Pressure drop (and system curves) for diverging branches

[@brendan714]

Nice! Only thing to say is that your negative flows don't indicate zero flow, but rather backflow from your destination :)

[@ProcesswithPat]

Fair! I guess I will need to pull that magical non return valve out my arse in response!

[@brendan714]

@@ProcesswithPat check valves to the rescue!

[@eronpaulo23]

But it will not give the flow that the polynomial is indicating!

[@ProcesswithPat]

Eron, that is critical input, thanks for pointing it out! The polynomial fit starts at that minimum pressure for each respective curve, not necessarily at zero pressure. You can fit any order polynomial you like and each one will give different negative values below the threshold value, so they are not indicative of the reverse flow rate.

[@ghmh]

Venturi effect?

[@gabrielamaral1900]

Really nice video! KZbin is kinda scarce on this subject when it is about system curves themselves. Could you talk next about closed loop networks (with one or two loops) with curves? Like the urban cold water pipe networks. Thank you in advance!

[@kyang9233]

I'm a mechanical engineer as well. I'd like to thank you. Your contents are great and brilliant. But why isn't your channel well known?

[@monikaklier6051]

Hey, thanks for the videos. Keep them coming, it's great content. Have a nice day!

[@panoskb9405]

You are an engineer that I would love to meet or work with. Your videos are so amazing. You cover all my "doubts"/thoughts.

[@ProcesswithPat]

If I’m making a video about it, it’s pretty much guaranteed it’s because I had the same doubts that I’ve clarified for myself. I want to speed up the process for others to do the same… Thanks for the supportive words.

[@MegaMarrero504]

This great! What size pump would you select for this? In your example at the end would be it 200 m3/hr at 3.5 bar?

[@3dprintwiz378]

Been trying my best to flip the axis on Google Docs, but for the love of God , I can't figure out how to do it. Can you please point me to the right direction.

[@banyatniyomvas]

Thank you

[@garymcilwaine5840]

Hey Pat. Love the videos. Starting as a process engineer in September. Is there any books that you would recommend for process engineers other than Perry's Handbook? Thanks again for the great content.

[@ProcesswithPat]

Good luck! For a non-technical read about engineering I would recommend Haldor Topsøe’s biography. If you want something that is technical but doesn’t read like a text book I’d go for Norm Lieberman’s Working Guide to Process Equipment.

[@garymcilwaine5840]

@@ProcesswithPat Thanks Pat. Just ordered the book by Lieberman. Looking forward to reading it soon. Thanks again.

[@zahraalgallaf6347]

I guess the original & destination points should be at fixed P

[@stevenreason7040]

This video is only correct if there were non return valves in each line, right? Below 2 bar you would have reverse flow from the header

[@ProcesswithPat]

Correct! It’s an important point, and that is why I have pinned an earlier comment by somebody else.

[@Bruh-vp6qf]

If you increase the opening of the valve to atmosphere, would the pump providing the pressure be able to maintain constant pressure to the branch point even though flowrate will increase? I'm curious about dynamic scenarios and if/how upstream pressure changes due to increased flow from a valve opening on a branch. Surely greater flow and greater pressure drop means a pump will not be delivering the same pressure?

[@brendan714]

The pump will have to adjust according to the pump curve - more flow means less head from the pump.

[@Bruh-vp6qf]

@@brendan714 Ok thought so which means iterating the calc based on flow and pump curve till the final flow is converged?

[@brendan714]

@@Bruh-vp6qf yes I think the way Pat is doing it with spreadsheets you would have to iterate by choosing an arbitrary delivery pressure, calculate the total flow you get from that pressure, use that total flow to then go back to the pump curve and see what pressure the pump gives, use that pressure to re-calculate flow, etc etc until it all converges. The other option would be to solve it using a process hydraulic simulation software like AFT Fathom, if you input all the details of the model (pump curve, line lengths, fittings, valve details, downstream delivery pressures), it will automatically iterate and calculate your flow rates. Pat's method is very much a hand calculation. Sometimes the systems get so complex that you can't solve it by hand and you need software to help. But it's always good to know how the calculation works, especially so you can do a quick sanity check when the software dumps out data.

[@Bruh-vp6qf]

@@brendan714 yep agreed just wanted to make sure I had the right thinking.cheers mate