Water demand using UPC and other approaches

[Mech110]

Hello everyone,

I am calculating water supply demand for a 3B+G+9 commercial+residential building. I calculated from UPC 2015; total fixtures 5815 which translates to 650 GPM. I have been reading a lot of articles and discussions on this forum as well as other sites. But my confusion is still there,

1. What does this figure of 650gpm calculated from UPC represent? Is this average demand of the building per minute or peak demand at any one time? does this also include diversity factor?

2. Pipe sizing and water tank sizing should be based on directly this figure 650 GPM or something else has to be done?

3. If I use average daily consumption of water per person (100 gallons per person per day), the water demand is too lower than 650 GPM. Similar is the case of using any other approach like area density or per unit dwelling quantity. Why SO MUCH difference in these approaches? Which approach should be used then? is UPC ok? safe? 650 gpm ok?????

Thanks in advance!

 

[PEDARRIN2]

I would recommend looking at ASPE (American Society of Plumbing Engineers) documents/references.

https://aspe.org/sites/default/files/webfm/ContinuingEd/PSD_CEU_177May11.pdf

https://www.aspe.org/sites/default/files/webfm/ContinuingEd/CEU_229_Nov15.pdf

Typically water supply demand calculated by methods in plumbing codes are peak demands which does include statistical diversity. In the US, it is based on studies performed by Dr Roy(?) Hunter in the early to mid 20th century. The Hunter's curve is what was and still is utilized.

But in many people's minds, it is overly conservative and there has been a push to refine or revise his approach due to better knowledge of the statistical use of fixtures. But it takes money to fund studies and that has not been forthcoming.

Then there is the issue of lower flow fixtures. But with the exception of toilet room lavatories and showers, the flow rates have not changed, just the flow duration, i.e. toilets and urinals. They still flow ~25 gpm, but the gallons per flush has decreased.

I usually calculate the flow rate per the Hunter's curve, then use a factor of 0.6 to 0.7 to account for these changes.

However, you need to consider the type of facility. I don't know what a 3B+G+9 commercial+residential building is specifically, but if you have a lot of showers which might be used early in the morning, or if you have a lot of large toilet rooms, i.e. sporting event or convention center type use, which might be used in a fairly short period of time, you have to take that into account because it could make your Hunter's curve peak load too small.

In my opinion, pipe sizing should be based on the max of these two flow rates.

Sizing water storage, however, should not. Unless you have meter information from a similar installation, you will likely ovesize the storage. I typically size water storage by the published gpd of facilities and add ~10% for safety factor.

Hope that helps.

 

[Mech110]

@PEDARRIN2

Thanks a lot for your reply. I read both of the above articles. Quite informative. I recalculated using IPC, it gives me a difference of approx. -167 gpm from UPC. I think I can go for 600-650 gpm to be on safer side.

Also, i now understand that both IPC or UPC give peak demand but including diversity/probability of simultaneous use so that the water supply pipe sizes are good enough to meet the peak demand whenever it happens.

Building type is a typical shops+restaurants+residential apartments building.

 

[PEDARRIN2]

With the residential use, you might have a higher peak flow due to the typical morning shower use. For this, I would look at about a 70%-80% diversity of the shower flow, i.e. total number of showers x (1.5 gpm to 2.5 gpm) as the potential peak flow. See which is higher. Also remember lower flow showers need a bit more pressure to actually seem like there is water flow. This may be the deciding fixture for your booster pump.

And if the restaurants serve breakfast, you will need to account for that in the peak load, since it will likely occur at the same time as the morning shower.

 

[Mech110]

@PEDARRIN2

Well it's getting more and more confusing now. Hard to reach any final satisfactory demand. So many methods, so many approaches. I wonder why UPC and IPC standards are silent on these issues. According to your suggestion, I have 382 showers which means 764 GPM considering 2GPM each.

I also calculated according to ASHRAE modified Hunter Curves. My Residential Apartments fixtures are approx. 4800 which gives 300 GPM. My Restaurants WFU is ~100 which gives around ~50. So total 350 GPM.

I also used the following link, the latest publication by IAPMO:

http://www.contractormag.com/codes/iapmo-develops-water-demand-calculator-residential-buildings

Used the calculator by IAPMO, my residential demand comes out to be 250 GPM. Assume 50 GPM to be the commercial load.

So the peak water demand is ranging from 300 GPM to 764 GPM according to different methods. I want to bang my head on wall!

