5 replies to this topic

[jango]

I’ve recently run into a design question regarding our cooling tower return line. We want to occasionally bypass the cooling tower by sending the water directly back into the basin instead of through the tower packing. The goal is to avoid over-cooling the process water during winter. I’ve attached a sketch of the setup— the bypass line is shown in red.

 

My question concerns the vertical drop on the return side (shown in blue). If I extend this vertical section farther downward, will the additional static head improve flow through the bypass line and allow me to select a smaller control valve? In other words, does increasing the vertical drop actually provide more driving head, even though additional pipe length also introduces more frictional pressure drop?

 

This feels counter-intuitive since longer piping usually means more pressure loss, but gravity should help in a downward line. Is it valid to expect increased flow from simply extending the pipe further down?

 

Attached Files

•  picture cooltower.png   16.23KB   0 downloads

Edited by jango, Today, 03:10 AM.

[breizh]

Hi,

Let me share my experience about cooling water equipped with similar set up. I've been production manager in a Soda Ash plant for 10 years in East part of France. The weather is quite cold in Winter, Hot and dry in Summer. The cooling water capacity was about 3500 m3/h.

We have similar set with a by-pass line from the return line to the basin. The flow was managed with a control valve based on the temperature after cooling. The idea was to prevent to shock the crystallizers using cold water, in winter. 

This was not a good idea, because there is risk to damage the packing (collapsing) due to ice formation and excess load.

In our case we solved or minimized this risk operating the fans at a lower speed. Check the possibility to use a VFD or mechanical means.

My advice, keep the flow running.

Breizh 

[jango]

Hi Breizh,

 

Thank you for your answer, fans are already speed controlled/shutoff depending on temperature of the coolingwater but this is still not enough. At the end of the year we run also at a lower capacity due to inventory reasons and have a lot less cooling requirements and the VSD en stopping the fan is just not enough. Idea is bypassing about 30% of the total liquid to the basin, I'm not worried about ice formation of the packing as there is still a large load going through the cooling tower. I'm still interested in the potential answer of the question for the sake of future project and myself. 

Edited by jango, Today, 05:19 AM.

[breizh]

Hi,

I guess you have multiple cooling water units with common header, why don't you isolate some of them?

About your question, I will run a separate line for the bypass. For self-venting pipe, the Froud number should be less than 0.3.

EDIT: Not knowing your role in your organization and background, be aware of risk of deposit of minerals on the packing if the flowrate is reduced too much. Check with the company selling chemicals for your cooling water.

Good luck

Breizh

[jango]

Hi Breiz,

 

Should this be a self vented pipe, as the pipe is fully liquid filled above the control valve and there is no entrained air? We also cannot have air going in the opposite direction. So self venting is not really necessary here. Is this correct? I'm a process engineer, scaling of packing is not a concern as we have very minimal hardness in our coolwater system as this is made up with deminwater/condensate.

[katmar]

For you to gain driving force from the extended drain pipe this pipe would have to run full. If you take the case where you would be diverting 300 m3/h through the bypass then the limit of where the gain in static head is exactly matched by the extra friction head would be when the ID of the pipe is 100 mm. In this situation the velocity is about 10.6 m/s and the Froude Number is 10.8.

If the diameter of the pipe is anything less than 100 mm then the extra length adds more friction loss than the static head gain. As the diameter is increased to greater than 100 mm there will be more static head gain than there is friction loss, provided that the pipe still runs full.

There are some conflicting ideas out there regarding the criteria to ensure full flow of a vertical pipe, but my experience is that for a Froude Number of 2.0 or more you can be fairly certain that the pipe will be full. If the ID is increased further and the Fr No decreases to less than 2.0 the pipe will be partly full, unless the bottom of the pipe is below the surface of the water in the basin and no air can enter the bottom of the pipe. In this situation the pipe will be full of water and you will score the full static height gain.

For all the above situations it is assumed that no air is being drawn down the cooling water return piping. If air is mixed in with the return water then there will be air in the bypass in all cases and the static head recovery will be decreased. If the initial vertical section of the return piping is designed for self venting flow then the air will be negligible.