Over 698,000 strictly plumbing related posts
Plumbing education, information, advice, help and suggestions are provided by some of the most experienced plumbers who wish to "give back" to society. Since 1996 we have been the best online (strictly) PLUMBING advice site. If you have questions about plumbing, toilets, sinks, faucets, drains, sewers, water filters, venting, water heating, showers, pumps, and other strictly PLUMBING related issues then you've come to the right place. Please refrain from asking or discussing legal questions, or pricing, or where to purchase products, or any business issues, or for contractor referrals, or any other questions or issues not specifically related to plumbing. Keep all posts positive and absolutely no advertising. Our site is completely free, without ads or pop-ups and we don't tract you. We absolutely do not sell your personal information. We are made possible by:
Author:
sum (FL)
Over the years I have read on this forum (and others) and there were always some controversial debates on the user of primer.
Some believes it does nothing except helping to line the manufacturers pockets.
Some do it simply because it's required by code and not because they believe it makes a better joint.
Some believe it makes a better joint because it softens the PVC materials, or because it "cleans" the material, or because it breaks the glaze, or other reasons.
Some apply the primer then the glue quickly, while the primer still wet. Some wait till the primer dries.
Some apply the primer only on the male end, some does both the male and female, some does female/male/female.
There is even a thread here a few years ago where someone claimed to have done some tests and found that joints assembled with primer is less strong.
[www.plbg.com]
My son is in the 10th grade and looking for a topic for a science fair project. I suggested to him why not do one that tests the primer?
He can assemble joints with and without primer, and test to see which one is better.
He can do both PVC and CPVC; and apply primer then glue and vary the wait time in between to see if the primer is more effective at various degrees of dryness; He can cut out assembled joints and examine under a microscope effect of the joint with and without primer.
Now in order to determine which joint is the strongest I am thinking a tool is necessary to either exert tensile stress or torsional stress to the joint until it breaks so that the stress that caused the failure can be quantified. I am trying to figure out what tool that would be and whether it is something that can be put together at home or I need to find some machine shop/lab.
Also, would be nice to do a leak test and see the joints may fail at different water pressure, not sure if that would work. A standard static head test would probably be passed by all of them.
Any thoughts? Comments? Suggestions?
If you know of previously done tests that relevant please point the way.
Edited 1 times.
|
Post Reply
|
Author:
KCRoto (MO)
I would make it a test with no primer, clear primer, and purple primer.
You could assemble the pipe with just primer no glue as well.
The primer has a cleaning agent in it, and leaves a coating on the pipe; I think the cleaner helps more than anything myself, but I use primer regardless. I generally use clear primer and glue for my projects, then take a purple sharpie marker after I am done and draw around the joint so it appears to have purple primer on it. The purple is just so that an inspector can see that the pipe has been primed. I worked with a guy that didn't use primer on his joints, and he had joints fail and leak within a year; I suspect that it was due to dirt or oil being on the pipe and the glue not bonding properly. I would be interested to see what you and your son find out, and how exactly you test the joints and so forth. Please keep us updated, I am sure I'm not the only one who is curious as to what you will find. Keep in mind that if you glue cpvc, it uses a different glue than dwv glue, and I suspect that you will break the pipe before you break a joint if it is allowed to cure properly prior to testing.
|
Post Reply
|
Author:
hj (AZ)
A cable puller with a scale between it and the fitting would test the strength of the joint. A water pressure test would probably shatter the pipe before the joint failed.
|
Post Reply
|
Author:
hj (AZ)
CPVC is NEVER used for dWv purposes, and dwv glue for PVC is the same as used for pressure piping.
|
Post Reply
|
Author:
WC (VA)
Test using a hydraulic press -- Harbor Freight 12 or 20 ton modified with a pressure gage via oil fill or pressure release. Gages with range of 0-6000 psi are available for $12. (Google gages) -- more $ gets a higher range -- Don't know what range would be needed. High pressure hose, and fittings are available at a local Hydraulic Service shop. Two small steel plates also needed for equally applying pressure with press.
TEST EXAMPLE:
Pipe and coupling should not be too small or big - IMO about 1-1/2 or 2 inches.
TO START (Trial and error)
Mark and insert glued pipe vertically into coupling ONLY about 1/2 or 1 inch. Applying pressure pipe "may" separate and "slide" into fitting but more likely all will break. --- However --- This may begin to ID the weaker method of assembly -- which is what you are looking for. -- It's a start.
NEXT
Mark and insert glued pipe into coupling about 1/2 or 3/4 of the depth of the fitting. Applying pressure, pipe and fitting will likely break. But --- again this may ID the weaker method of assembly.
The above is just a idea of course, I'm sure improvements in test method could be made with more thought.
This test would possibly ID the weaker assembly method for mechanical strength. Liquid pressure test would require assembly using additional threaded parts for plugs etc. Plumbing is mostly of low pressure and IMO probably all of the assembly methods using the full and proper inserting of pipe / fitting would hold. Revealing if one method or other has a better / or worse effect on the actual mechanical strength of fitting would be interesting to find out. --- IMO comparing mechanical strength of assembly would be relevant to what pressure different assembly's could hold. However -- depending on how things seem to fall out with mechanical testing, additional liquid pressure testing may should be done.
Consistence is required.
Same length and square cutting of pipe ends. (Use a chop / miter saw)
Time allowed between assembly methods and test etc etc.
Documentation -- Documentation of materials and assembly
Safety glasses / equipment worn at all times.
|
Post Reply
|
Author:
Paul48 (CT)
I'd use pvc, just not pipe or fittings. I would think it would be much easier testing with pvc blocks, and give you the same information.
|
Post Reply
|
Author:
sum (FL)
At the present moment, he is considering two types of pipes. PVC and CPVC. Not sure what size to use, I am not sure that doing the experiment using multiple sizes will be too telling.
