Its been a while since I have used WaterCAD but I believe it works in the same manner at Boss International WaterNetworks which is what I typically use.
In the past I have run the model in one of two ways:
Method 1 - Use Reservoir and Pump combination - Place the reservoir elevation at the invert elevation of your tie in point. Connect the reservoir to your tie-in location with a pump. The pump would be defined by a three point curve developed from your fire flow data. Place the appropriate
roughness coefficients on both existing pipes and proposed pipes of the model. Then place the appropriate demand, say Flow 1062 gpm for FH1 (with no demand on FH2) and then 998 gpm to FH2 (with no demand on FH1) to validate your model against your fire flow test. If these check then your pumps are adequately defined to throttle the water from the reservoir to your system (in addition to assued roughness coefficient values). You can not move on to place your expected demands to verify system adequacy.
Method 2 - Use a reservoir only - Place the reservoir at the expected elevation connected by a 'dummy pipe' to your system
The reservoir elevation based on FH1 would be:
85 psi x 2.31 = 196.35 ft
Therefore 196.35 ft + 2.5 ft (approximate elevation from FG to nozzle) + FG elevation = Reservoir Elevation Water Level.
Once the reservoir is defined based on this information, the place the appropriate roughness coefficients to both existing and proposed pipes. Then place the demand of 1062 gpm to verify the expected residual pressure is obtained. If the residual pressure is not obtained, then adjust the roughness coefficient as necessary to obtain the residual pressure from the test. Once it checks, then your system is ready for the service demands to be tested. You can develop your reservoir definition the same way based on FH2 but I would only use it as a check, you can run the test with one reservoir alone.
Regardless of which method you use, verify that your static head + nozzle height + FG elevation falls within the city's or county's tank operating water levels. If at all possible get the tank's water elevation on the day the test was run to further verify your static head numbers. This will ensure that you validate your test data as much as possible.
From the data you provided, it appears to me that there is quite a substantial drop in static head from FH1 to FH2. Now, I don't know the topographical features of your site or where these hydrants are located, but if this was a flat site I would be concerned that there may be either a bust in one of the readings or that there is more than one water system on the road, street, which each hydrant connected to a different system or tank.
Another item to look out for is to verify that the fire hydrant location where the fire flow was run matches the address for the hydrant you are showing on your model. I just completed an fire line analysis last week, when the agency that ran two fire flows at two different locations, had given me the wrong address for one of the hydrants. Furthermore, and even funnier they had no record of they themselves running the second test. When I requested to verify location they gave a surprised look and asked why. I explained I was just dotting my i's and crossing my t's. When they realized I had a fire flow result for a test they had ran themselves and they had no record of, the said "can we make a copy?" and then said "thank you".
Also don't forge to define minor losses on your system. By the way, I tend to prefer to use method 1 but I do use both depending on the site size and complexity.
