PVC Piping For Problem Chiller 1

[BenjaminM]

What are the down falls of PVC pipe for large flows?

I have been tasked with the job of re-designing a circulation loop for our 270 ton McQuay chiller. The chiller needs 640 GPM but we currently only giving it 250 GPM.

We've found a pump capable of delivering the capacity which the chiller needs. I've redesigned the loop, cutting out ALOT of extra piping and increasing the pipe diameter from 3" and 4" to nearly all 6" Sch 40 C/S pipe.

My estimated project cost is about $14k. They want a cheaper alternative. The suggestion of running it all in PVC pipe came up and everyone seems to be sold on this idea.

I have not priced PVC, but I can expect the material cost should drop appreciably. The labor cost, which was a third of my estimate will disappear because we will not need the welding contractor to do the fabrication for us.

As attractive of idea as running this line may be from a pricing point of view, I am rather gun shy of using PVC pipe for anything past a sink drain. That said I have no real foundation for this opinion.

So, has anyone had good luck with PVC? Or are we setting ourselves up for trouble down the road?

 

[ApC2Kp]

myrdale,

The PVC pipe should be OK for cooling water service. There are more plastic cooling towers that have PVC pipe in the distributors inside the units.

Now if you are considering PVC for the chilled water loop piping, then that is a different service. The PVC material would be more brittle at low temperatures (less than 40F) and some people add heat tracing to avoid the problem of brittle failure of PVC. I would avoid chilled water in PVC.

Support spans will be shorter with PVC pipe. You might want to increase PVC diameter to Nps-8 to reduce velocity and the startup slug loads?

 

[ash9144]

Chilled water is usually ethylene or propylene glycol there may be compatibility issues.

I would be leary of the glued joints in a pvc circuit. Startups would be difficult and I would expect at some point the line would crack from the slugging.

 

[786392]

I don't know the low temperature range;

but the non-ductility of PVC in this context must be explored.I assume.

Best Regards
Qalander(Chem)

 

[moltenmetal]

Return water temperatures and excursions thereof are also a concern. PVC is limited to 60 C maximum service temperature, and CPVC to about 80 C (some push it a bit farther than that, but it's noticeably softened already at 80 C). We've made that mistake before, and had some saggy, nasty-looking return water piping. Never lost containment, but needed a LOT of support to avoid looking like spaghetti...

CPVC at 6" is also pretty spendy.

If indeed it's not a cold glycol/water service where the brittleness of PVC would make it worrisome, or a service where you could see high return water temperatures, PVC can definitely be a reliable and cost-effective solution.

If either of these two no-nos are encountered, consider Victaulic grooved-end fittings. Welded pipe for cooling water is overkill.

 

[Montemayor]

myrdale:

PVC will work - with some reservations, engineering details, and trade-offs. Be aware that PVC probably has one of the highest expansion coefficients around -- just ask any vinyl siding contractor!

You can experience some rather severe ruptures in your piping network if you don't take the precautions to install sufficient off-sets and/or expansion loops. The stuff will contract and expand such that it will a lot of careful detailed engineering. This, together with the added supports, the strength of bonded joints, and the loss of mechanical integrity are going to be some of your main issues when considering trade-offs you will have to accept. Yes, as you might have guessed it, there ARE trade-offs, and you might as well discuss them out in the open before anyone that recommended the PVC expresses "surprise".

 

[rconner]

You may also be interested in the perspectives also expressed by others in the concurrent thread at

http://eng-tips.com/viewthread.cfm?qid=240882&page=1

and/or page 6 case at

http://www.lcwlegal.com/newspublications/PDFs/ISM_2008_11.pdf

etc. (toughness etc. with respect to future construction in congested areas could also be a consideration).

 

[BenjaminM]

Thankyou all for the responces. The temperature range we are looking at is around 40*F. Even under full load, I can not say I've seen the temperature above 60*F maximum.

Montemayor's point about the expansion of PVC is part of my concern. However the pipe runs are not very long, they will not have very much length to grow. I have a run of about 15 foot horizontal and a run of probably 10 foot or so verticl. There will be a couple of 90's on both the supply and return. Am I correct in believing the 90's will absorb some of the growth / contraction? I will have a couple of pipe stands to support the weight. Let me see if I can post a sketch of the layout tomorrow.

I am also curious if we should heat trace the line. It does get into the lower teens a couple nights of the year in NW Georgia. However we leave the pump circulating continusly. I'll have the line insulated, but that will be less for freeze protection and more to prevent heat loss.. well rather gain.

Thanks for the links rconner!

 

[StoneCold]

I have some 6" PVC in cooling water service, but it is free draining from cooling towers into tanks. We have had some joint cracks develop over the last 6 years of service. I would not use it in a pressure piping situation. I would spend the money and go with Red thread fiberglass pipe. I would also avoid the CS if at all possible. Refineries love the CS piping but keeping the bugs killed in there can be a major chore.

Regards
StoneCold

 

[zerosum]

We didn't even use PVC pipe for our numerous chilled water circuits in our PVC (resin) plant. Expansion/contraction, pipe support, and UV degradation require a lot of attention to detail for a system to hold up over time.

We used welded C.S. chilled water and cooling tower water piping. There are perhaps some savings with groved-end piping. I have never used fiberglass pipe but it might make sense in your application.

 

[rconner]

I wasn't exactly sure what you meant by "large flows", so I rough took a look at the flow velocity. Even if you upsize the piping, it appears that you may be talking about flow velocity of 7 fps (2+ mps) or so. I think you would find with a quick web search that you have approached a level where many authorities, apparently based on experience, appear to be uncomfortable with this kind of velocity in pvc piping.