As far as the actual primer/glue, we will attempt to use the following:
Cleaner
Purple Primer (can be used with both PVC and CPVC)
As for the cement there are many products, you have for just PVC the regular glue, the medium duty, the heavy duty, the all purpose for both PVC and CPVC, the rain or shine, the red hot blue etc...need to pick a few not all. He is even thinking of trying the spray on one step PIPEWELD.
So he can will have these:
No primer no cement only pipe and fitting pushed in until bottoms (as a control)
cement only
cleaner + cement
primer + cement
cleaner + primer + cement
and he may try to prime and immediately apply cement versus prime wait a few seconds then cement
each unique combination will have a number (say 6) of identically made joints for testing.
we will post results as we go, and may be show in youtube videos how he makes the joints so you guys can make sure he is doing it right and consistently.
The biggest issue I see is what is the best method to create a joint failure that can be easily quantifiable. Tension to pull them apart? Torsion to twist the pipe off the fitting? Compression to break the assembly?
Back in college we had to do in civil engineering lab concrete mix test. Each student made a concrete cylinder and let it cure for 28 days then we put it into a compression machine and it increases pressure with a read out 1800 psi, 1805 psi...crack! and we can record the strength of the concrete cylinder. We don't have access to such expensive electronic machines here.
Furthermore, leak is a whole different thing. I suppose a joint can be very strong mechanically but leaks. But I am not sure it is practical to test for that without some elaborate setup.
|
Post Reply
|
Author:
sum (FL)
a cable puller. hmmm.
possibly, however I worry about that may be dangerous. In order to read the scale or strain gauge, and to apply the tension, at the point where the failure occur, there will be a piece of pipe/fitting attached to a cable full of tension flying into someone's head at a high speed.
Since it's solvent weld, wouldn't the perfect joint actually fail at the pipe and not at the joint, since the fitting and pipe becomes one and is thicker there?
|
Post Reply
|
Author:
sum (FL)
I see where you are getting at, by assembling the joints NOT at full depths, you can use compression to cause the joint failure and it may be an easier test.
While it may quantify the relative strengths of the joints, not bottoming out on a tapered joint may invite complaints that one cannot really rely on test results from improperly made joints, also just marking a line on the pipe does not mean my son can consistently hit that line with the cement's tendency to push out etc...having a hard stop is important I think.
However what you proposed gave me an idea of a possible variation. Use a repair coupling that has no stop inside, have him insert half way and let cure, then compress the assembly to cause the failure.
|
Post Reply
|
Author:
WC (VA)
A metal hose clamp (with a mark) could be used around the pipe to act as a external stop to ensure consistent depth insertion.
Partial insertion is suggested - at least for some of the first tests mainly IMO the entire assembly will crack / break when pipe is fully inserted and compressed under hydraulic pressure, UNLESS of course one of the procedures IS much weaker than others. When testing is done at full insertion that is a good idea to use a repair coupling with no stop. (A hose clamp with a mark should still be used to make sure all fittings pipe had the same depth insertion.)
True, with partial insertion, some would say because the fitting is not fully inserted it is not a accurate test. However it would prove any differences, (or not), of mechanical bond of the same materials using different methods of bonding. This would indicate which method is the stronger - and better - IF there is any difference.
I think any assembly method that would pass a reasonable hydraulic press mechanical test would hold the normal liquid low pressure used in plumbing.(At least usually.) This is assuming that when assembled it was done properly with the twist to spread the glue etc. Liquid prssure testing a few assembly's should confirm that if there is any doubt - and to answer possible questions.
|
Post Reply
|
Author:
hj (AZ)
To test a joint he has to apply TENSION, not PRESSURE. Pressure just forces the pieces together until they shatter.
|
Post Reply
|
Author:
hi (TX)
it will be hard to test strength with the inexact standards that you can establish for making a solvent weld.
I might suggest doing a weld on several pipes 2" in size and then using a band saw to cut the joint in two. Carefully remove the pipe from the fitting and then take pictures of the bonding area on the remaining fitting using the Macro on the digital camera. You can get some idea of how big the bonding/solvent welded area is.
PVC pipe plumbers commonly use is white because of the clay filler. I believe there is clear PVC that could be used to photograph the weld area also.
|
Post Reply
|
Author:
WC (VA)
"To test a joint he has to apply TENSION, not PRESSURE. Pressure just forces the pieces together until they shatter."
Applying evenly distributed hydraulic pressure against the end of a pipe -- attempting to PUSH it INTO a coupling would cause a shearing action between the pipe and the coupling wall. It would not have a crushing effect. Testing would be done with only partial insertion and/or using a repair coupling which has no shoulder to be pushed against. It would have the same effect as attempting to pull the pipe from the fitting as previously suggested ---- only in the opposite direction. It would also be safer as previous noted. Hydraulically compressing - end to end - could be more easily controlled and pressure measured when assembly breaks/shears apart. When pulling apart, breakage/failure of assembly would be sudden and dangerous for anyone nearby taking measurements.
Testing as suggested may very well always cause the assembly to shatter and never shear apart (just as pulling) -- unless one or more of the assembly methods is very inferior. When assembly shatters or shears apart the pressure would be known and compared with different methods of assembly.That unknown is what the testing is about.
|
Post Reply
|
Author:
vic (CA)
The biggest issue I see is what is the best method to create a joint failure that can be easily quantifiable. Tension to pull them apart? Torsion to twist the pipe off the fitting? Compression to break the assembly?
MANY years ago while taking some plumbing courses at L.A. Trade Tech at night a rep from Spears plastics (if my memory serves me correctly which is getting to be not so reliable) came to one of the classes to teach us a bit about solvent welding.
He taught and proved to us how using primer was far superior to a joint without using primer.
Products and formulas have changed (hopefully all are better) and so I can't make any claims today with facts of today's products however I can share how he proved it to us.