WHICH METHOD TO ADOPT?

 

[PEDARRIN2]

The reason in my opinion the UPC and/or the IPC are silent is they are code, not engineering design guides (mostly) The UPC (which I don't use much as I work in IPC areas) includes water pipe sizing in the actual code whereas the IPC has it in an appendix.

And the ASHRAE modified curves are for hot water only and are used to calculate a balance between hot water generation and storage, but really have nothing to do with flow.

The IAPMO (which is the base organization which publishes the UPC) publication is for residential (single and multiple (4 max?) family dwellings. I would hesitate to use it for a larger situation because I don't know that it would just be a multiplication factor.

I had a hotel I designed that had ~530 rooms, three restaurants, a pool, and other associated items. My water supply fixture load was 8633 and my domestic water service was 8", so a higher peak flow rate (~800 gpm) is not out of the question.

 

[Mech110]

I used both IPC and UPC as I stated above. IPC gives even a lower flow than UPC. IAPMO's calculator just restricts its use for Residential but has not talked about the scale. Any number of fixtures can be entered. However, I am also not willing to rely on it at the moment. 530 rooms of hotel means 530 bathrooms which comprise the major load. In my case, I have 382 bathrooms demanding the major chunk of flow. So, by all this discussion, I think I can go with 650-700 GPM max. flow demand.

Did you use a tube-well (bore) to save water in storage tank? Could you also give your expert opinion on this thread:

https://www.eng-tips.com/viewthread.cfm?qid=441867

as once again I am getting multiple opinions on that.

Thanks once again for your kind support.

 

[PEDARRIN2]

My application did not have a storage tank. It was supplied from a municipal main.

 

[dbill74]

I am curious to see your calculations Mech110.
You mention getting a range of 300-764 GPM at one point. I have to admit I find it hard to believe UPC and IPC would generate such a large difference.

Keep in mind that when using fixture units and the Hunter's Curve, there is a built-in diversity factor. Do a goolge search for Hunter's curve and look at the images that come up, it is not linear. This takes into account that with more fixtures in a system, the diversity is going to go down.

Pedarrin mentioned using a diversity factor or 70-80%; IMHO this wuold probably be good for up to about 100 bathroom groups, for 382 bathrooms I think a better diversity factor would be closer to 45-50%.

Going back to your original questions:
1. What does this figure of 650gpm calculated from UPC represent? Is this average demand of the building per minute or peak demand at any one time? does this also include diversity factor?
A: It is the peak demand at any one time. Yes it does include a diversity factor. (The building will NOT use 650 GPM 24/7.)

2. Pipe sizing and water tank sizing should be based on directly this figure 650 GPM or something else has to be done?
A: Yes and no. This figure is used to size the building main, what connects to the street and what the project's civil engineer needs.
You can use the same method for sizing the mains within the building as well. But as you get to indivual fixtures and groups of fixtures you need to look at the individual fixtuers.
Water tank sizing should NOT be based on this figure alone. For tank sizing you need to consider what the tank is for and how it will be used. Other factors will be how quickly you intend to refill it, where will the tank go and what structural support is available/needed.

3. If I use average daily consumption of water per person (100 gallons per person per day), the water demand is too lower than 650 GPM. Similar is the case of using any other approach like area density or per unit dwelling quantity. Why SO MUCH difference in these approaches? Which approach should be used then? is UPC ok? safe? 650 gpm ok?
A: Think about it. Is each shower going to be used continuously throughout the day? Are the toilets being flushed every other minute?
Using daily consumption per person number is appropraite for helping to size a water tank, but as I mentioned above, not the only factor. Your peak demand is used to size the piping coming from the tank. It may even be appropraite to require the tank be able to provide the peak demand for a limited period of time (30-45 minutes for example for peak shower usage). Before you can size your tank, you need to define its function and what it needs to serve.

 

[Mech110]

Thanks for your reply. Very helpful.

 

[tbonebanjo]

Pipe sizes are based on peak flow, the maximum flow expected in the system. We need to know the function of the tank in order to help with sizing parameters. Is it an elevated gravity tank being fed from a water supply? If so, then the make-up flow rate to the tank vs the hourly usage need to be taken into consideration. Most of the buildings I work on do not have a supply tank. We typically use a pressure booster system from the municipal street main. If the building is on a well or the street main cannot keep up with the building demand, then storage is required.