The finishing.com Hotline: Serious Education ... plus the most fun you can have in metal finishing. Ted Mooney, Webmaster
PVC glue breakdown
At what "temperature" does a standard PVC plastic glued joint come apart? The application is in a water source heat pump system that supplies 85 degrees water to a heat pump unit and returns 95 degrees water to a cooling tower. If the cooling tower goes down and the water source heat pump unit continues to run, the water temperature to the unit increases, at what temperature will the glued joint break/fail? This sequence has happened over a period of several times a month for several years. Does the cooling down and heating up process of the joint over a period of time cause the glue to breakdown and fail? Any and all comments are welcomed.
Thank You,Jim Faiella
consulting - Winter Park, Florida, USA
I think that you are asking the wrong question. Fixing the cooling tower is the correct answer. You did not say how high the temp would go to. 120F is about as high as you would want to go with regular PVC. Plastic has a rather high coefficient of thermal expansion, so if the pipe is ridgidly secured it is going to be stressed. You can replace the pipe with CPVC (costs a lot more) It can be glued, or glued carefully and welded or welded. I would certainly put a temperature sensing alarm in the system. The high temp can not be good for what you are calling a heat pump. Copper pipe may be an option and stainless certainly would work as far as standing the temperature changes and high temp.James Watts
- Navarre, Florida
Hi Jim !
The joints should NEVER come undone if, IF they were properly cemented.
Yes, temperature can cause a fairly large expansion or contraction (3.5 x l0-5 per degree F) but where that could happen, then use expansion joints and/or allow bends to flex, i.e., never put restraining clamps close to the bends.
Methinks that something is wrong with your cementing process. Go either to the fin.com library ... or just type my full name down on Google for an article on how NOT TO cement!
One last point. You live in a HOT area!. Assuming that this piping is outdoors, I'd try to use a white PVC instead of the grey Sch. 80 PVC ... then the sun wouldn't affect it so much, eh?
I have only come across this phenomenon once in my life ... but there some twit used a 'cleaner', i.e., Varsol instead of a PVC primer... but that was in the 60's.
The max. max. operating temperature for uPVC pipe is 140 ° F. But going to CPVC gives you a higher operating temperature but the thermal expansion is going to be very similar to PVC.
White Rock, British Columbia, Canada
(It is our sad duty to advise that Freeman passed away April 21, 2012. R.I.P. old friend).
With respect to the question the glue will never fail with correct priming and application as the glue (in simple terms) is PVC suspended in a hydrocarbon base, so as the hydrocarbon evaporates from the joint area the PVC remains completing the bond, but be warned, there are glues and there are good glues and due to some really average results that I have personally been involved with (monumental failures) I now only use and recommend TANGIT glue and primer for critical applications. Use the hardware shop, garden variety glue at your own peril.Jeremy Gray
- Melbourne, Australia
Interesting input! I was involved ONLY with the top end, the best cements ... and never ever would have considered an ordinary store type 'glue'. .
On cementing large dia. pipe, 8" to l2" dia., it's interesting to note that after ONE YEAR, if one makes a slice in the belled, sic. cemented end, those minor gaps, so necessary for a decent interference fit, are filled solid BUT you can run a screwdriver along those gaps and they will indent. We used to pressure test the cemented connections at a fairly high psi. I forget? 400 or 800 short term psi. In those early days of plastics (l960-1) we made our own cement BUT bought THF solvent in 45 gal. drums from Dupont as in those days most 'suppliers' of cement were not using THF ! The lathe offcuts supplied the 'body'!
White Rock, British Columbia, Canada