We all used clear cement and "glued" a pvc fitting to pipe as well as we each did another fitting/pipe where we used purple primer first and then cement.
Talk about a "high" classroom .... lots of vapors.
Then he gave us a lecture about pvc, his company and so on and after waiting as long as possible we all cut the fitting/pipe (in other words the solvent welded joint) in half.
Half ways where both sides looked like a big C.
If I recall we could visually see the difference as well as he took a hammer to the C's shaped half fittings and pounded a couple of them with a hammer.
Not very scientific, certainly no measurements were involved however it was easy to see how much stronger a weld was when using primer.
My gut (I have no proof these days) says that even today "wash and wax" all in one can still isn't as good as washing a car and then waxing it .... and so this old fart still believes that primer first and then solvent cement is superior to any all-in-one-can solvent cement.
I can't wait to read what the results are and how all was done.
Edited 1 times.
|
Post Reply
|
Author:
WC (VA)
"The biggest issue I see is what is the best method to create a joint failure that can be easily quantifiable. Tension to pull them apart? Torsion to twist the pipe off the fitting? Compression to break the assembly? "
Exactly -- A safe test which can be easily quantified is what is needed.
Applying pressure "pushing" the pipe INTO the fitting (without a shoulder) will cause failure at some point. At the point that a pipe and coupling either shatters or "shears" apart the hydraulic pressure would be known and recorded. I have never seen this test done so IMO I'm guessing a fully assembled pipe and coupling is more likely to shatter that shear apart. That is why I suggest also testing with pipe inserted at various shorter measurements into a coupling, only a inch or so. With only short sections it is more likely to get a shearing action instead of shattering, especially if one of the methods has less bonding and is weaker. This pressure can be measured for compassion. No matter how or when the assembly fails, comparing the pressure that it fails will ID if any assembly method has a mechanical advantage. This test does not rely on visual interpretation, which some may or may not agree.
This test will not answer to everyone's satisfaction as to what assembly method is the better. It WILL give quantified, directly comparable results.
(Agreeing to what they mean will probably not happen until the cows come home.)
Actually I would also like to see a measurable torsion/twist effect test applied, but I don't know of a easy cheap safe way to accomplish that. Why haven't manufactures done that -- or if so made results known? I suppose it is in the manufactures interest to sell several different combinations of glue/primers etc. It would be more convenient and cheaper for the consumer if it was known for a fact that some of the presently sold/used products were not really needed or had no real measurable advantage.
|
Post Reply
|
Author:
KCRoto (MO)
I would be curious as to how cell core vs. sch 40 pvc would hold up as well.
|
Post Reply
|
Author:
sum (FL)
WC thanks for the hose clamp idea. We are thinking this may be a good way to do it.
I have lots of older pipes and fittings so I will have my son practice with it. I am sure he will earn a few purple stained shirts before he is "proficient" in doing a joint consistently.
We were discussing various project ideas last week and I showed him the thread by hlee and said there seems to be some real controversy on this subject and we couldn't find any manufacturers' test results (which is actually surprising). I suggested this project to him and at least I believe it will be somewhat meaningful instead of some obscured bacteria petri dish type subject.
|
Post Reply
|
Author:
sum (FL)
vic, I will see if we can integrate some visual cut outs into the report.
It helps to have quantifiable results as well as visual evidence. I hope there is a way to tell the difference visually between "softened PVC welded together" and "unsoftened PVC glued together" when looking at a cut out.
|
Post Reply
|
Author:
sum (FL)
hj, if the coupling has no stop in it then the tension or compression would be testing for a similar failure. I too would prefer a tensile test if one can be done that's measurable and no danger of snaping and becoming a projectile.
|
Post Reply
|
Author:
sum (FL)
Right now we are thinking SCH 40 SOLID CORE PVC and CPVC. We could add CELL CORE. However, I worry about using softer pipes as the test (whether it's compression or tension or torsion) to create a failure is only going to be conclusive if the failure happens at the actual assembled joint.
In other words, when assembled with primer and cement, the joint was pulled apart at X pounds of force, and when assembled with only cement the joint was pulled apart at Y pounds of force. If X > Y we can conclude the primer makes a significant difference.
However, if both test results in the actual pipe being fractured, then the tests would be inconclusive. A softer or thinner pipe may have a higher chance of that happening. We could try one or two and see if it's good enough to include it into our sample.
Edited 1 times.
|
Post Reply
|
Author:
sum (FL)
"Actually I would also like to see a measurable torsion/twist effect test applied, but I don't know of a easy cheap safe way to accomplish that."
I have been thinking about this and here is an idea.
I will use two of these 1/2" plugs, with a section of pipe in between.
So PLUG - PIPE - PLUG
Once the joints are made and cured, I can fix one end of it with a vise. Then if I can find a hex socket to fit over that hex shaped plug, I can turn the assembly with an adjustable socket wrench. Assuming there is a socket that fits around the hex shaped plug, I can incrementally adjust the socket wrench until one of the two joints fails. I will not have a precise failure torque, but I know it's between say 40#in and 50#in for example. Would that work?
This will only work if the joint fails. It will not work if the pipe in between gets twisted, or if the torque strips and rounds the hex shaped plug.
What do you all think?
|
Post Reply
|
Author:
WC (VA)
Using two pipe hex headed plugs for a torsion/twist test looks to be a interesting test. Is it your intention to just force the assembly to break for inspection purposes? Or do you mean you will be using a torque wrench to slowly increase torque in increments to when "something" finally happens.
I would suggest having zero exposed pipe between the two plugs. I think the pipe should be the exact length needed to insert into and connect the two plugs. This may keep the pipe from twisting/breaking before the assembly does.
If a torque wrench is used the problem is having some idea what range torque wrench you will need. (? to ? foot pounds.)
