I am a structural engineer in the mid-west and have a project that would put snow drift load on an existing structure. The existing roof construction is bar joists @ 4'-0" o.c. +/-, bulb-tee purlins @ 32 5/8" o.c., 1/2" gypsum board, 2" poured gypsum, 2" insulation & BUR. I'm looking for resources that would let me determine capacity of this system for additional snow loading.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Capacity of a gypsum roof 1
- Thread starter vobs
- Start date
Gypsum strength declines with the ater to dry gypsum ratio.
wc (water/dry gypsum) Limit Tensile strength
0.45 22.7 kgf/cm2
0.5 17.7 kgf/cm2
0.6 14.7 kgf/cm2
0.7 10.6 kgf/cm2
0.8 9.2 kgf/cm2
1 8 kgf/cm2
Compressibe strengths are far bigger, maybe around 8 times the tensile.
The main thing here is to ensure a PERFECT impermeabilization of the roof to be kept closely watched, because water will distroy the gypsum plates, and even condensation studies need be made to avoid condensation produce the same effect. Water plus gypsum plus steel also will give severe problems the mass of gypsum here being so big.
The attached papers of the gypsum board somewhat will hamper the effective contact and I wouldn't rely on composite action.
wc (water/dry gypsum) Limit Tensile strength
0.45 22.7 kgf/cm2
0.5 17.7 kgf/cm2
0.6 14.7 kgf/cm2
0.7 10.6 kgf/cm2
0.8 9.2 kgf/cm2
1 8 kgf/cm2
Compressibe strengths are far bigger, maybe around 8 times the tensile.
The main thing here is to ensure a PERFECT impermeabilization of the roof to be kept closely watched, because water will distroy the gypsum plates, and even condensation studies need be made to avoid condensation produce the same effect. Water plus gypsum plus steel also will give severe problems the mass of gypsum here being so big.
The attached papers of the gypsum board somewhat will hamper the effective contact and I wouldn't rely on composite action.
-
1
- #3
jheidt2543
Civil/Environmental
Lutfi,
I know of at least one contractor in Illinois that still does this type of work, so if you need a contractor, I can give you his info.
Thanks in advance.
I know of at least one contractor in Illinois that still does this type of work, so if you need a contractor, I can give you his info.
Thanks in advance.
The system you described was very popular in the late 50's -60's. Unfortunately, if any water got into the system, you might as well replace the gypsum deck because it is usually worthless after it gets wet. The other popular roof deck material was "Tectum".
jheidt2543
Civil/Environmental
Here are a couple of URL's with information on gypsum roof deck systems from a quick Google search just to get you started. I did not download the .pdf file myself so, I really don't know how on target that one is.
Two manufacturers of gypsum bulb tee systems were W.R. Grace Co. (Zonolite) and National Gypsum (Gold Bond). You might try their websites or call their Technical Services Departments for information.
Good luck!
Two manufacturers of gypsum bulb tee systems were W.R. Grace Co. (Zonolite) and National Gypsum (Gold Bond). You might try their websites or call their Technical Services Departments for information.
Good luck!
- Thread starter
- #6
Thank you everyone for responding, but the info doesn't really address my question. I need some resource for determining the capacity of this roof system for additional snow loading due to drift.
I'm aware of what can happen if water gets into the system & that is another concern. I'm not sure yet I want to add load to this roof that will later melt and possibly get into the deck. I know this roof is not "Tectum" so that's not a concern.
Jeidt2543 maybe your contractor contact can help. If you could provide his info, that would be my next step. My e-mail address is:
dwaine.voboril@davisdesign.com
Thanks
I'm aware of what can happen if water gets into the system & that is another concern. I'm not sure yet I want to add load to this roof that will later melt and possibly get into the deck. I know this roof is not "Tectum" so that's not a concern.
Jeidt2543 maybe your contractor contact can help. If you could provide his info, that would be my next step. My e-mail address is:
dwaine.voboril@davisdesign.com
Thanks
vobs,
I certainly appreciate your concern. My thoughts are that, if you haven't already done so, you first see where you are considering the capacity of the joists. Odds are they aren't going to be adequate for additional load. (One fix I've seen used is to add joists in alternate (every other) deck spans to theoretically reduce the tributary width to each joist.) If the cost of the fix for the joists can't be justified, you're done; otherwise you'll still need to figure out what it takes to reinforce the gypsum construction for the added load and, unfortunately, I can't be helpful on that aspect without doing a good bit of digging myself.
sto04078
I certainly appreciate your concern. My thoughts are that, if you haven't already done so, you first see where you are considering the capacity of the joists. Odds are they aren't going to be adequate for additional load. (One fix I've seen used is to add joists in alternate (every other) deck spans to theoretically reduce the tributary width to each joist.) If the cost of the fix for the joists can't be justified, you're done; otherwise you'll still need to figure out what it takes to reinforce the gypsum construction for the added load and, unfortunately, I can't be helpful on that aspect without doing a good bit of digging myself.
sto04078
- Thread starter
- #8
Well, maybe I was not clear enough. I simply wouldn't use gypsum as a structural material. But if someone wants to use gypsum to his risk, I would recommend the limit strengths above, using only the new thickness, maybe a safety factor of 3, (that is, the values aboved divided by 3 would be the allowable stresses) and determining the stresses elastically as plates or beams.
