Subscribe to TIC

Everything Moves

I’ve heard carpenters and trim installation contractors complain that PVC trim expands and contracts too much. My comment to them is: Yes, cellular PVC trim does move, but so do all other exterior building products, and many of them just as much as, if not more than, cellular PVC.

A Note from the Publisher:


If you are sensitive about articles that seem to favor a particular manufacturer, then DON’T read this one!! It’s written by the President of Versatex! But after years of experience, I’ve found that some manufacturers know more about their products than anyone else; if a carpenter wants to learn the best way to install a product, sometimes the best source of information is the manufacturer. In the future, look for more carefully-screened articles from manufacturers.


The fact is, movement can be caused by different forces of nature. For wood, wood composites, and fiber cement, movement is all about the moisture content of the product. Wood expands and contracts with changes in the surrounding humidity and, to a lesser degree, the temperature. More humid air will cause wood to expand, while drier air will cause wood to contract.

Wood does not move in all directions equally. In fact, the greatest movement will always be across the grain. If you read the installation instructions for fiber cement, or composite wood siding and trims, you’ll see the manufacturers recommend gapping between boards. Why would you need to gap something unless it’s going to move?

Coefficient of Thermal Expansion

For building products made from cellular PVC, aluminum, steel, or other polymer-based materials, it’s all about the temperature at the time of installation compared to the temperature swings the product will experience throughout the year.

You need to learn how to deal with this phenomenon, and I’d like to tell you how without getting too deep in the weeds.

First of all, let’s discuss product movement. Just about every material—be it a natural resource, or man-made product—has a coefficient of thermal expansion. Wow! Those are some big technical words. So, what does it mean?

The coefficient of thermal expansion describes how the size of an object changes with a change in temperature. Specifically, it measures the fractional change in size per degree change in temperature at a constant pressure.

There are several types of thermal coefficients: volumetric, area, and linear. Which one is used depends on the particular application, and which dimensions are considered important or critical to the material. For solids, like cellular PVC, one might only be concerned with the change along a length, or over some area. Some common coefficients of thermal expansion for some standard building product materials are:

For cellular PVC trim, as well as most exterior building products, the focus is on linear movement, since movement along the product’s length is what needs to be controlled—especially where there are long runs of trim. If your cellular PVC trim is going to move, it will be most noticeable in the fascia, frieze, or rake boards on a house.

Determining Amount of Movement

Before getting into how to best control movement, let’s look at how to determine the amount of movement for a given set of conditions. Let’s say we’re installing cellular PVC trim when the outside temperature is 50° F. The boards are 18 feet long, and the house is in Maryland, where the temperature can reach 100° F.

To determine the maximum amount of linear movement, we need the coefficient of linear thermal expansion for cellular PVC, which is 0.000032 in/in-F, the length of the board in inches (216), and the maximum temperature swing the product will be exposed to during the year—in this case, 50° F. The formula to determine movement is as follows:

The Change in Product Length (unrestricted) = The coefficient of linear thermal expansion for cellular PVC x the length of the trim x the maximum change in temperature or (Temperature at time of installation – Maximum Temperature product can reach on any day during the summer).

Δ Length = 0.000032 in/in-F x 216 in x (100° F – 50° F)

Δ Length = 0.3456” (unrestricted) or 0.1728” (when properly nailed) which is between 5/32 in. and 3/16 in.

When I say “properly nailed,” I’m not talking about an 18 gauge or even a 16 gauge trim nail. We recommend an 8d, 12 gauge trim nail.

Now, you’re probably saying: I can’t get 8d, 12 gauge nails that I can gun. Yes, you can. Swan Secure (now part of Simpson Strong-Tie) makes such a nail that is branded their “TRIfecta” nail. These nails come in a strip and are collated, so there is no problem gunning the trim to the framing members of the home.

Further movement can be reduced by using an adhesive in combination with the fasteners. For instance, gluing the cellular PVC fascia board to the sub-fascia with Liquid Nails sub-floor or heavy duty construction adhesive can reduce the board from expanding or contracting.

Best Installation Practices

Here are some “best installation practices,” given to us by contractors and remodelers with years of experience putting up long runs of cellular PVC trim (i.e. fascia, rakes, frieze boards):

  • Screws restrict movement more than nails
  • If you can bend the fasteners you plan to use to secure your trim between two fingers, they are too thin.
  • If practical, you can further restrict movement on long runs by reducing the on-center fastener spacing to 12 in. A good example here is a fascia board where there is a wooden sub-fascia allowing a tighter on-center fastener spacing.