Select a torque wrench that has the required torque to be applied falling around the middle or well within the wrench’s high-low torque range. Make sure the selected wrench’s capacity is rated above the torque required. Do not exceed or go below the torque value range of the torque wrench. When a "clicker / release" type wrench is used, do not exceed the torque setting (when it clicks / releases) or wrench will be damaged. If increase is wanted you must stop and change the setting to a higher range.
If not familiar with use and selection of torque wrench's Google and youtube it. There are important rules that must be followed for use and storage.
|
Post Reply
|
Author:
KCRoto (MO)
automotive and farm supply places generally have large sockets like that. That is where I got mine to fit elements on water heaters.
|
Post Reply
|
Author:
sum (FL)
"Using two pipe hex headed plugs for a torsion/twist test looks to be a interesting test. Is it your intention to just force the assembly to break for inspection purposes? Or do you mean you will be using a torque wrench to slowly increase torque in increments to when "something" finally happens."
The latter. That way I can have an approximate torque at the failure point for the joints made with vs without primer.
It may also be telling, may be the primer helps with tension/compression but not as much with torsion.
However, the following may happen.
The hex headed plug gets stripped or rounded before anything happens.
or, I couldn't find a socket to fit that hex head snugly, which causes it to strip or slip.
or, the pipe in between twists and ruptures before anything happens - yes I think shortening the length will help with that, thanks.
now, there is also a chance that without primer, the joint would twist apart under torque. With primer the pipe twist or break. If this happens, then I don't need to buy an "adjustable" wrench.
|
Post Reply
|
Author:
WC (VA)
"The hex headed plug gets stripped or rounded before anything happens."
Use a 6 point socket (not 12 point) for more secure fit and better chance of not deforming hex on plug.
|
Post Reply
|
Author:
sum (FL)
Back to the drawing board.
I went to home depot and got the plug. I then walked over to the tools section and tried to match a socket size. I then realized the plug is not a hexagon, its an octagon!
Edited 1 times.
|
Post Reply
|
Author:
WC (VA)
Seems there is always a surprise. I did not check any plumbing supply but I did notice the few copper/brass bushing I have laying around are 6 sided hexagons. Maybe check around other PVC bushing/caps etc, possibly a different brand / Mfr make what you want. --- Next --- there is always a tool for every/any thing for the right $ --- someone probably makes a 8 sided octagon socket --- "maybe" in the size you want. Be flexible, the best fit may be SAE or Metric.
|
Post Reply
|
Author:
hj (AZ)
If you turn the plug hard enough, the fitting will crack LONG before the joint fails.
|
Post Reply
|
Author:
sum (FL)
hj, that may be the case, but if that hdppens, then one can conclude that solvent welded joint was made and strong as can be. However if primer is used ans the plug cracks, if primer is not used the joint twists apart, that is useful to know.
|
Post Reply
|
Author:
sum (FL)
I seem to remember some time ago, there was an adjustable socket, it looks like one with many tiny pins and needles, and the claim was that you can use it for any shape and size. You press down on the nut and the needles over the nuts recesses while the rest of the needle stay, conforming to the shape of the nut. What is that thing called? I might just have a use for this.
|
Post Reply
|
Author:
KCRoto (MO)
a universal socket I think. I don't think it goes larger than an inch or so on the one that I saw a few years ago. I googled universal socket and saw that craftsman has one, and one called Gator Grips.
|
Post Reply
|
Author:
sum (FL)
Yes KCRoto, I can't find one larger than an inch also. Looks like I need to go to PLAN C.
|
Post Reply
|
Author:
sum (FL)
This is the current thinking on the compression test.
There will be a short piece of pipe, and two joints made on each end, with repair couplings so no stop in the middle. One end will be with primer + cement, the other end without primer. A three piece assembly.
Then I will attach a 4X4 to the concrete block wall with a series of Tapcon screws. At the bottom I will have a bathroom scale, on top of it a bottle jack, then a piece of wood, then the pipe assembly test piece, like this. As I crank the bottle jack the bathroom scale should register the force until the failure point.
The reason to create an assembly like this is to also test both joints at the same time to see which one fails first. Once one end fails, the failed fitting removed, the other joint will be subjected to the test again until it fails too.
not sure if the bathroom scale would be adequate. A scale would register up to 300 pounds, if it goes above that then I won't be able to get a reading.
Is there a bathroom scale for a gorilla?
Edited 1 times.
|
Post Reply
|
Author:
sum (FL)
I found an interesting article with some data.
[www.harvel.com]
seems to be data based on tests performed by IPS. Here is a chart from the link.
For example, they showed a 4" Sch 80 PVC or CPVC Pipe. 4.5" O.D. 2.25" socket depth having a minimum cement shear strength of 300 PSI.
Cement surface area (in2)=(O.D.)(PI)(socket depth) = 31.81 in2
Minimum cement shear strength = 300 PSI
Resistance to shear in pounds force = (31.81) (300) = 9543
This is a lot of force almost 10000 pounds.
Now if I don't use 4" pipes but say only 1/2", the surface area reduces to 1.57 in2
Resistance to shear in pound force = 1.57 * 300 = 471
which is more than what a bathroom scale can handle. I need something with at least a 600 pound capacity.
I was also hoping I could use more than 1/2" pipe as I think there is a chance the joint may not fail before the whole pipe assembly buckles.
Edited 2 times.
|
Post Reply
|
Author:
KCRoto (MO)
how about use 2" pipe, but cut down the fittings so that only 1/2 inch of fitting can make contact, or use the hose clamp idea posted previously. The larger pipe would be more stable during tests; I suspect the surface area v. bond strength follows a more exponential path until it becomes untestable as the pipe itself shatters. (or the fitting)
you could test the depth of the fitting in 1/8 inch increments to find out
|
Post Reply
|
Author:
WC (VA)
Good idea to use a bottle jack in lieu of a hydraulic press, it is certainly a lot cheaper. I would suggest the following.
