Polycarbonate Tops... a scientiffic analysis! - Page 2

theetruscan · 03-13-2003, 12:24 PM

Quote:

Originally posted by unregistered

never apologize for 'talking technical', unknown terms will serve as a stimulus to those with an interest
(others use the Back button)

Seriously, we know how to Google. And if I didn't want to learn anything I wouldn't be reading this. Don't worry about going over our heads, it increases the likelihood of eventual comprehension.

TerraMex · 03-13-2003, 01:17 PM

>My question is: couldn't you just use a heating process,
>like a microtorch?

Something like this? Or ...?

WinFlex · 03-13-2003, 01:39 PM

Nice to know taht people are curious about this sort of stuff. Regarding heat bonding, please read the link I posted earlier, about 4 post back. It is not as reliable as chemical bonding, and I imagine it would also be more complicated to accomplish!

bigben2k · 03-13-2003, 01:50 PM

(not) That's what I meant, but now that I think about it, why not use a hot tool to press open a hole, instead of drilling it? Isn't it the drilling action that creates the microfractures?:shrug:

WinFlex · 03-13-2003, 01:52 PM

Very interesting suggestion, bigben2k... the amount of deformation that would occour using a hot tool is probably unacceptable, however. Nevertheless, howe about drilling, then inserting a hot pin that is a couple of thousands wider than the drill? This would melt shut any big cracks and completely erradicate smaller ones!

On a sidenote, I took my block up to 150 psi last night (yeah, i know i'm crazy) and it's absolutely rock solid at 150 psi for over 12 hours now... nor preassure drop, no observable deformation, no cracks.

nicozeg · 03-13-2003, 03:08 PM

BillA, why you say that solvent wiping the threads dont work?

It seems to me that the solvent would penetrate by capilarity the micro fractures and make a good job sealing them. In that case the polycarbonate is going to be able of holding an incredible tensile stress without damage.

The source of that suggestion seems very reliable: a manufacturer of polycarbonate products.

TerraMex · 03-13-2003, 03:09 PM

Doesnt the drilling of the plastic also heats it up due to the friction ?

And if so, doesnt it mean that at least some (if not most) of the microfissures will be removed naturaly by that process?

BillA · 03-13-2003, 03:25 PM

nicozeg

I seriously doubt the polycarbonate mfgr said that wrt the dimensional tolerances needed for threaded connections,
and the same problem arises with 'flame' annealing

threading these plastics is rather pointless as there are easier/cheaper ways to affix connectors that are also more reliable

what is the goal ?
simple is always better

go back to the first post
stop speculating
get some threaded pieces, a blacklight, and a scope
you are pushing a rope uphill

bigben2k · 03-13-2003, 04:34 PM

Quote:

Originally posted by unregistered
I seriously doubt the polycarbonate mfgr said that with regards to the dimensional tolerances needed for threaded connections,
and the same problem arises with 'flame' annealing

threading these plastics is rather pointless as there are easier/cheaper ways to affix connectors that are also more reliable

I gotta hand it to Bill: when he's right, he's right.

-Using a hot tool wouldn't work, because it doesn't allow you to create threads. (plus it could get messy!)

-Using the chemical solution could give inconsistent results.

-trying to hot melt a thread into place would be impractical.

In order to have a thread in plastic, it would have to be molded around a threaded shape, so that a good thread can be made without fractures. Since that's beyond our means, we're left with few options:

-Use an adhesive, to fasten say, a PVC fitting on top of the hole. This would involve drilling a hole in the top, which can be sealed/fixed by any means.

-Drop the whole idea of using a thread, and just glue a straight tube. Slip the tubing over it, and use a hose clamp. ref the polycarb barbs.

-Use a brass (or other favorite metal) threaded sleeve, and JB Weld (actually, use the proper adhesive) it into the top (after the opening has been sealed/fixed). Then screw in whatever you want.

nicozeg · 03-13-2003, 04:48 PM

Of course I like more the bonded tube solution, in fact, none of the blocks i've made have barbs.