  • Shiplap joints offer a superior joint to scarf or miter cut joints. They increase the adhesive surface area while also aligning the face of the boards, thereby preventing any offset.

  • Allow the cellular PVC trim to acclimate to the outside temperature before installing. If possible, install any long runs on a house when the outside temperature, and the temperature of the cellular PVC trimboard, is between 60° and 65° F.
  • Double-fasten on both sides of any board-to-board joint using the recommended number of fasteners based upon the width of the board (see image, right).
  • Pick inconspicuous spots away from sight lines for expansion joints that will compensate for any movement in the cellular PVC trim.
  • Southern exposures, or areas where the product is in direct sunlight, can result in slightly greater product movement due to the heat gain potential for the trim in these areas.

So, there you have it. Everything you wanted to know about why cellular PVC trim moves, but were afraid to ask. I hope the information and recommendations provided here help you with your cellular PVC trim applications, thereby making installation easier, and providing a finished project that meets or exceeds the homeowners’ expectations.

• • •


John Pace has more than 20 years of experience in the design, development, production and installation of rigid and cellular vinyl building products for residential and light commercial applications. He is a founder of Wolfpac Technologies, Inc., an extruder of cellular PVC sheet and board materials that has been serving the building products industry since 2003. He is also the President and Chief Operating Officer of VERSATEX Trimboard, a subsidiary of Wolfpac Technologies, Inc.

While many companies have recently implemented environmentally responsible strategies, John has been a driving force in the implementation of green practices for decades. The company was recently awarded the Green Seal of Approval from the NAHB for meeting the requirements of certain mandated practices specified in the National Green Building Standard.

John regularly collaborates with customers and designers in the field, and maintains strong associations with the Vinyl Siding Institute, the Moulding & Millwork Producers Association, the National Wholesale Lumber Association and the National Coil Coaters Association.

John holds a BS in Civil Engineering from Lehigh University, as well as a Professional Engineers License.


38 Responses to “Everything Moves”

  1. Aaron

    Just want to say thanks for the consistently fantastic content on here.

  2. John pace

    Content is from a wide range of builders, contractors and remodelers along the east coast. We went to the source to make sure we provided up to date information on the proper installation techniques for cellular PVC trim.

  3. Gary Katz

    I’m glad you enjoyed this story. It was a step in the ‘different’ direction. I have always learned a lot from John and thought other carpenters would, too.

    Thank you for contributing. As I suspected it would be, this is an illuminating article.

  4. John

    We have been using the cortex screws. They hold great and look fantastic but the price and slow installation speed are significant enough drawbacks to cause me to give those Trifecta nails a try.

    • John

      In the event that the TRIfecta nails are overdriven, what is the best product to fill the nail holes (finishes smooth, possibly sandable, won’t squeeze or pop out…)?

    • John pace

      Do you paint your cellular PVC trim? If you try not to paint it the best fastening system is the cortex screws and plugs. The plugs are typically made from the manufacturers sheet product and are a perfect color match. if you’re painting your PVC trim contractors on Long Island use DAP’s Crack Shot or DAP’s Fast N’ Final. They are easy to use, have very little to no shrinkage, are easy to sand, and paint without flashing. There are a host of other nail hole fillers that can be used with cellular PVC these are just two of the more common ones I’ve seen on projects.

  5. woodworkbykirk

    this is great, in the last 2 years i think all but one deck ive built has been clad with this type of trim and the decking was composite. at break time we always had debates on what the nailing pattern should be as to going easy on nails to let it move so it doesnt break or to nail the snot out of it so it cant move..

    this really clears things up

    • John pace

      Here is another tip I learned from some of the better contractors. To help control board movement, especially in long fascia, frieze and rake board runs, use 2 fasteners across the width of a nominal 4″ and 6″ board, 3 across the width of a nominal 8″ and 10″ board, 4 fasteners across the width of a nominal 12″ wide board and 5 across the width of a nominal 16″ wide board. Fasteners, whether they are nails or screws should be spaced a maximum of 16″ on center along the boards length. I don’t think it’s right to promote a products brand in a forum like this. However, Versatex has a back pocket contractor handbook specifically for cellular PVC. It tells you the best sealants, adhesives and fasteners to use with this product along with some installation tips from the experts. You can contact Versatex at 724-857-1111 or and have a handbook mailed to you.