Add a pressure gage and bleed off valve (if needed) with hose to the bottle jack via the oil fill / pressure release. I don't think a bathroom scale will be able to register near high enough. Example if a 300 lb person were to stand and balance on a vertical pipe end, I don't think anything would happen (to the pipe anyway).
Add a metal plate to the bottom of the wooden 4X4 pressure point.
The 4X4 must be really secured to the wall extrealmy well or be blocked in the overhead, to be able to resist the applied pressure. Other wise the wall could be damaged if attachment does not hold.
|
Post Reply
|
Author:
sum (FL)
Possibly.
I could reduce the surface area of the joint in order to put a limit on the force necessary to break the joint.
Using the 300 PSI, and if I assume a 1" O.D. pipe, the surface area would be 1"*PI*SocketDepth.
and if I use a maximum force of 300 pounds, and calculating the socket depth required to fail at 300 PSI, I end up with a socket depth of about 5/16". I believe this is too shallow a joint. For some reason I am thinking at least an inch would be required to create a reliable joint for the test.
|
Post Reply
|
Author:
sum (FL)
"Add a pressure gage and bleed off valve (if needed) with hose to the bottle jack via the oil fill / pressure release"
I am not familiar with how this is done. I have a cheap Harbor Freight bottle jack I bought some time ago to hold up a trusses while I made my repair. I don't remember if there is an oil fill & pressure release.
If this could be done I assume the reading will be in PSI?
also, I am wondering how meaningful a torque test is. Even though we call it a compression test because it is compressing the pipe assembly, what is being tested is the shear strength of the welded joint.
By torquing the pipe or fitting to cause it to come apart, the test is actually on shear force required to tear the joint apart, it's really the same test with the force applied in a different direction. I wonder by doing this test whether the resulting data would be redundant, or useful.
|
Post Reply
|
Author:
WC (VA)
A bottle jack should have at least one if not two threaded connections.
1. A pressure "bleed off" near the bottom side. This is the small threaded valve which you use to depressurized the jack. IF this is the only threaded connection than in addition to a pressure gage with fittings a tee with another valve would be needed to be added with the assembly --- used to depressurize the jack. If this is the only available threaded connection it will have to be determined if removal of the bleed off valve will support using that for the connection. I have never done it -- so am not sure. Also when you bleed off the pressure it will not be released internally but actually out of the new valve, and will have to be contained and replaced after each operation. If this is the only existing threaded fitting, it may be better and more practical to add another one for the gage.
2. Some jacks in addition to the above bleed off valve have another threaded plug for adding hydraulic oil. This plug could be removed and a pressure gage attached with required fittings and possibly a short hose (if needed or wanted). No additional valve would be needed as the existing bleed off valve would not be removed.
Understand that at my present location I do not have a bottle jack in hand to verify the above -- I am going by memory.
Sizes of required fittings to be determined. A local hydraulic supply shop could prefab the required assembly --- for a price.
The gage can indicate whatever you care. I would suggest one that reads in PSI (Pounds per Square Inch).
As previously indicated gages with range of 0-6000 psi are available for $12 (Google gages).
I "think" a test applying pressure to the end of a fitting - attempting to push it into the fitting, (IMO a attempt to cause a shearing action) would be different than attempting to twist (torque) a pipe and fitting to cause separation. I believe either or both would give a indication of the superiority of one type of assembly over the other -- IF -- there IS any difference. Most likely a fully assembled pipe and fitting will break (shatter) apart during the test unless one type of assembly IS much inferior. IMO the better chance to compare mechanical bond without the fitting shattering will be to test pipe inserted into a fitting for only short comparable measurements as previously mentioned. This type of test could more likely ID a inferior type of assembly -- causing it to shear or break in only two pieces instead of multiple breaks.
Either of the tests will determine.
A) Fully assembled pipe and fitting will always completely shatter into pieces when enough pressure or twist/torque is applied.
A inferior type of assembly will not shatter but shear or twist apart.
C) Partly assembled pipe and fitting will shear or twist apart -- but some will hold together better than others ---- AS THE ASSEMBLY OF THE PIPE AND FITTING ARE INCREASED IN DEPTH. This, IMO would ID the inferior type of assembly (IF there is one). This is what the test is about.
However -------- Be prepared ----- no matter what your results are, it is unlikely to convince anyone who actually uses PVC pipe and fittings that they "Really mean anything".
|
Post Reply
|
Author:
KCRoto (MO)
you could make an extremely strong, self contained frame with 4*4's and thread rod. I found a picture of what I was imagining (kinda) Then you don't have to worry about securing it to a wall or ceiling.
|
Post Reply
|
Author:
WC (VA)
SUM ---- Important change to previous suggestion.
Follow-up installing pressure gage to bottle jack.
I believe the only connection point would be a oil fill plug if the jack has one. With more thought, removing the bleed-off for a gage connection would defeat it's purpose and the internal jack oil flow would stay in the bleed-off mode and not allow the jack to function.
|
Post Reply
|
Author:
hj (AZ)
I can't believe this discussion is still going on. In the first place you are trying to test a "proper" joint, and if the pipe is NOT inserted fully, (so you can do a compression test), then it is NOT a proper joint and the results will be flawed. The "fill plug" on a jack is on the 'no pressure' reservoir side so it will not register any pressure and if you remove the "release" valve there is also no way to create pressure in the jack.
|
Post Reply
|
Author:
hj (AZ)
quote; For some reason I am thinking at least an inch would be required to create a reliable joint for the test
To create a "reliable joint" for the test, it should have FULL insertion, and then subjected to a tension stress NOT compression.
|
Post Reply
|
Author:
sum (FL)
hj,
I don't see the difference between a tension test versus a compression test in this particular case. Since we are using a repair coupling, the only thing holding the two pieces together is the solvent welded portion, there is no middle stop. When compression is applied, the shear force will tear the joint apart at the seam of the joint (at least that's the theory). In this case, tension or compression are simply force applied in opposite direction, the failure is due to shear, and shear applied outward or inward doesn't matter. If you look at the IPS test, this is also what they tested, shear. My compression test is just a way to test the shear strength of the welded joint. Now if a compression actually breaks the pipe or fitting, or a tension actually pulls a pipe apart not at the seam of the joint, then the test is moot anyways.
