calibration hammer model

Question

anyone have an experience in calibration Hammer model? 
 
 Thx

Tom Walski · Accepted Answer

For that short pipe, the pressure wave is traveling back and forth about 200 times per second. So you would want a pressure sensor that responds on the order of one reading every 0.0005 s. I personally don't know of any equipment that does that but hopefully you do. 
 
Similarly, when you run your HAMMER model, you'll need to use a time step on the order of 0.0001 s which means you would need to set the duration of a run to 1 s or else it will take a very long time to run. 
 
This all sounds very challenging. 
 
What most researchers do is use a much longer pipe. If you had 300 m of pipe, you could get readings on the order of one every 0.05 s which is feasible and there are devices that can do this. 
 
What some researchers have done to get a long pipe into a small lab is buy a spool of small diameter HDPE pipe and stack it up. Copper tubing would also work but that would be more expensive. You may want to look at the papers of Bruno Brunone and others who have done this kind of testing. 
 
Good luck.

Sushma Choure · Answer

Hello Fadel, 
 
 Could you please provide us more information on what kind of transient analysis or calibration do you want to perform using Hammer? 
 Here is some useful information for you to get started with Hammer, if you haven&rsquo;t gone through it already. 
 http://communities.bentley.com/products/hydraulics___hydrology/w/hydraulics_and_hydrology__wiki/10348.learning-resource-guide-for-new-user-of-bentley-hammer-v8i 
 http://communities.bentley.com/products/hydraulics___hydrology/w/hydraulics_and_hydrology__wiki/3031.hammer-general-faqs

Scott Kampa · Answer

Hello Fadel, 
 
There is a model calibration tool available in WaterGEMS, which uses the same file types that HAMMER uses. This tool is called Darwin Calibrator. You can find more information on Darwin Calibrator here: communities.bentley.com/.../5910.using-darwin-calibrator 
 
Darwin Calibrator is not included in HAMMER because it is assumed that you have a working, calibrated model before you run the transient analysis. The following link has information on other steps you can take to prepared you model to run in HAMMER: communities.bentley.com/.../23535.preparing-a-model-for-analysis-in-bentley-hammer 
 
Regards, 
Scott

Tom Walski · Answer

Scott's advice is correct if you are trying to calibrate the underlying network model which you need before you get into the transient analysis. We also have a great deal of material on such calibration in our book "Analysis of Water Distribution Systems". 
 
If you are trying to calibrate the transient aspects of a HAMMER run, you need data collected at a very fine time resolution. Data from an ordinary SCADA system or data logger may not be sufficient. There are special loggers than record pressures at a fraction of a second time interval that you would need. 
 
Once you get this data, if the model doesn't match the field data, you adjust model input to match the field data. One of the first things that users adjust is wave speed since the actual wave speed is often different from the theoretical speed.

Tom Walski · Answer

It's pretty much impossible to measure transient pressure on a 3 m long pipe. 
 
The pressure wave would have travelled 180 times back and forth in a second. 
 
Unless you have some amazing pressure and flow meters, it's doubtful that your calibration measurements are very accurate.