  6. David

    I’d like to use PVC for a high exposure garage door jamb. Is this possible?

    • John pace

      Cellular PVC trim is an excellent product for use as a garage door jamb. It won’t rot, it’s durable and in a new construction application you can run the trim below the level of the concrete floor and pour up against it. I know several contractors that use cellular PVC when forming for a concrete pour. It’s great when you have a radius in the pour because it bends so easily.

  7. Paul Young

    I’m always looking for a good adhesive for cellular PVC to wood framing.
    John recommends Liquid Nails Sublfloor or Heavy Duty construction adhesive.
    PROBLEM is manufacturers Tech Data sheet has PLASTIC LUMBER and PLASTIC, VINYL in the Not Recommended sections.

    • John Pace

      Paul. I received the information on Liquid Nails Sub Floor and Liquid Nails Heavy Duty Construction adhesive direct from the manufacturer. I’ll send you a copy of their e-mail addresses to me next week when I’m back in the office. Other adhesives or sealant adhesives that have performed well when bonding cellular PVC to wood are OSI’s new TRIMTeQ™ TeQ Mount™ Mounting Adhesive, and Solar Seal #900 by NPC. I prefer the Solar Seal because of its excellent weathering characteristics and flexibility. There are numerous others but these are just two other products that perform well. I believe you will find their tech data sheets or their technical support center approves using their products to bond cellular PVC to wood.

  8. Sonny Wiehe

    Your article contains some fantastic information; particularly the chart showing coefficient of expansion of familiar building materials. This material phenomena is something I only formerly learned about only during my architectural education; not in the field. However I did not know the exact linear relationship between cellular PVC and wood and that was eye opening. I mention this because while we carpenters do learn a great deal just from empirical feedback of working with materials and putting them together, we often do not have the opportunity to study or observe their behavior over time or over the course of different temperature cycles (I suppose we just don’t like hanging around the client’s house that long after getting paid). We may only learn of this phenomena when there is a problem like a leak or sprung joint which is (hopefully) rare.
    Anyhow, while you show the chart, I think it is important to highlight the fact that PVC has a coefficient of expansion that is 10X’s that of wood! And aluminum coil (which many of us have made a transition from as a trim cladding product) is 40X’s! This chart shows that the thermal coefficient of wood is comparatively negligible and that has been why this issue has been, well… neglected. With cellular PVC being more widely used than ever, this contrasting relationship can no longer be ignored.
    I think we have all easily seen the results of poorly detailed aluminum cladding (buckling from over restrained fastening) or maybe pieces falling off (under fastening). I believe it’s easier for carpenters to understand the thermal behavior of aluminum (and its 40Xs factor) being an issue because of its metal (hey, it gets super hot in the sun dude!) and skin-like application. But when it comes to using cellular PVC it appears (and is marketed) to most as working like wood (cutting and routing mainly). I think that is where the “10Xs” factor of PVC should be made more widely known. This way carpenters can be more proactive in avoiding issues instead of avoiding the material all together.

    • John Pace

      Sonny your response is spot on. I used to manufacturer trim coil back in the 80’s and 90’s under the brand name Napco. The biggest issue with wraping a wood fascia board with aluminum coil was the difference in the expansion and subsequent contraction between the two substrates. If the coil was painted a dark color you could expect buckling. Since cellular PVC also moves but to a lesser degree than aluminum coil, steps need to be taken to minimize board movement. Those steps include reducing heat gain (lighter paint colors), installing the trim at an ideal temperature relative to the products coefficient of thermal linear expansion and through an aggressive fastening schedule. Thanks for your response. I thoroughly enjoyed reading it. John

  9. Joe Stoddard

    Another good article- and I for one appreciate the disclaimer Gary . Always good to understand the source of the info- it doesn’t diminish it, but does keep things in perspective.

    My ( rhetorical) question – wouldn’t devising a blind fastening system that allowed the PVC movement to happen in a controlled way ultimately be a better solution? It seems to me one well-milled and consistent ship-lap joint every 12-16′ is going to look much better over the long haul than imperfect rows of fasteners gunned in every 12-16″ , maybe streaking, maybe filled with caulk or other filler that is capturing grime from the air/rain 24/7.

