Printable View
Dear All,
I am doing a 150kWr 30% propylene glycol system. Primary pump is pumping from a 1000lit buffer tank to chiller & the fluid for process (secondary circuit) is picked up after the chiller. Wondering if its best to tap the return lines from secondary pumps (process) into the buffer tank itself or at the suction of primary pump near the buffer tank. Any suggestions? not sure if it makes much difference to chiller performance !!! :confused:
regards
GS
Hi GBRD
If you are using a single tank it's probably best to bring the return from the secondary back to the tank directly over the suction of the primary pump and the return from the heat exchanger back to the tank directly over the suction of the secondary pump. This gives you the coldest brine to the process pump suction and the hottest brine to the heat exchanger pump suction.
The tank will have some variation in temp but this shouldn't cause a problem. You can still control the chiller on tank temp.
Paul
Hi, GBRD :)
we use to install a kind of closure (to prevent too much mixing of primary and secondary liquid, in the same time tank became more rigid;)) with holes big enough to keep the same level in both parts of tank ... piping is the best as explained by PaulZ ...Quote:
Originally Posted by GBRD
don't connect return lines direct to suction of primary pumps, you need buffer tank as expansion vessel
Best regards, Josip :)
If I understood what you meant you will pass the glycol return (hottest temperature) from the services directly to the heat exchanger which would increase the heat exchanger efficiency and also on the the aircoolers (due no existing of mixing). I also do it like this way. Notice that if you use only the secondary pump for all series circuit you will need more head pressure for the same glycol flow so as you say you should connected the pumps in series on the way you describe.
...but the buffer tank must be pressurized or else the pumps must the connect in series mounted one downstream the other.
Check the hydraulics, maybe you will need a separate pump.
Well, there is only one common primary pump at the tank, no matter how many secondary pumps are there till flow of primary pump is more than secondary pumps altogether. I do agree with Paulz, but agree with Josip too. What Sandro has mentioned is something I was thinking of that sending the hottest fluid directly to chiller would increase it's performance. However, there is no need to pressurize the tank...
I do not know what split you are looking at, but the buffer is so very small, even with a 10C diff on your stat you only have a few minutes of buffer. Go Bigger. You do not really want to look at any more than a maximum of 6 starts an hour, on equipment like this.
What you should have is a "split buffer" basically a wall in the middle, cold form chiller into one side, drawn from this side to the process. warm from the process into the other side, drawn this side to the chiller.
the wall does NOT go all the way to the top. Any inbalance in the flows, flows over the top of the wall. Or you can use 2 tanks, similar set up, but you have a "large" balance pipe between the 2 (full flow with little to no pressure drop)