DFU to GPM 2

[CivilTyro]

Hi all,
I am working on a project where I am trying to calculate gpm from each building(SFR, Condo, Restaurant, Hotel, Bar etc) and convert to cfs to size the sewer lateral for each building and sewer main lines on proposed onsite streets to finally calculate total sewer flow rate from the project to a proposed lift station. I have received DFUs for each building from the Architect. I am having hard time calculating gpm for each building from the given DFUs. When I look up the plumbing code, I see 1 DFU = 7.5 gpm and this seems very excessive. I have also seen 1 DFU = 1 gpm or 0.5 gpm in my online research.
I have seen the earlier posts, I have not gotten a simple answer for this. I am not concerned about internal plumbing lines, and their sizes but rather the sewer lateral that connects the buildings to main lines.
This project is big and my total DFU count is 5,275 and I would like to make sure I am calculating the numbers correctly so I don't oversize the sewer infrastructure.
I am used to applying a sewer generation factor for each land use an calculate the gpm/gpd/mgd/cfs for the project, but this is my first time dealing with calculating flow based on DFUs.

Any help here is appreciated. Thanks.

 

[fel3]

CivilTyro…

All the plumbing codes I have seen over the past 40 years show a non-linear relationship between the number of fixture units (both WSFUs and DFUs) and peak flow (Qp), so neither 1 DFU = 7.5 gpm or 1 DFU = 0.5 or 1.0 gpm is correct. Because the relationship is non-linear (which I will discuss more below) the sum of the buildings' peak flows does not equal the peak flow for the entire development.

I don't have a plumbing code handy, just some notes and a few copied pages in a binder, but I think the WSFU-Qp and DFU-Qp relationships are the same mathematically, it's just that buildings almost always have different WSFU and DFU counts, so their peak water and sewer flows are different. I hope someone else can verify or correct this point. I mostly work with water, so my binder is skewed that way.

Also, below 1,000 WSFUs (and DFUs???), it matters whether the lavatories use flush valves or flush tanks, with flush valves having higher peak flows. BTW, from here-on-out, I will mostly discuss WSFUs, with the assumption that the same thing or something very similar also holds true for DFUs

The recent plumbing codes I am familiar with all use a graph to show the WSFU-Qp relationship up to 5,000 WSFUs. Fortunately, the 1979 ICBO Plumbing Code has a matching table for this relationship and this makes curve fitting much easier. For the data in the table from 1,000 WSFUs to 5,000 WSFUs, I derived the following regression curve: Qp = 2.106 * WSFUs^0.6638, with r = 0.99967. I came up with this curve years ago for the first project I worked on that had more than 5,000 WSFUs (a 5,000-bed state prison). This curve allowed me to extrapolate well beyond 5,000 WSFUs as well as interpolate between 1,000 and 5,000 WSFUs.

The way to use the WSFU/DFU-Qp relationship is to first determine Qp for each building from its fixture unit counts, then size that building's water and sewer services accordingly. For the sewer system and for a stick-type water system, size each pipe based on the numbers of WSFUs and DFUs that are upstream. For a loop water system, I prorate the total peak water demand based on the WSFUs per building.

So, let's say you have four buildings, upstream to downstream, with the following fixture unit counts:
- Bldg 1: 50 WSFUs ( 50.0 gpm) and 40 DFUs ( 46.0 gpm)
- Bldg 2: 100 WSFUs ( 67.5 gpm) and 80 DFUs ( 61.2 gpm)
- Bldg 3: 200 WSFUs ( 90.0 gpm) and 160 DFUs ( 77.0 gpm)
- Bldg 4: 1,000 WSFUs (208.0 gpm) and 750 DFUs (177.0 gpm)
-----------------------------------------------------------
- TOTALS: 1,350 WSFUs (252.0 gpm) and 1,030 DFUs (210.6 gpm)
(as you can see the development totals are only about 60% of the sum of the buildings' peak flows)

Also, Bldgs 1+2 = 150 WSFUs (79.0 gpm) and 120 DFUs (73 gpm) and Bldgs 1+2+3 = 350 WSFUs (~119 gpm) and 280 DFUs (~101 gpm).

After all this, it's important to remember that the Fixture Unit Method is an estimate and may not always tell the correct tale. For example, years ago I did the site civil design for a 40,000 sf medical office building next to a hospital (this was for my one developer client). This building was in full operation Monday through Friday from 8:00 a.m. to 5:00 p.m. and partial operation from 6:00 a.m. to 8 a.m. and 5:00 p.m. to 8 p.m. The only thing open 24/7 was a small pharmacy. My client was a curious sort, so he replaced the head on the water meter to one with real-time digital data recording. Imaging our surprise when we found that the weekeday peaking factor was something like 40:1. I had guessed 10:1.

Any way, I hope this helps.

Fred

==========
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[bimr]

In addition to what fel3 posted, the local building code probably specifies minimum sizes for sewer laterals. You should have a discussion with the local AHJ.

 

[CivilTyro]

fel3,
Thanks for the great reply, although I did not follow your formula, what is r?

 

[fel3]

r = correlation coefficient. 0 = no correlation; +1 or -1 = perfect correlation.
r^2 = coefficient of determination

==========
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[fel3]

I just revisited this thread and realized that my use of "upstream" isn't entirely accurate. In fact, it only applies to the sewer system, which is a stick system. For a stick water system, the upstream end is opposite that of the sewer system. I wanted to clear this up for some future reader who may get confused with my explanation.

==========
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill

 

[CarlB]

fel3,

The flow vs WSFU for the 4 buildings fairly closely matches plumbing code/Hunter curve numbers, for a system with flush valves. But it doesn't match the equation you provided, but kind of matches flow for non-flush valves, can you elaborate? Thanks!

 

[fel3]

CarlB…

I derived the above equation using data only from 1,000 FUs to 5,000 FUs for the purpose of extrapolating beyond 5,000 FUs. I would not expect my equation to match the chart below 1,000 FUs. This is because there are two curves on the chart below 1,000 FUs (flush valves and flush tanks) and shapes of both curves do not appear to be the same as the curve above 1,000 FUs.

I have attached the data I originally used. The first page is the fixture unit vs peak demand chart from the 1997 Uniform Plumbing Code. The current California Plumbing Code (my jurisdiction) uses the exact same chart. It is possible to scale the curve above 1,000 FUs to derive a power curve for extrapolation, but scaling a small chart isn't the most accurate method and it's time consuming. Instead, I used the table in the second page, which is from the 1979 ICBO Plumbing Code. It matches the chart, but it was much easier to work with.

I hope this helps.

Fred

==========
"Is it the only lesson of history that mankind is unteachable?"
--Winston S. Churchill