Design for Flow

Hi,

Please do - post the image of the system for details. Makes it far easier to comment.

If I understand correctly you want to divert flow from one heat exchanger (or promote the flow into a specific heat exchanger), by installing a flow restriction in the flow with excess flow.

You mention the supply pressure is only 40 PSI. Is this a fixed pressure? What supplies this pressure? Is there a pump sending this water or is it head pressure from a tank or reservior?

If there is a pump, can you supply the pump curve for this pump?

Resticting the flow can be done in a number of ways. Installing an oriface plate is one way, and it is easy to calculate what the diameter should be. But this method is fixed control, any changes to flow / pressure etc will change the effectiveness of the control. In practive, we've installed a blank plate in the line (between flanges), and during commissioning simply take it out and gradually drill it out to achieve the correct flow.

Other more precise methods is to use a control valve (suggest globe valve) to accurately control the flowrate to what is required. This also will accomodate future changes.

Hope this helps.

Anthony

Hi
thanks for the comment.i will send you pictures of a pipe line that is branched out of a common header which supplies
sea water for coolers installed on the power generator(alternator)section.There are two coolers both on either side i e ,one on the drive end and the other on the non drive end.The line going to the drive end cooler is tapped from a 6inch line,which is 3"in dia.And the same 6" header is further reduced to 3" and goes into the non drive end cooler.Did u get my point.The inlets of both the coolers are 3" dia,but one tapped on a 6"dia pipe and the other from the same header which is reduced to 3".Here the flow of water
recorded in the drive end is 139 u s gallons/minute in the drive end,and the non drive end cooler recieves only 32 u s gallons of water. My query is1- where and what size of oriface can sort out the issue2- we just cannot stop the electric generator supplying 14 mw for plant and accomodation,so the oriface has to be somewhat precise
3- can we remove the reduced portion of piping and replace it with a 6" extension and then tap a 3" pipe on it similar to the
drive end.Will the velocity drop further? or will there be more flow4- Can you suggest a different solution or design

It seems to me that you are not supplying enough water for the distribution design you have....If it has worked in the past ( ie; this is not a new installation) then it would seem the supply has been partially compromised....If you are using seawater as you state this is probably due to mineral deposit accumulation in system...My advice would be to clean it out or replace it, or some combination, which ever is most efficient cost wise...

hell

i have uploaded some profile pictures ,but it is not featuring here.

awaiting your response

Dear Nelson,

pictures are not to be uploaded to profile, but to the post itself when writing it. Use the green camera icon for that.

Regards,

abel

please see theseprofile pictures.I am offshore and having net problem.

can you pls calculate the flow in us gallons on both the tappings on the suggested modified lower pipe

Dear Nelson:

The 171gmp (Us gallon=3.78 liters) in 6" pipe results in a flow speed of 0,6m/s. (Pressure drop/head loss around 0,06psi for 20m pipe).

For 3", 171gpm, speed 2.4m/s (estimated pipe lenght 3m) loss would be bigger, around 0,3 psi.

Thinking that your target is to divide evenly the flow (85gpm each end) with the new design, the second part of the pipe, now 6", would have a speed of 0,3m/s, head loss 0,02psi (estimated lenght 3m)

I would say that, with such flow speeds, pipe diameter is not the issue here and regarding your info that after flush there are no significative changes, escaling should be dismissed. It seems a basically low flow.

This considering that there is no information of standard operation values posted.

My personal approach, with the information available, would be a valve as marked in the sketch.

If a valve is not an option being offshore, the orifice plate mentioned by ajwinemaker in the post #1 would do.

Salu2, and let us see how it goes!!.

Hello ,

I think you have calculated on the wrong pipe.This one is my idea of modification.The existing pipe is the top one having a reduced end at the nde ares.The parameters given was for that pipe.So can you please refer that drawing nce again and reply fast,as we are expecting a crew change day after tomorrow and the issue will get stuck.So the flow in the de is 171us gallons and nde is only 32 us gallons.

Can you calculate please calculate the flow on pipe no2(the modified pipe ) and tell me what the flow will be if we switch to the new pipe modification.

