View Full Version : Multiple Process Pumps on Chiller Supply Loop

04-14-2015, 11:10 AM
We have our chiller loop set up per the Pro Chiller diagram with a dedicated recirc pump and a single process pump going to the supply header. A VFD with input from a pressure sensor in the supply line controls the pumps speed.

The pasteurizer at the end of the line sometimes has flow issues if a lot of tanks are open. Could I put in another pump with another VFD and pressure switch for the pasteurizer? The pump would be between the drop from the supply header and the inlet of the cold HX section on the pasteurizer. It could also take a higher pressure than the tanks so the pressure setpoint could be raised.

Any issue with doing this?

Any limit to the number of pumps you can add similar to above for future tank additions that may start to get farther away from the original process pump at the chiller?

04-14-2015, 03:33 PM
I'm not an expert on this, but my suggestion would be to use a second pump running a second header that's dedicated to the pasteurizer.
I would think adding a pump onto the end of the line won't help you if there's not enough glycol flow to the end of the supply header.
That way you don't run the risk of not having enough glycol to supply the HX, I would run a dedicated line for that particular unit of the glycol supply tank. The other option I can see is turning up some tanks temporarily to increase the glycol pressure in the header.

04-14-2015, 03:41 PM
This matter of using VFDs in this respect is relatively new and in many cases is not being correctly applied.
For odd reasons, everyone thinks these days that everything needs a VFD on it.
I don't find them to be a necessarily good idea for this application unless they are set up in some very strict parameters.
A VFD should not take the place of a bypass regulator. You simply will not get the flow rates you need if the RPMs on the pump are not right for the load. The loop should be balanced with your last device on the line being taken into account.
Things that should be quite simple are being over engineered and at a cost.
We run over 500 BBLs of Fermentation on a 1 HP loop pump with 2-120s at the FAR end of the run without any problem at all. Easily 100 Ft. from the main well.
Without seeing your system or knowing more about it I would say there is a problem with the way it is configured or the way you are modulating the pump speed which is very likely too low. I'd be curious to know the pressure in PSIG on the supply header.

The booster pump arrangement you are describing is not the correct way to approach this type of problem.
If a second pump was going to be run for whatever reason it should be piped separately off the main well.

04-14-2015, 04:21 PM
We have a pressure bypass. We are doing first in last out. We have 4 x 105 BBL FV's, 1 80 BBL Uni, and the pasteurizer at the end of the loop. There is a pressure gauge at the end which I want to say off the top of my head reads around 5 PSI most of the time (the VFD is set to maintain 15 PSI, from a 3 HP 3450 RPM pump which spends most of its time between 35-40 hz).

The HX streaming diagram calls for 20 GPM of glycol with a 9 PSI pressure drop through the plates. I don't know what it actually is. When we turn off all the tanks on the loop the temperature can be as much as 5 degrees F lower exiting the HX which is what leads me to believe we are not getting sufficient flow by using the VFD to control the pump speed based only on the pressure input at the beginning of the system.

The end of the loop where the pasteurizer resides is a good 100-120' of 2" piping away from the pump. It would be a pain (and expensive) to run a separate supply line straight to the pasteurizer.