you not turn a groove on the lid and use an o' ring?

That way the o'ring will be able to take up any change in dimension.

Just a thought.

sgt york
Posts: 331
Joined: Mon Oct 01, 2007 7:21 pm UTC
Location: Chapel Hill, NC
Contact:

Re: Physics problem (not homework....work work)

O-rings: I can't find any that are compatible with tissue culture. Found some that make the claim, but according to some of the other people around here, they're not. They all leak stuff that's toxic to these cells. And yeah, I need tight tolerances. As close to gas tight as possible.

Temp changes : I nuked it 7 times, tracking the sizes of the three critical diameters after each cooling. It looks like the acetal is changing shape pretty much randomly. My current solution is to trash this piece and mill a new one for chemical sterilization instead of autoclaving. I might be able to salvage the base, but the lids are no good now, they won't hold the culture support. Chemical sterilizing is a bit of a pain in the ass (the sterilizer is in another building), but not that big of a deal. Experimental runs are weeks to months long, and I only need it sterilized between runs.

Anyway, thanks for the help, everybody.

Seraph
Posts: 342
Joined: Mon Jul 16, 2007 4:51 pm UTC

Re: Physics problem (not homework....work work)

My current solution is to trash this piece and mill a new one for chemical sterilization instead of autoclaving.

I guess it doesn't matter too much now, but Polyoxylmethylene (aka Delrin, POM) is fairly well know for being dimentionally unstable when machined. Was it annealed before you had it machined? That's pretty important with large POM parts. In general there are a lot of residual stress in the blank you start with, and if you don't get rid of it before machineing it'll come when the part it heated later.

sgt york
Posts: 331
Joined: Mon Oct 01, 2007 7:21 pm UTC
Location: Chapel Hill, NC
Contact:

Re: Physics problem (not homework....work work)

Seraph wrote:I guess it doesn't matter too much now, but Polyoxylmethylene (aka Delrin, POM) is fairly well know for being dimentionally unstable when machined. Was it annealed before you had it machined? That's pretty important with large POM parts. In general there are a lot of residual stress in the blank you start with, and if you don't get rid of it before machineing it'll come when the part it heated later.

Well, crap. Wish I had known that before. No, it was not annealed. It was machined straight from the blanks. Although I wouldn't call this large; the whole device is maybe 25 x 10 x 5 cm. On the large side for tissue culture, but I'd imagine small to average by most other standards.

The guys down at the shop said acetal would work, and I just assumed they knew what they were talking about. But it does matter still....I've been working on a grant until yesterday and have not tried to make the device again. If there is a way to do this and have it autoclave-tolerant, I'm willing to hold off on starting. It will save time in the long run.

So, do you have any suggestions? Teflon is out; some of the parts are pretty small and since PTFE is so soft, it won't machine well (PTFE was actually my first choice, but we had to abandon it for that reason).

Silicon-based things are out as well due to permeability issues (mostly "unknown with reason to believe it would be bad").

I need to stick with plastics, as most metals will interfere with what I'm doing. Al is a possibility, maybe. I'm doubtful, but I'd have to check. But anything with iron in it is out. Fe is a tightly controlled element in the experiment. And to a bio-type, tight iron control means below nanomolar level. Media goes through some pH swings, and leaching is an issue. Even if it doesn't cause any problems, I'd have to spend some time proving it doesn't cause any problems.

Yes, I know I may be SOL. But there's always chemical sterilization.

====
Edit: Do you have a reference for that? I'd like to take it with me when I head back down there.

oxoiron
Posts: 1365
Joined: Fri Jul 13, 2007 4:56 pm UTC

Re: Physics problem (not homework....work work)

sgt york wrote:I've been working on a grant...
You have my condolences.
I need to stick with plastics, as most metals will interfere with what I'm doing. Al is a possibility, maybe. I'm doubtful, but I'd have to check. But anything with iron in it is out. Fe is a tightly controlled element in the experiment.
Now, you have me curious. With what are you working?
"Whenever you find yourself on the side of the majority, it is time to reform (or pause and reflect)."-- Mark Twain
"There is not more dedicated criminal than a group of children."--addams

sgt york
Posts: 331
Joined: Mon Oct 01, 2007 7:21 pm UTC
Location: Chapel Hill, NC
Contact:

Re: Physics problem (not homework....work work)

oxoiron wrote:Now, you have me curious. With what are you working?
Bacterial infections. Give bugs iron, and they grow like crazy. As a result, most animals are really good at sequestering iron. One of the things I'm looking at is the ability of different bugs to potentially cause and exploit microhemorrhages as an iron source. A lot of that is already known, so it's not the main thrust of the project, but it is an important aspect and something I have to control for.

oxoiron
Posts: 1365
Joined: Fri Jul 13, 2007 4:56 pm UTC

Re: Physics problem (not homework....work work)

I was hoping you were looking at a specific enzyme. I always like hearing about new, iron-based biochemistry.
"Whenever you find yourself on the side of the majority, it is time to reform (or pause and reflect)."-- Mark Twain
"There is not more dedicated criminal than a group of children."--addams

sgt york
Posts: 331
Joined: Mon Oct 01, 2007 7:21 pm UTC
Location: Chapel Hill, NC
Contact:

Re: Physics problem (not homework....work work)

oxoiron wrote:I was hoping you were looking at a specific enzyme. I always like hearing about new, iron-based biochemistry.

Nah...nothing novel. Got some interesting scavengers from different bugs that may work together in cool ways, but nothing novel in itself.

Neat mechanisms, though. For example, some of the pyoverdines from Pseudomonas have such a high affinity for iron that they can actually strip it from transferrin. YOINK!