Regarding whether the joint is done at full depth, I agree, full depth is ideal. However, I don't think a perfect joint is absolutely necessary. I think if we can keep all variables the same, then a primed joint with 1" socket insertion should be stronger than an unprimed joint at 1" socket if it's done at the same temperature, cement, primer etc... applied the same way, within the same time, allowed same time to cure etc...I do understand, without inserting the pipe at full depth (half length of a repair coupling), it will probably be dismissed simply because the results are not reliable due to the joint being improperly made. I can see that being a concern. But bear in mind this is a 10th grade project, and I can't buy an industrial test machine, so we will have to make do with creativity and compromise in rigging something together.
That's why I started this thread is so I can get opinion to have this done best we can - within the limits of our resources, we hope the results would be interesting and hopefully can confirm a few things.
Edited 1 times.
|
Post Reply
|
Author:
hj (AZ)
Now all you have to do is add in the factor of age/time, because I have had unprimed joints last 20 years before they came apart.
|
Post Reply
|
Author:
sum (FL)
True.
This is why the tests we are going to do is simply going to conclude whether a primed joint is stronger than an unprimed joint. Nothing more.
It could very well be that both primed and unprimed are both adequate.
It also does not do anything in terms of telling which joint would be more leak proof. That would have been a good test to do too but I can't think of any way to do that sort of tests with "home grown" tools.
When I was going to college our civil engineering lab had all kinds of test machines that can easily perform these tests. I thought may be a phone call or two I can obtain access to local university labs, boy was I wrong. I think 20 years ago that would have been true, today they simply can't do this sort of things due to liability issues, while I understand, is kind of disappointing.
|
Post Reply
|
Author:
sum (FL)
I found a socket that the 1/2" octogonal fitting will fit snugly in. A 15/16" 12 point socket. Will it twist the joint apart (as I hope) or will it round out the octogon? I don't know.
Also I found this gadget at Harbor Freight.
[www.harborfreight.com]
Get precision torque control with any ratchet! Three color LED and progressive audio notification when reaching specified torque.
TRACK mode displays torque value as pressure is applied, prevents over-torquing
PEAK mode displays highest torque value after load is released
Accurate within +/- 2%
Torque range: 29.5 to 147.6 ft. lbs. (40 to 200 NM)
Auto shutoff in 80 seconds
50 memory settings
Anyone has experience with this? I wonder if this may be useful to measure the torque, and more useful then an adjustable torque wrench.
Edited 1 times.
|
Post Reply
|
Author:
KCRoto (MO)
I have a very strong suspicion that you will put more torque on that pipe than what the meter can handle. [www.physicsforums.com] I found a forum that might be able to offer some guidance as well. I was looking up pvc compression strength and found a house built with foam and pvc, and this physics forum. Figured any additional resources couldn't hurt.
|
Post Reply
|
Author:
hj (AZ)
IF the joint is done properly, you are going to have to apply more torque than that meter can measure OR the "hex" joint will withstand without cracking, unless you glue the plug into the adapter. You may also need a couple of 350# linebackers or a 600# gorilla to pull on the torque wrench to get the joint to fail.
Edited 1 times.
|
Post Reply
|
Author:
WC (VA)
Wheew ------- I was worried that due to complaints this interesting thread had stopped. Another recent thread concerning how to disassemble / assemble a union has been getting all the attention and is fast catching up on the post count ---- and may yet pass it.
I know cost is a major concern and I understand that. For applying pressure at the end of a assembly I have seen on Amazon several hydraulic Bench Presses that have gages ALREADY installed. One is a Dake model for $299.65. Others are as "little" as $244. A Bench Press would also be convenient to move around if needed for demonstration. If serious, read all the reviews as one complained the gage failed early.
------ After use, some cost recovery may be made via resale (craigslist).
Applying torque to cause a failure will not be easy. It also may be a little on the wild side when failure is abrupt. Safety glasses / shield, heavy apron required. Expect PVC shards. IMO to be successful a 6 (or 8) point socket and a torque multiplier in conjunction with a high range torque wrench will be need. Possibly in the thousands of pound range. This IS cost prohibitive but maybe somewhere rental is available. I have personally used one at over 5,000 lbs (and they go much higher) but the US Gov provided it.
PRACTICAL LOW COST SUGGESTION FOLLOWS:
As a preliminary torque measurement test. It may be useful to force failure of a fully glued assembly using ONLY a socket, breaker bar and pipe (long) extender AND safety equipment --- just to get a "feel" for what it takes. From that you may get a better idea as to what equipment is needed to take measurements while forcing failure. It may be that breaking a small assembly via torque is easier that expected. That's what testing is about.
|
Post Reply
|
Author:
KCRoto (MO)
A cheater bar on the wrench would do it. You can snap a 7/8" grade 8 bolt from torque alone like that.
|
Post Reply
|
Author:
KCRoto (MO)
I agree that testing one joint to failure would be the fastest and easiest way to get an initial starting point. A bench vise, long lever, and a simple method of gauging force would work to get your ballpark, then start testing at about 30% lower than the initial break point.
|
Post Reply
|
Author:
sum (FL)
Well, some preliminary results.