    • Gary Katz

      I’m glad you enjoyed the article and the ‘notice.’ I know some folks are sensitive about ‘conflicts of interest,’ so I wanted it out front. John knows more about PVC trim than anyone I’ve met. I can’t imagine a better source for this story. In fact, you’ll see that he DOES recommend ship-lap joints toward the end of the article.

  10. Emanuel

    Hi John, Great article with lots of good tips. When I install my PVC trim I use pocket screws to join two lengths together with pvc glue. I also fasten them with screws, 4 to 5 per row every 12″. Then I insert plugs into the holes cut from pvc scraps which I then cut smooth with a Tajima saw of 18 tpi. I’ve been doing this for a while and had great results. Thanks again for a great detailed article.

    • John Pace


      Your installation techniques are outstanding and what I consider the best of the best. You should have written the article. If you don’t mind I’d like to share your ideas with other contractors. Thanks for your input.

  11. Emanuel

    Hi John:

    Thank you for the kind words, that means a lot coming from you. I would be honored for you to share my ideas with other contractors. I am interested in using your product for upcoming projects and I was hoping to get some more information from you. Thanks for your time

  12. Sam Marsico

    Great Article, thanks for posting.
    Has anybody else had problems with the Hitachi NV 65 coil nailer? (The one that shoots the Trifecta nails). Mine has misfired or double fired and jammed ever since it was new. I have used both the plastic and wire collated coils, in stainless and galvanized.
    I really like the cortex screws, but there is a learning curve. They don’t like to start, stop, and start again because the head is so small and they easily strip. I have given up on my 18 volt tools when using cortex and bought a cheap drill driver with a keyless chuck. The variable trigger and slightly higher torque seem perfect for driving the cortex to depth in one shot.

    • John Pace


      Thank you for your feedback on the coil nailer and the Cortex screw and plug system. Although I spend quite a bit of time in the field, I seldom hear about issues with the equipment. If you don’t mind, I’d like to forward your comments on to Simpson Strong Tie who now owns Swan Secure and FastenMaster the producers of the Cortex screw and cellular PVC plug. If I receive any feedback from either company, I’ll be sure to forward it on to you.

  13. Mike

    I’m a little confused about managing the linear movement in long segments of cellular PVC. I am installing PVC crown molding which must be glued at the miter joints. Therefore, I am cutting a scarf joint to allow for expansion. It would seem like I’d want to fasten the crown moldings every 12″ to make sure the board doesn’t buckle, but I’dwant to predrill holes slightly larger than the fastener to allow the linear movement of the board over the scarf joint bit to keep it held tight to the substrate. It seems that I wouldn’t want to use adhesive to attach it to the substrate unless the adhesive would just hold it tight to the substrate but still allow some linear movement at the scarf joint so that the material didn’t buckle.

  14. John Pace


    Your idea of reducing the fastener spacing from 16″ on center to 12″ on center is correct. It will help to constrict the movement of the board and reduce the chance of buckling during periods of rises in temperature (summer). Pre-drilling the fastener holes to allow movement is spot on. I disagree with the idea of not using an adhesive to secure the moulding to the substrate. This is just another means of mechanically constricting the movement of the moulding whether it is contraction or expansion of the moulding.

    Keep in mind movement is an issue with long runs. By long runs I mean more than two boards and typically limited to fascia, frieze and possibly rake board applications. Most trim on a home will not experience “visible” movement due to trim or moulding length, location on the home (north and east walls see far less sun than the south and west walls), temperature at time of installation (ideal temp is 60 to 65F) and installation techniques utilized.

  15. Steve

    Great article, but it makes wood sound like it doesn’t move which we know it does. There needs to be a correlation of the thermal expansion of PVC with the expansion of wood based on humidity.

  16. John

    Steve, first of all great question. I’m not sure how to correlate the linear thermal expansion of cellular PVC with the shrinking and swelling of wood based upon humidity change/moisture content. Although both move the manner in which they move are totally different. Cellular PVC expands and contracts along its length. Wood shrinks and swells across its width. Cellular PVC contracts and expands as a function of temperature (difference between the installation temperature and the low and high temperature it might encounter) and product length. Humidity has no effect on the movement of cellular PVC trim. Wood shrinks and swells with changes in moisture content in response to daily as well as seasonal changes in the relative humidity of the atmosphere. Moisture content changes of only a few percent are sufficient to cause significant shrinking and swelling of wood. In the case of sugar maple, for example, a change of 6 percent in moisture content may result in a dimensional change of about 2-1/2 percent. In general, the amount of shrinking and swelling which takes place is directly proportional to moisture content changes in the wood. I have some good articles on the movement of both products as well as the formulations used to calculate their movement. Let me know if you’d like them.