But still lots of people likes brass barbs, and it seems to me that improving the reliability in this case is good. Also, the microfractures problem is present in every machining step made on polycarbonate: Sawing, milling, drilling, etc.

I'm sure that none of the block manufacturers that produced the cracking tops that polycarbonate a bad fame tried the "solvent trick" Maybe it don't avoid completely the problem but sure helps a lot.

If someone wants to use solvent in machined surfaces there's another precaution needed: No cutting oil can be used, It's going to penetrate the cracks and ruin any solvent job.

Cova · 03-13-2003, 05:02 PM

I said barbs earlier mostly because I believe they make a better seal with the tubing than a smooth tube, of course either would work and tube is obviously simpler.

As for holes used to connect the top to the base, assuming an o-ring is being used to seal the connection, wiping the holes with solvent after drilling should not be any problem. The holes should be over-sized enough anyways to allow the screws to fit through easily as well as account for the screws moving a bit due to the different thermal expansion of the top/bottom.

If you really want to use brass fittings with a plastic top, I don't see that we're going to come to any good solution to it. Perhaps coat the threads with solvent and while it is fluid screw the fitting into place - the plastic should harden into shape following the threads while sealing any microcracks, then if needed remove the fitting (it shouldn't actually be glued in unlike using a plastic fitting as well) and apply your favorite thread sealer eg. teflon tape.

bigben2k · 03-13-2003, 05:17 PM

Quote:

Originally posted by nicozeg
Of course I like more the bonded tube solution, in fact, none of the blocks i've made have barbs.

But still lots of people likes brass barbs, and it seems to me that improving the reliability in this case is good.

If someone wants to use solvent in machined surfaces there's another precaution needed: No cutting oil can be used, It's going to penetrate the cracks and ruin any solvent job.

I opted for an all plastic construction, from the beginning, but I have pp barbs, in a plastic-to-plastic connection (actually PVC). Otherwise, an o-ring will take care of my only plastic to metal joint.

We have to steer people away from the brass barbs! But otherwise, I'd opt for gluing a brass sleeve into the top. I don't see how brass barbs improve reliability.

Good tip on the solvent/oil.

Alchemy · 03-13-2003, 07:48 PM

Quote:

Originally posted by WinFlex
You do however bring up a valid point. WE NEED TO FIND THE CRACK TIP ENERGY for polycarbs in order to continue this analysis on a scientific basis.

You'd be better off with a crack growth rate, I think. I have one for an "average" acrylic - only an example graph, no specs on the material.

How about fracture toughness?

I have data on a sample polycarbonate with a fracture tougness of 3,000 psi -inch^(1/2). Yield stress is 8400 psi.

I don't have the geometry factor for radial flaws, unfortunately, so someone more skilled in mechanics should take it from there. If I spoke any further on this, I fear BillA might set my house on fire.

/me yells "What's that!" at BillA so he looks away and doesn't read the next paragraph.

What you really want, I think, is a crack growth rate. Assume for a second the forces are enough to lead to some crack propogation, and that flaws are already present. You know that there is a cycle of stress on the material as it heats up and cools down. This can tell you how long until some sort of failure occurs. If the forces are far too low or far too high for slow crack growth, it will be obvious at this point. You also need to know a coefficient of thermal expansion and stress-strain curve to find out what that cyclic stress is, as well as the size of any intial cracks.

Most of my knowledge is in fluid deformation rather than solid deformation, so that's all I got. Just be lucky you're dealing with unidirectional stresses.

Alchemy

BillA · 03-13-2003, 08:25 PM

no Alchemy, you're safe on this one; its many years since I played in this sand box

re the nomenclature:
when I say crack propagation I am indeed referring to the extension of an existing crack
- yes, cyclic stress regression is how the components are tested
(too many years in doing that)

and what we need to be thankful for is that the matl is isotropic

MadDogMe · 03-14-2003, 04:03 AM

Double post :shrug: ...

Might as well use it...