First, it was a pretty interesting experience to teach my son how to create PVC joints. I put together some old fittings and pipes and had him practice about half a dozen joints.
So today we wanted to do some preliminary testing just to get a better idea so he can put together all the needed pieces for the real tests later on.
First for the compression test we could not find any repair coupling smaller than 1.5". So after some debating we got 3/4" couplings and SCH 40 3/4" pipes. We glued together everything as depicted below...but using regular 3/4" couplings but only inserted 3/4" into the sockets.
We waited 90 minutes before we did the tests, to give it some time to set. Manufacturer suggested 24 hours, so our real tests we will use 24 hours.
We then used a bottle jack and cranked on the PVC assembly and was able to cause the joints to fail. THE UNPRIMED JOINTS failed first, then the primed ones. However, as expected, a bathroom scale was not enough. I am guessing the forced to cause it to fail was about 350 pounds or so. If we have a 400# scale that would work.
So we need to locate a 400# scale, and see if there is a source for 3/4" or 1" repair couplings so the joints can be made by inserting the pipe to full socket depths.
On the torsion test it was a total failure.
I did find a hexagon fitting, and it fitted snugly into a 1-1/16" socket, and I had a digital gadget to measure torque, one end of it we put into a vise, the other end we put the socket on it and turned. We couldn't tear the joint apart. Not the primed joint, not the unprimed joint. The hexagon got totally rounded out by the metal wrench and just "skipped".
I then took out my 24" plumber plier and cranked on it, same result, the plastic got chewed up but the joint stayed together.
The only thing I can think of is again to reduce to depth of penetration and see what happens.
Edited 1 times.
|
Post Reply
|
Author:
KCRoto (MO)
Your test on the torsional force would be testing 3.14 times as much glued area in the direction of the force applied. I think. It has been a few years since physics classes. If it took 350# of force to shear the joint with straight line force, it will take much more (I would think) to cause the same rotational shear. I don't think any plastic fitting will take that much abuse.
|
Post Reply
|
Author:
vic (CA)
Sum .... I love it. Thanks so much for sharing your results/experiences so far.
|
Post Reply
|
Author:
hj (AZ)
quote; The only thing I can think of is again to reduce to depth of penetration
And thus NOT test a "proper" PVC joint.
|
Post Reply
|
Author:
vic (CA)
I'd also love to see tests done using different brands of solvent welding. In other words grey versus clear versus Red Hot solvents and so on from different manufacturers.
Sum are you up for the task?
|
Post Reply
|
Author:
sum (FL)
hj,
I agree. If the pipe is not fully seated into the socket, then it's not a "properly" made joint.
Right now it looks like the torsion test is not going to happen because there is simply no way to tear the joint apart with the parts being PVC. One could argue that this is to be expected since they "fused" together.
So the next thing to consider is, is there any value in testing this if the conditions are not perfect? For example pipe cut a bit too short so they are not fully seated, or someone used a hand saw and the pipe is not at 90 degree but at a skew causing one side to seat fully and the other side not so...
If I can test a joint seated say half way in and can document the amount of torque to tear the joint apart when only cement is used, versus primer + cement, versus cleaner + primer + cement, is it useful?
Let's say the numbers say that a primed joint seated half way into the socket is stronger than an unprimed joint seated half way in, can one conclude that the same result - relative speaking - will occur if it's seated all the way in? I am inclined to think so.
|
Post Reply
|
Author:
sum (FL)
vic,
I was asking my son yesterday while we were doing this trial run. We used regular clear cement because it seems to be the most popular.
I then asked him if he intends to cut across the joints and look at the "seams" on the joints in a microscope, just to see how much softening goes on with the primer. Perhaps with the primer he would see a "wavy" pattern of the fused plastics. He said yes, I then asked him how he would be able to see the joint clearly if clear cement is used...so we started to think may be we ought to try Rain or Shine or Red Hot Blue glue. So that if the cement is applied WITHOUT primer, you can see the seams clearly.
We also debated about testing CPVC pipes.
The limiting factor is we have to make do with using rigged together DIY contraptions. Not the most professional.
|
Post Reply
|
Author:
hj (AZ)
It will give you a comparative value, but not an actual one. For example the unprimed one MIGHT fail at 50% of the primed one, but if they were proper joints, that might still be more the tested primed joint.Then too, the results may be irrelevant if the failures are well beyond any value the joint would be subjected to in the "real world".
|
Post Reply
|
Author:
KCRoto (MO)
This is a test to see the effects of having primed versus unprimed joints. 'proper' is irrelevant in the scope of experimentation for the purposes of empirical data. After data from different depths of fitting is collected it could be graphed and extrapolated to get results that would normally only appear after mechanical failure of the plastic. You and I both know you aren't going to change how you think of and perform plumbing due to the results of an experiment. In the interest of knowledge itself, I am curious to see the results even if you (hj) want to fuss about it not being 'proper'.
Edited 2 times.
|
Post Reply
|
Author:
sum (FL)
Since a PVC fitting socket is tapered, one would expect the deeper the insertion, the "better" the joint, and probably the strength will not increase linearly with the depth of insertion. It should be interesting to see.
The real world is not very pretty. I happen to think the pros posting in this forum are heads and shoulders above the average plumber...I know because otherwise I wouldn't have found my way here to seek advice after being repeatedly disappointed by the quality of work I got from the pros I hired.
In the real world I see people stepping on sprinkler line glued 5 minutes prior, and I have seen plenty of people not bottom out the sockets in a fitting because they didn't have enough room, or put in a 90 elbow and on one side they couldn't do the quarter turn to spread the cement because the pipe is already fixed.
|
Post Reply
|
Author:
sum (FL)
We did another trial tonight. This time with CPVC pipe as we think perhaps the PVC fittings may be stronger than the PVC counterpart. If I put a wrench on the hex shaped CPVC male adapter it may not chew it up.