  17. Jim

    I have seen some sites mention using spacers when placing PVC facia to cover a rim joist for a deck to promote air flow. Do you recomend these? If so how thick and what type of spacer material would you use?
    I don’t imagine this would have a big effect on the expansion /contraction issue

  18. Markus

    Are you only concerned about movement with high temperatures? What about low temperatures in a city like Chicago where you have fluctuations between -40 – +100 degrees? What is the ideal install temperature?

  19. John Pace


    The ideal installation temperature in any region is the midpoint between the high and low. Although -40 may occur for a day or so it is not a temperature that lasts for an extended period of time. The temperatures used to determine the midpoint should be the average highs and lows for your area of the country. Assuming an average low of 0F and a high of 120F, the midpoint would be 60F. Thus, the ideal installation temperature on long runs of trim should be 60 to 65F.

    PVC expands and contracts along the length. Thus the longer the length the greater the potential for movement. You should not be concerned with expansion and contraction of doors, windows, corners, etc. provided they have been properly installed with the correct fasteners since they are most likely less than 1 board in length.

  20. Don Jeanrenaud

    You mention that shiplap joints increase the surface area for adhesion, and mention use of Weldon, indicating you would want these joints cemented. And once they are cemented they wouldn’t be free to slide on one another. In addition to the cement, you state that reducing centers for nailing, and using construction adhesive between pvc and subfascia would all help. What I can’t understand is that it doesn’t seem possible to stop the expansion of the pvc molecules, I don’t think physics would allow it, so if we anchor the pvc in these ways, the expansion/contraction would still have to occur, resulting in either/or some flexing of the fasteners, cracking of the cemented shiplap, or internal stresses in the pvc. Please give me your thoughts on this. I’m planning on installing some very expensive vented fascia, and I don’t want any of it to crack due to expansion/contraction.


  21. John Pace

    Good morning Don:

    As you will note in my article, “Everything Moves”, I also addressed how best to handle expansion and contraction. So my article wasn’t stating that you can “completely” restrict product movement but that you can control it. Also, concern over expansion and contraction should not be an issue when lengths are two board lengths (36′ = one joint) or less. If it is you’re using the wrong product or the wrong installation technique. First of all, you can chemically reduce the movement of cellular PVC due to the chemical make-up of your trim. There are ingredients when added to cellular PVC that can reduce the coefficient of thermal linear expansion. Thus, not all cellular PVC trims have the same coefficient of thermal expansion. Some have values 10% to 14% higher than others. Thus, it is possible to alter this value. Work continues to find an ingredient that will have a greater impact on this value. Secondly, you can minimize the amount of expansion and contraction in cellular PVC by installing it when the board temperature is around 60F and 65F which is the mid-point between what I consider the average high and low the board may be subjected to in a specific region of the country. For example, in the northeast, the average highs and lows of a PVC trimboard are 0F (low) and 118F (high). Thus 60F is the midpoint between the high and low resulting in the least amount of expansion and contraction. Obviously, this ideal temperature will vary region to region. Is it always practical on a jobsite to install long runs of trim at the ideal temperature? Probably not. However, it shouldn’t be discounted. Does using the right fastener, adhesive and joint design help to constrict trim movement? Yes it does. On an 38′ x 38′ outdoor living area that I worked on with a local contractor, we bonded the trim to the sub-fascia with Liquid Nails Sub-Floor Adhesive, we used screws and plugs versus nails and we used the slip lap joint. Did we glue all the joints? No. First, we used two 16′ boards with the joint glued tight, and created two smaller joints at the end of the run that were not glued but were still created using the same shiplap. Thus any expansion and contraction would occur at the end of the run and in an inconspicuous area away from the main steps to the outdoor living area. We also mitered and glue the two 3′ lengths to form the corner. This assured us the miters would not open since the board lengths were so short. Finally, the structure was built in June. We had other PVC trim on this project so made it a point to install any long runs in the am before the boards heated up. Over the past 4 years, this outdoor living area has been exposed to highs air temperatures of 90F too 95F and -15F to -20F. During the summer the joints not glued are tight. During the winter the same joints will accept no more than the thickness of a piece of paper to no more than 1/64″. This is less than what the board should move based upon the length of the run and the coefficient of lineal thermal expansion. Thus, in my opinion and based upon this and other installations I’ve been a part of you can restrict movement and control but not eliminate product expansion and contraction.