Does drilling a poly top cause much/any cracking(micro flawing/whatever)?, or is it just threading that's the real bear?, it's the 'sideways' load/pressure that threading causes that causes the cracking is'nt it?. what different types of threading tool are there?...

Are BSP treads tapered or just NPT?...

MadDogMe · 03-14-2003, 04:04 AM

Quote:

I said barbs earlier mostly because I believe they make a better seal with the tubing than a smooth tube, of course either would work and tube is obviously simpler.

If you think about it the smooth tube provides a better seal(more of the two surfaces mateing) than 'ribbed' barbs, barbs are ribbed to prevet airhoses being pushed off of them, or normal hose being pulled off, not to aid seal. As long as a smooth tube is clamped evenly it's seal is better, we're not talking high pressure here are we?...

************************************

One thing I wonder about is when sealing/glueing a copper tube into a poly top is how big to make the hole oversized to allow for adhesive?, too big and you're relying on the adhesisves tensile strenght as well as it's bonding strenght? (did I put that right?)...

I 'spose it all depends on how fluid the adhesive is?...

Volenti · 03-14-2003, 04:48 AM

When I made my micro pin block, which has 15 holes drilled in the acrylic top, 11 of them threaded, I used a slightly oversized drill for the the threaded holes (after some expriments on scrap material). The resulted I feel in a less stressfull thread cutting process, with less than full height threads cut into the acrylic.

Also the hose barbs were only screwed in with light finger pressure, with 2 part epoxy sealing the threads, thus avoiding the expansion cracking caused by trying to seal with plumbing tape.

I have examined the block closley and have found no visible cracking (which should be pretty easily seen with the led lighting in the block)

Plus having an almost 3/4'' thick peice of acryic probably isn't hurting things either

WinFlex · 03-14-2003, 08:03 AM

Quote:

coefficient of thermal expansion...

This has already been established earlier in the thread by me... first page, I think. Also, I used a UV diode and a stereo microscope to inspect my blocktop for microcracks. I found NONE. This may be attributed to mere blindness on my part or the fact that all holes were machined using endmills of lesser diameter than the holes themselves.
Anyhow, I think the whole discussion about poly tops cracking is slightly overrated. In my oppinion, the chance of getting a leak from a poorly bonded straight through tube connection is much more likely.
How many DangerDen blocks have cracked...? perhaps 5 (just a guess)... how many have been sold? 10000 perhas?
Let's see some statistics... and also can anyone locate the data that I requested earlier?

WinFlex · 03-14-2003, 10:24 AM

Lol... I just mentioned to my machinist the possibility that Lexan could crack when it has fittings screwed into it... like billa said, a complete split would occour, caused by catastrophic failiure.
He just grinned, picked up a half inch piece of antistatic lexan (the type i'm using in my blocks) and a large ball peen hammer...
He then proceeded to beat the living crap out of it for about a minute... pretty amusing. He then handed it to me... I peeled back the protective layer and except for a few microscopic indentations (note, indentations, not cracks) there was absolutely nothing to be seen. I will take this piece home tonight and inspect it further using UV light, but this pretty dramatic demonstration (be it as non-scientiffic as it was) has me thinking that Lexan is a whole lot tougher than we all think

BillA · 03-14-2003, 10:46 AM

different properties, different failure mechanisms
- and quite different test methods to differentiate between the two

indeed Lexan is incredibly tough, that is well understood
I think you looped back on yourself
your choice of words will also influence the answer
"a complete split would occur, caused by catastrophic failure"

is that what I said ?
don't think so
try instead the propagation of an existing micro crack (resulting eventually in a leak)

you need more accuracy
I will leave you to your 'proofs'

WinFlex · 03-14-2003, 11:28 AM

true... I agree that the method of testing here doesn't exactly simulate what happenes in our case... all it does is show the amazing impact resistance this material has. Nevertheless, I do think that you were taking about a complete split insdead of a large enough crack to allow water to penetrate the top.

Also, I NEVER claimed that this is proof. It is merely a piece of information to take into consideration when postualting a theory regarding the problem we are trying to solve. Lousy choice of words here, BillA.