- Thread starter
- #10
It's not my choice to use this material, but it is part of the existing building that would get additional load due to snow drift due to a proposed addition.
Now if i understand you correctly, at best I could get 22.7 kgf/cm2 tensile strength which equals 323 psi approximately. With a factor of 3, that gives me 107.667 psi for an allowable tensile stress. Let's just say 105 psi. Now assuming the 1/2" gyp board plank is acting alone (non-composite), the most it can carry on a 32.625" span would be 4.75 psf. Self weight of the decking system itself is 28-35 psf depending on how much balast is present. Something else is happening here. While your info may be technically correct, it's not getting me any closer to a solution.
Now if i understand you correctly, at best I could get 22.7 kgf/cm2 tensile strength which equals 323 psi approximately. With a factor of 3, that gives me 107.667 psi for an allowable tensile stress. Let's just say 105 psi. Now assuming the 1/2" gyp board plank is acting alone (non-composite), the most it can carry on a 32.625" span would be 4.75 psf. Self weight of the decking system itself is 28-35 psf depending on how much balast is present. Something else is happening here. While your info may be technically correct, it's not getting me any closer to a solution.
craigory28
Structural
Vobs,
I have come across this many times before. I'm just going to address the joists since everything else seems to have been beaten to death.
I first determine the live load capacity of the existing joist. Our state code used to have a minimum LL of 30 psf. With IBC that value can become a bit lower depending on where you are located. I calculate the new drift load per IBC and taper it down to the new IBC LL. Hopefully the new LL is less than what the joists were originally designed for. I then compare shear and moment diagrams. If the new load falls into the existing I will say ok. (I would assume if you are some percent i.e. 5% over you might be able to say ok but I have never found this anywhere. I will lead this up to engineering judgement.) However as someone stated earlier it will probably not.
If it does not here is what you can do:
- go back to the original joist manufacturer and get their fix. There usually are joist tags at the end of ever joist. There is usually a cost to this and lots of expensive welding. I have not found an owner yet willing to pay for this.
- We typically add new joist between the existing. I would call a joist manufacturer to talk about this and get some comfort level with it. I typically design the new joist to take all the load and ignore the existing. (It won't matter on cost.) Also when adding joist in existing a couple things you can do are have them bolted at midspan so installed easy, smaller joist seat so they can slide into position then shim as required, and apply regular X bracing to brace the top chord. (Remember the top chord will not be attached to deck.)
You will also have to look at what the joist is bearing on. Typically if it a wall the load increase will not be enough to cause a problem. A steel beam you can weld extra material as required to get it to work. A joist girder might cause some problems and require you to go back to the original designer.
I have come across this many times before. I'm just going to address the joists since everything else seems to have been beaten to death.
I first determine the live load capacity of the existing joist. Our state code used to have a minimum LL of 30 psf. With IBC that value can become a bit lower depending on where you are located. I calculate the new drift load per IBC and taper it down to the new IBC LL. Hopefully the new LL is less than what the joists were originally designed for. I then compare shear and moment diagrams. If the new load falls into the existing I will say ok. (I would assume if you are some percent i.e. 5% over you might be able to say ok but I have never found this anywhere. I will lead this up to engineering judgement.) However as someone stated earlier it will probably not.
If it does not here is what you can do:
- go back to the original joist manufacturer and get their fix. There usually are joist tags at the end of ever joist. There is usually a cost to this and lots of expensive welding. I have not found an owner yet willing to pay for this.
- We typically add new joist between the existing. I would call a joist manufacturer to talk about this and get some comfort level with it. I typically design the new joist to take all the load and ignore the existing. (It won't matter on cost.) Also when adding joist in existing a couple things you can do are have them bolted at midspan so installed easy, smaller joist seat so they can slide into position then shim as required, and apply regular X bracing to brace the top chord. (Remember the top chord will not be attached to deck.)
You will also have to look at what the joist is bearing on. Typically if it a wall the load increase will not be enough to cause a problem. A steel beam you can weld extra material as required to get it to work. A joist girder might cause some problems and require you to go back to the original designer.
