Hello Mohammed,
They have to do with convergence of the network, as you've noted. I've particularly found them useful when valves and check valves are preventing convergence from occurring on large models with a lot of pressure zones and facilities with control setpoint pretty close to each other, but I try to exhaust other options before adjusting these. They can effectively give the model more time to converge. There's a good summary on this forum posting:
http://coffee-table.4894995.n2.nabble.com/Unknown-properties-in-Hydraulic-Option-td7081021.htmlThe commercial packages include some additional documentation on using these values; there's some details in the very popularly viewed post here for the damping limit parameter. There's some specific recommendations for these parameter values for troublesome models in the third post down the page. (The engine referred to in this posting is EPANet, so it's still applicable):
http://forums.innovyze.com/threads/122-System-Unbalance-Question?highlight=damping+limitI think the reason it's missing from documentation is that looking over the source code, it was added in a later version. Here's the notes on the usage in the source code (hydraul.c):
"Status checks on CVs, pumps and pipes to tanks are made every CheckFreq iteration, up until MaxCheck iterations are reached. Status checks on control valves are made every iteration if DampLimit = 0 or only when the convergence error is at or below DampLimit. If DampLimit is > 0 then future computed flow changes are only 60% of their full value. A complete status check on all links is made when convergence is achieved. If convergence is not achieved in MaxIter trials and ExtraIter > 0 then another ExtraIter trials are made with no status changes made to any links and a warning message is generated."
Thanks,
Brian