Anyhow... where are the leaks? I have never talked to anyone who experienced one!

http://www.material.freeserve.co.uk/mats.htm#plas1
http://www.material.freeserve.co.uk/mats.htm#plas2

Ok. Here are finally some numbers, and BillA, I may just be getting closer to some of the proof you said I already stated

MadDogMe · 03-14-2003, 11:54 AM

IMO the micro fissures produced as the result of threading, (which I don't doubt), still only propagate as the result of user error, 'idiots using a pipe wrench to tighten barbs till they won't shift no more' is what springs to mind

. expansion~contraction under normal fitting of barbs is probably well under the critical point. I think I've only ever seen a couple of articles where there has been failure in a 'poly' top and I doubt they were Lexan as oppossed to plexiglass, the one sticks inmind featured severe discolouring/clouding as well, the bloke slapped on aload of epoxy and was satisfied with his 'fix'...

Pure conjecture I know so I'll shut up now

...

WinFlex · 03-14-2003, 11:56 AM

Thanks for bringing that up. Yes, plexi is much more prone to discoloration than lexan is, so that would point us to the conclusion that these tops were indeed not polycarbonate... however, more examples are needed

Ok. Combined with some calculations that I will perform tonight, this link should lay the entire conversation to rest:

http://www.geplastics.com/resins/dat...lexan/101.html

Since87 · 03-14-2003, 12:00 PM

Quote:

Originally posted by WinFlex
Nevertheless, I do think that you were taking about a complete split insdead of a large enough crack to allow water to penetrate the top.

Why in the world do you think that Bill was talking about a complete split?

Any crack with the potential to become large enough to cause leakage should be considered a serious flaw IMO.

WinFlex · 03-14-2003, 12:05 PM

There must have been a misunderstanding here... billa was talking about "...but fail instead by brittle fracture..." earlier. I guess he was refering to brittle fracture on a microscopic level, not a macroscopic one as I assumed. Infact, I do not think Lexan can experience brittle fracture on a macroscopic level under normal conditions, unless it is exposed to tremendous forces.

03-13-2003, 01:17 PM	#27
TerraMex Cooling Savant Join Date: Nov 2002 Location: Portugal, Europe Posts: 870	For bb2k >My question is: couldn't you just use a heating process, >like a microtorch? Something like this? Or ...? __________________ "we need more cowbell."

03-13-2003, 01:50 PM	#29
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	(not) That's what I meant, but now that I think about it, why not use a hot tool to press open a hole, instead of drilling it? Isn't it the drilling action that creates the microfractures?:shrug: __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

03-13-2003, 03:09 PM	#32
TerraMex Cooling Savant Join Date: Nov 2002 Location: Portugal, Europe Posts: 870	But... Doesnt the drilling of the plastic also heats it up due to the friction ? And if so, doesnt it mean that at least some (if not most) of the microfissures will be removed naturaly by that process? __________________ "we need more cowbell."

03-14-2003, 04:03 AM	#40
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	Double post :shrug: ... Might as well use it... Does drilling a poly top cause much/any cracking(micro flawing/whatever)?, or is it just threading that's the real bear?, it's the 'sideways' load/pressure that threading causes that causes the cracking is'nt it?. what different types of threading tool are there?... Are BSP treads tapered or just NPT?... Last edited by MadDogMe; 03-14-2003 at 04:09 AM.

03-14-2003, 04:48 AM	#42
Volenti Cooling Savant Join Date: Jan 2002 Location: in a nice cool spot Posts: 427	When I made my micro pin block, which has 15 holes drilled in the acrylic top, 11 of them threaded, I used a slightly oversized drill for the the threaded holes (after some expriments on scrap material). The resulted I feel in a less stressfull thread cutting process, with less than full height threads cut into the acrylic. Also the hose barbs were only screwed in with light finger pressure, with 2 part epoxy sealing the threads, thus avoiding the expansion cracking caused by trying to seal with plumbing tape. I have examined the block closley and have found no visible cracking (which should be pretty easily seen with the led lighting in the block) Plus having an almost 3/4'' thick peice of acryic probably isn't hurting things either __________________ feel free to icq/msn me, I'm always willing to toss around ideas.