So what we did was a 90 elbow, a short piece of pipe, and a male adapter.
After two hours of curing, I put the elbow on the vise and applied some torque to it.
The result.
The entire thing twisted off. The joint didn't break, the pipe did, and I didn't have ANY pipe length in between.
This seems to suggest it was truly solvent welded right? The portion inside the fitting is twice as thick (pipe thickness plus fitting thickness), if it's one piece then the weakest section is the pipe in between.
Edited 1 times.
|
Post Reply
|
Author:
KCRoto (MO)
I guess that would lead back to a modification of the original thesis: Does glueing a pvc joint without primer really create a solvent weld at all? without primer it may just be glued pipe. If you look at the wiki for solvent welding it states that surface preparation is the first step to a solvent weld. It cites iso 472 for this, so it would appear by definition that without primer, the joint isn't solvent welded. So perhaps it comes down to glued v. solvent welded joints: Is there a substantial difference?
|
Post Reply
|
Author:
sum (FL)
I am going to have my son repeat this test with CPVC with primer and without primer, have them cure for 24 hours and repeat the test and see.
|
Post Reply
|
Author:
hj (AZ)
quote; Since a PVC fitting socket is tapered,
IF that be the case, then a "slip coupling" would not make a good test because it has to have a constant diameter in order to slide all the way onto the one pipe.
|
Post Reply
|
Author:
sum (FL)
hj, I was told (by someone here) PVC fittings are tapered ever so slightly so the fit is more snug as you push deeper into the socket. Is this not true?
I have wondered about slip couplings or repair couplings, I assume they are not tapered, or may be they are but makes no difference when the PVC is being softened by primer and cement.
Theoretically, to simply compare primed joints versus unprimed joints, I don't have to use pipes and fittings. I can simply use two pieces of square PVC "plates" or "tiles", and compare the results. As long as all the conditions are identical, I think we should be able to get some interesting results to chew on, knowing the exact parameters of the tests.
|
Post Reply
|
Author:
sum (FL)
Took another trip to the Home Depot.
My son and I spent a bit of time looking at all the different cement.
We noticed that even all the cement are made by Oatey (except for Christy's red hot blue glue), but the directions on blue lava hot glue, rain n shine glue says "No primer needed on non-pressure DWV, where local codes permit." whereas the rest of them like regular cement, medium duty gray, heavy duty etc...do not say that. Very interesting. So we will test a bunch of different cement too.
|
Post Reply
|
Author:
vic (CA)
Interesting. Thanks sum.
Maybe start a new thread rather than continuing on this one as before we know it this one will be on the next page.
|
Post Reply
|
Author:
hj (AZ)
At the rate it is going, it will not only be ON the next page, but it will BE the next page.
|
Post Reply
|
Author:
sum (FL)
Vic, I plan to start a new page once we get to the next milestone.
The project is due in December, so the plan is to perform the tests some time in October, and have most of November for him to analyze results, prepare reports, and possibly contact manufacturers for comments if results surprise us.
September is kind of mobilization, do a few trials, get the right tools, parts, have all the various test scenarios laid out. We are getting to the tail end of that.
Once we finalized on how many types of cement, what size pipes, varies depths of insertion into the fitting sockets, we will start a new thread with that as a starting point.
For example, we are still debating if we ought to include pvc cleaner into the test, to use before priming. However we are using new piped, so cleaners may do nothing. We are also wondering if we should do two primer tests, one with the cement applied immediately after primer is applied, and the other being wait a few seconds for it to "work". We saw some plumbers said the primer needs a few seconds to soften the pvc before brushing on the cement, that may help to find out.
My son will be doing all the work, I will assist him. The only thing I will do for him is cutting the pipes with a chop saw.
|
Post Reply
|
Author:
vic (CA)
Thanks sum. We can all learn so much from each other and you're such a fantastic resource for us all.
One thing to consider .... primer most likely will prove to be great for aiding strength to solvent welding however it MAY be that it weakens the pipe a bit.
So I suggest that you not assume that pipe is the same strength when using primer. For fun maybe something else to test. Pipe breaking point by itself versus pipe with having placed primer on it (and nothing else).
|
Post Reply
|
Author:
sum (FL)
Vic, would be happy to test the effects of primer on a pipe.
I presume what you meant was that when primer is applied it softens the pipe and fitting, when it hardens again, there may have been some changes in the properties of the pipe that makes it not as strong as it used to be.
How to test that is a challenge.
Any suggestions?
One possible test I thought of, is to dip one end of a pipe into the primer container, let it coat inside and out. Let it cure. Subject the pipe to a compression test, and see if the primed end crumbles first?
Edited 1 times.
|
Post Reply
|
Author:
vic (CA)
I'm no expert on how to test the strength of pipe and so on however possibly a crude way I was thinking is that I might take a piece of pipe (not sure what length, maybe at least 3 or 4') and have one end in a fixed stand/place and then take the other end and move it one way until the pipe breaks. Then take another piece and in the middle place some primer on it for maybe a few inches, let it dry and do the same thing.
How to test the pressure as well as safety equipment needed is beyond my expertise.
My hypothesis is that the pipe might be weaker after having primer placed on it.
|
Post Reply
|
Please note:
- Inappropriate messages or blatant advertising will be deleted. We cannot be held responsible for bad or inadequate advice.
- Plbg.com has no control over external content that may be linked to from messages posted here. Please follow external links with caution.
- Plbg.com is strictly for the exchange of plumbing related advice and NOT to ask about pricing/costs, nor where to find a product (try Google), nor how to operate or promote a business, nor for ethics (law) and the like questions.
- Plbg.com is also not a place to ask radiant heating (try HeatingHelp.com), electrical or even general construction type questions. We are exclusively for plumbing questions.
Search for plumbing parts on our sponsor's site:
Special thanks to our sponsor:
|