- Thread starter
- #13
As long the existing gypsum board is capable of sustaining the new added gypsum, (2"
it is this what I was counting on supporting new load, mainly. You may resort to more ordinary and lower safety factors, and calculate the plate(s) as supported on 4 sides. Using a 3 factor is just showing my disrespect for such downgradeable material put to structural use. Attainable and inherent uniformity of the gypsum surely allows (in disregard of its chemical potential bad behaviour) a far more ordinary safety factor from just a mechanical viewpoint.
Respect the inadequacy of the material I think most of those here coming are not thinking in chastising those who ask -even if sometimes may appear so- but precise what they think is correct. Everything I have written about is tantamount to say that it is maybe better to leave such job to other. Before that, one tries to convince the deciding parties of the inconvenience of following their stated intent, offering other alternatives. Then if not reasonable I would walk out of the thing. I have found some unreasonable people between the owners, maybe 10% a bit and 1% amazingly UNreasonable, but almost no one of those within the trade.
it is this what I was counting on supporting new load, mainly. You may resort to more ordinary and lower safety factors, and calculate the plate(s) as supported on 4 sides. Using a 3 factor is just showing my disrespect for such downgradeable material put to structural use. Attainable and inherent uniformity of the gypsum surely allows (in disregard of its chemical potential bad behaviour) a far more ordinary safety factor from just a mechanical viewpoint.Respect the inadequacy of the material I think most of those here coming are not thinking in chastising those who ask -even if sometimes may appear so- but precise what they think is correct. Everything I have written about is tantamount to say that it is maybe better to leave such job to other. Before that, one tries to convince the deciding parties of the inconvenience of following their stated intent, offering other alternatives. Then if not reasonable I would walk out of the thing. I have found some unreasonable people between the owners, maybe 10% a bit and 1% amazingly UNreasonable, but almost no one of those within the trade.
VOBS,
Typcally, the poured gypsum deck is about 2" to 3" thick, (not 1/2" as you previously mentioned) and is supported by bulb tees. I have generally seen the tees spaced about 30" to 36" on center.
According to Fields Corp., the compressive strength of the gypsum is 500 psi. If you know or can determine the actual thickness, and whether or not it is reinforced, you should be able to calculate the allowable loads.
Another way is to use the load charts, for a given span and thickness, back figure the allowable stress.
I would be surpised if the allowable load on the deck exceeds about 40 psf.
Remember, the alloawble stresses are only good if the deck has not been exposed to water. Once the gypsum gets wet, you have nothing left.
Typcally, the poured gypsum deck is about 2" to 3" thick, (not 1/2" as you previously mentioned) and is supported by bulb tees. I have generally seen the tees spaced about 30" to 36" on center.
According to Fields Corp., the compressive strength of the gypsum is 500 psi. If you know or can determine the actual thickness, and whether or not it is reinforced, you should be able to calculate the allowable loads.
Another way is to use the load charts, for a given span and thickness, back figure the allowable stress.
I would be surpised if the allowable load on the deck exceeds about 40 psf.
Remember, the alloawble stresses are only good if the deck has not been exposed to water. Once the gypsum gets wet, you have nothing left.
- Thread starter
- #16
CSEllc
Refer to my original post.
"The existing roof construction is bar joists @ 4'-0" o.c. +/-, bulb-tee purlins @ 32 5/8" o.c., 1/2" gypsum board, 2" poured gypsum, 2" insulation & BUR."
I would assume the 2" of poured gypsum is reinforced, but there is nothing on the original plans to indicate that is the case.
I agree with your statement that gypsum is worth 500 psi in compression. What I'm not sure about is the strength of gypsum in bending. The load charts is exactly what I'm looking for.
Anyway, we have decided to remove the gypsum deck in the area of the drift load and re-roof with metal deck, insulation and an adhered EPDM. The maximum drift load is 85 psf and the roof is spongy in areas, making us suspect water has gotten into the system.
However, if someone has design information, including load-span tables for such a system, I would greatly appreciate it. Thanks.
Refer to my original post.
"The existing roof construction is bar joists @ 4'-0" o.c. +/-, bulb-tee purlins @ 32 5/8" o.c., 1/2" gypsum board, 2" poured gypsum, 2" insulation & BUR."
I would assume the 2" of poured gypsum is reinforced, but there is nothing on the original plans to indicate that is the case.
I agree with your statement that gypsum is worth 500 psi in compression. What I'm not sure about is the strength of gypsum in bending. The load charts is exactly what I'm looking for.
Anyway, we have decided to remove the gypsum deck in the area of the drift load and re-roof with metal deck, insulation and an adhered EPDM. The maximum drift load is 85 psf and the roof is spongy in areas, making us suspect water has gotten into the system.
However, if someone has design information, including load-span tables for such a system, I would greatly appreciate it. Thanks.
- Status
Similar threads
- Locked
- Question
- Question