03-13-2003, 01:39 PM	#28
WinFlex Cooling Neophyte Join Date: Feb 2003 Location: Boston, MA, USA Posts: 51	Nice to know taht people are curious about this sort of stuff. Regarding heat bonding, please read the link I posted earlier, about 4 post back. It is not as reliable as chemical bonding, and I imagine it would also be more complicated to accomplish!

03-13-2003, 01:52 PM	#30
WinFlex Cooling Neophyte Join Date: Feb 2003 Location: Boston, MA, USA Posts: 51	Very interesting suggestion, bigben2k... the amount of deformation that would occour using a hot tool is probably unacceptable, however. Nevertheless, howe about drilling, then inserting a hot pin that is a couple of thousands wider than the drill? This would melt shut any big cracks and completely erradicate smaller ones! On a sidenote, I took my block up to 150 psi last night (yeah, i know i'm crazy) and it's absolutely rock solid at 150 psi for over 12 hours now... nor preassure drop, no observable deformation, no cracks.

03-13-2003, 03:08 PM	#31
nicozeg Cooling Savant Join Date: Sep 2002 Location: Santiago, Chile Posts: 403	BillA, why you say that solvent wiping the threads dont work? It seems to me that the solvent would penetrate by capilarity the micro fractures and make a good job sealing them. In that case the polycarbonate is going to be able of holding an incredible tensile stress without damage. The source of that suggestion seems very reliable: a manufacturer of polycarbonate products.

03-13-2003, 03:25 PM	#33
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	nicozeg I seriously doubt the polycarbonate mfgr said that wrt the dimensional tolerances needed for threaded connections, and the same problem arises with 'flame' annealing threading these plastics is rather pointless as there are easier/cheaper ways to affix connectors that are also more reliable what is the goal ? simple is always better go back to the first post stop speculating get some threaded pieces, a blacklight, and a scope you are pushing a rope uphill

03-13-2003, 04:48 PM	#35
nicozeg Cooling Savant Join Date: Sep 2002 Location: Santiago, Chile Posts: 403	Of course I like more the bonded tube solution, in fact, none of the blocks i've made have barbs. But still lots of people likes brass barbs, and it seems to me that improving the reliability in this case is good. Also, the microfractures problem is present in every machining step made on polycarbonate: Sawing, milling, drilling, etc. I'm sure that none of the block manufacturers that produced the cracking tops that polycarbonate a bad fame tried the "solvent trick" Maybe it don't avoid completely the problem but sure helps a lot. If someone wants to use solvent in machined surfaces there's another precaution needed: No cutting oil can be used, It's going to penetrate the cracks and ruin any solvent job.

03-13-2003, 05:02 PM	#36
Cova Cooling Savant Join Date: May 2002 Location: Canada Posts: 247	I said barbs earlier mostly because I believe they make a better seal with the tubing than a smooth tube, of course either would work and tube is obviously simpler. As for holes used to connect the top to the base, assuming an o-ring is being used to seal the connection, wiping the holes with solvent after drilling should not be any problem. The holes should be over-sized enough anyways to allow the screws to fit through easily as well as account for the screws moving a bit due to the different thermal expansion of the top/bottom. If you really want to use brass fittings with a plastic top, I don't see that we're going to come to any good solution to it. Perhaps coat the threads with solvent and while it is fluid screw the fitting into place - the plastic should harden into shape following the threads while sealing any microcracks, then if needed remove the fitting (it shouldn't actually be glued in unlike using a plastic fitting as well) and apply your favorite thread sealer eg. teflon tape.