    On this same project, there is modified MDF on the front of the house. These pieces are no more than 12′ long. During these same temperature conditions, the joints have gaps 1/4″ to 5/16″ wide. I’ve also seen the same size gaps on fiber cement and other wood and inert composite trims. Why am I saying this? Because as the title of the article says “Everything Moves”. Unfortunately, the expectations placed on cellular PVC is greater than on other composite trims.

  22. Kendall Everett

    PVC does cause movement but like you said, there are other building materials that do the same thing. One thing that you mentioned that I always try to keep in mind is the temperature, different materials react to varying temperatures. Sometimes it’s difficult to consider your budget with getting the correct building materials but it’s always worth it to get the correct material to begin with. Thanks for the tips!

    • John Pace

      Good afternoon Kendall:

      Excellent point on the correlation between temperature and movement. In testing done in the northeast and southeast we find a white or light colored PVC trim board facing due south will reach a max. temperature between 120F and 125F. Most regions of the country see min. winter temperatures around 0F. Sure there are a few days below 0 but not for any extended period of time unless you live in Fargo, ND or Bangor, ME. Thus the ideal temperature for installing long runs of cellular PVC trim is the midpoint temperature between the high and the low or 60F to 65F. I understand it may not be practical in the field to install your trim at this temperature. However, installers should consider installing long runs of cellular PVC in the am when the product is cool rather than during the mid-afternoon when the boards are there hottest. This will cut the expansion and contraction in half. Also, there should never be a concern with expansion or contraction of window surrounds, door surrounds or corners, or wherever the boards or sheets are two boards or less in length. I designed and acted as GC on a 36′ x 36′ outdoor living area for our home and used cellular PVC for the soffit, fascia, frieze, ceiling, columns and cabinetry. The only area were we have seen any board movement has been on the fascia and frieze. At -15F, the gap between the two boards was the width of a piece of paper. In the summer the gap closes up. The paint color on the trim was a beige tone. The gap is 1/3rd the size of the gap in the modified MDF used on the front entryway to the home where the boards are only 10′ in length. The modified MDF was on the home when I bought and is ready to be replaced this time with cellular PVC.

  23. Fred Gallo

    I am using Versatex WP4 tongue and groove trim board vertically in the North East where there is a change of temperature of 100 degrees F. From your expansion calculations in 23ft of length I could see a change of 7/8 of an inch. This does not present a problem. My problem is the width of 33′-6″ which could expand up to an inch and a quarter. If I was using a 1×6 trim board it would be about 80 boards wide. Should I leave approx. 1/16 of an inch between every trim board to allow for expansion?

  24. John Pace


    First question is are you using it for an exterior ceiling application or as a vertical siding. If its being used as a ceiling product, the temperature differential will be far less than 100F. Secondly, cellular PVC expands along its length not across its width, thus the reason the factor is called the “coefficient of thermal linear expansion”. Thus, you need not be concerned with any expansion or contraction across the width of the board. Good luck with your project. If you need any further help let me know and I’ll put you in touch with one of our field technicians in your area.

  25. Mike

    Hey John, thanks for the great article! I am in the process of finishing my basement and decided to go with pvc base molding. I completed about half of molding so far and didn’t realize that pvc trim expands more than wood. It’s about 61F in my basement right now and I made all the corners fairly tight. I attached the molding with construction adhesives and screws. My basement maybe gets to 75F tops in the summer, so a change in around 15 degrees. Based on your formulas, I should have left a gap of 1/16 on my longest run (240″). Do you think I will run into problems when things warm up?

  26. John Pace

    Good morning Mike:

    No need to worry about the base moulding. I’m confident you will not run into problems with expansion and contraction. In my 14 years in cellular PVC I’ve never seen an issue with expansion and contraction when cellular PVC is used in an interior application. The change in temperature is typically too small to create movement issues. I believe the reason is the lack of the direct sunlight (UV). Your base mould is going to warm by radiant heat. I wish I had cellular PVC when I finished the basement of our home. My Dad was a carpenter. I’ve always enjoyed building things. Unfortunately, my time is now focused more on the family (grankids), business, and educating those interested in cellular PVC. Good luck with your project.