Your suggestion of the oriface is my top priority ,yet i would like an answer for the modified pipe for my satisfaction.ThankI

Contact Rig Machinery Consultants, dont miss your crew change...

28/28 or what?

Yup...yes not rig oil and gas platform....

Thanks all

You have made a significant contribution and i feel these calculations have helped to

come closer to the issue.We will be inserting an oriface and record the exact flow both sides and then proceed.

Good plan, given your current situation. I would appreciate an update, if time permits.

A few parameters will help determine where your problem actually is, this will help;

1- Water temp in

2- Water temp out

The way you have described it I believe this is an air over type cooler (radiator) correct? If so what is your air flow?

You do not need high pressure, you only need good flow of both water and air. Most importantly the system needs to be actually removing the heat.

A- Are your cores fouled by scale build up?

B- Are you moving the proper amount of air across the cooler?

C- Are the Cooler fins clean, allowing good air flow and proper radiation?

all the parameters have been checked as suggested.Since the flow is mimimal on the non drive end the compartment temperature is high near alarm.Secondly the water outlet temperature is also higher than the drive end outlet

On the non drive end the water temp is around 30degrees in and 70 degrees out.

The drive end is okay.30-45

At 170 gpm with a 20F rise in water temperature, you should be able to remove around 1.7 million BTU/ hour.

139 gpm with a 27F rise is about 1.9 million BTU per hour.

32 gpm with a 70F rise in water temperature is about 1.1 million BTU per hour.

How much heat removal does the equipment require? Then the water requirements can be calculated.

As others have suggested, with sea water foulling, scaling, and debris build up are prime suspects.

Dear Augustinec,

With such a differences of flow, I agree with other posts about the insuficience of water supply to the system. Also, about scale build up, it is greatly enhanced by slow flows.

You have 139+32= 171gpm of feeding flow (we suppose both coolers are the same design/size?). Prior to redistribute the flow, decreasing in the 139 to increase in the 32, please consider how much you can decrease in the 139.

How far from temperature limit are each one?

Please fill in:

Actual temperature in Drive end:

Trip temperature in Drive end:

Temperature air exhaust cooler Drive end:

Actual temperature in Non drive end:

Trip temperature in drive end:

Temperature exhaust Non drive end:

Salu2

Has this cooling deficiency always existed or is it something that has recently manifested?

I am the second person to ask this. (#4 previously asked)

It matters because we need to know if it is a design or a maintenance issue.

Hello

We have six nos john brown make gas generators in our field and all the machines are having tis problem from the start.When we back flush the coolers it gives some more cooling nut not much.I think it is a design problem.

What is the GPM flow recommendation per cooler? If you have had this problem since installation,ie;( it has never worked properly), and you have recommended water flow available(85 GPM per cooler), then indeed it is a distribution problem....I would restrict the first line out as has been suggested here....

The present flow is 139 us gallons/min on the drive end

and 32 us gallons/min on the non drive end

I think the header would be constituting 139 us gallons/min

and our expectation is 100 on drive end and atleast 70 on non drive end.

"I think the header would be constituting 139 us gallons/min" What?

139+32=171 not 139

Introduce some obstruction (through a valve) in the drive end to divert flow to the non drive end. You have just enough flow to share per your 100/70 expectation.

If there is a lot of fluctuation that requires too much manual attention then consider automating the distribution. That could be the start of a whole new topic.

That this has been like this since the beginning of service means that extraordinary maintenance activities are probably not called for but if they are due anyway and the risk of service interruption is acceptable and the risk of a repair scope escalation or creep is unlikely then do that first. We're not sure if this deficiency is really a legacy issue or just anecdotally so. (log books?)

Why wasn't this dealt with when first commissioned?

Ive been down these same old roads may time on the rigs, rather than getting back to the basics our new generation of maintenance personnel would rather re-invent the wheel.

Clean the system as has been suggested many times by now, maintain the system... Insure the system is operating AS DESIGNED first, then move on from there.

Pro-active beats re-active maintenance any day.

Tim

barnacles are growing in the pipes?

or Zebra Mussels. or scale. or other debris.

Something (or several things) appear to be affecting both flow and heat transfer rate. There are several posts with opinions of what to try to fix or look for.