03-13-2003, 08:25 PM	#39
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	no Alchemy, you're safe on this one; its many years since I played in this sand box re the nomenclature: when I say crack propagation I am indeed referring to the extension of an existing crack - yes, cyclic stress regression is how the components are tested (too many years in doing that) and what we need to be thankful for is that the matl is isotropic

03-14-2003, 10:24 AM	#44
WinFlex Cooling Neophyte Join Date: Feb 2003 Location: Boston, MA, USA Posts: 51	Lol... I just mentioned to my machinist the possibility that Lexan could crack when it has fittings screwed into it... like billa said, a complete split would occour, caused by catastrophic failiure. He just grinned, picked up a half inch piece of antistatic lexan (the type i'm using in my blocks) and a large ball peen hammer... He then proceeded to beat the living crap out of it for about a minute... pretty amusing. He then handed it to me... I peeled back the protective layer and except for a few microscopic indentations (note, indentations, not cracks) there was absolutely nothing to be seen. I will take this piece home tonight and inspect it further using UV light, but this pretty dramatic demonstration (be it as non-scientiffic as it was) has me thinking that Lexan is a whole lot tougher than we all think Last edited by WinFlex; 03-14-2003 at 10:40 AM.

03-14-2003, 10:46 AM	#45
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	different properties, different failure mechanisms - and quite different test methods to differentiate between the two indeed Lexan is incredibly tough, that is well understood I think you looped back on yourself your choice of words will also influence the answer "a complete split would occur, caused by catastrophic failure" is that what I said ? don't think so try instead the propagation of an existing micro crack (resulting eventually in a leak) you need more accuracy I will leave you to your 'proofs'

03-14-2003, 11:28 AM	#46
WinFlex Cooling Neophyte Join Date: Feb 2003 Location: Boston, MA, USA Posts: 51	true... I agree that the method of testing here doesn't exactly simulate what happenes in our case... all it does is show the amazing impact resistance this material has. Nevertheless, I do think that you were taking about a complete split insdead of a large enough crack to allow water to penetrate the top. Also, I NEVER claimed that this is proof. It is merely a piece of information to take into consideration when postualting a theory regarding the problem we are trying to solve. Lousy choice of words here, BillA. Anyhow... where are the leaks? I have never talked to anyone who experienced one! http://www.material.freeserve.co.uk/mats.htm#plas1 http://www.material.freeserve.co.uk/mats.htm#plas2 Ok. Here are finally some numbers, and BillA, I may just be getting closer to some of the proof you said I already stated Last edited by WinFlex; 03-14-2003 at 11:47 AM.

03-14-2003, 11:54 AM	#47
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	IMO the micro fissures produced as the result of threading, (which I don't doubt), still only propagate as the result of user error, 'idiots using a pipe wrench to tighten barbs till they won't shift no more' is what springs to mind . expansion~contraction under normal fitting of barbs is probably well under the critical point. I think I've only ever seen a couple of articles where there has been failure in a 'poly' top and I doubt they were Lexan as oppossed to plexiglass, the one sticks inmind featured severe discolouring/clouding as well, the bloke slapped on aload of epoxy and was satisfied with his 'fix'... Pure conjecture I know so I'll shut up now ...

03-14-2003, 11:56 AM	#48
WinFlex Cooling Neophyte Join Date: Feb 2003 Location: Boston, MA, USA Posts: 51	Thanks for bringing that up. Yes, plexi is much more prone to discoloration than lexan is, so that would point us to the conclusion that these tops were indeed not polycarbonate... however, more examples are needed Ok. Combined with some calculations that I will perform tonight, this link should lay the entire conversation to rest: http://www.geplastics.com/resins/dat...lexan/101.html Last edited by WinFlex; 03-14-2003 at 12:02 PM.

03-14-2003, 12:05 PM	#50
WinFlex Cooling Neophyte Join Date: Feb 2003 Location: Boston, MA, USA Posts: 51	There must have been a misunderstanding here... billa was talking about "...but fail instead by brittle fracture..." earlier. I guess he was refering to brittle fracture on a microscopic level, not a macroscopic one as I assumed. Infact, I do not think Lexan can experience brittle fracture on a macroscopic level under normal conditions, unless it is exposed to tremendous forces.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)