Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations IDS on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Recirculation line?

Status
Not open for further replies.

Engineering1012

Chemical
Apr 18, 2016
29
New maintenance engineer here, working at a chemical plant.

Currently we have three vertical pumps for a sump in my unit. A primary always runs to maintain sump level at 40% and secondary pump kicks on when sump rises above 60%. The third is a backup. So the main issue is that the primary pump runs below or at minimum recommended flow(334 gpm) for 95% of the time, it isn't until we have rain or other dumps that allow the pumps to open up closer to their BEP (1300 gpm). All three pumps feed into one control valve.

Now there is a small 2" recirculation line that only provides about 120 gpm according to tests and actually dumps into the pumping chamber right at the suction of a pump which has its own air entrapment issues. With this poor design I am looking into sizing a new recirculation line to dump back in the settling chamber and then getting a Vfd for the primary (due to cost).

With the CV 100% open, the discharge pressure off the primary pump was 22psig. There is no other points to get a pressure reading downstream of the valve when it is loaded during normal ops.

My question is , what should my plan of attack be to size this line? Any advice or tips for a young engineer would be great. Thanks in advance!
 
Replies continue below

Recommended for you

Assume the flowrate thru' operating pump is maintained by the control valve, correct?
What is the maximum inflow to the sump during rain periods?
Are the pumps oversized for the application?
Consider changing primary pump for a lower flow unit to operate near the normal condition (95% of the time)
Size the by-pass line to larger dia. possibly with a manual control valve to allow setting during normal operating periods.(move the discharge from the by-pass line to elsewhere or discharge it well below the sump surface level)
You need to analysis the costs involved of fitting a VFD and the increased running costs (VFD inefficiency) compared to running the pump operating near to its BEP with a by-pass line in operation, you might be surprised with the outcome between the 2 options.

A lower flow unit seems to make some sense, provided the other 2 units can maintain flow during peak demand.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
 
- Yes the flowrate is maintained by the control valve.
- Dont have the maximum inflow yet, I will be getting this tomorrow.
- The pumps aren't oversized when they need to run at full capacity, but the other 95% of the time, yes they are.
- A lower flow unit for the primary is an option, however this would also require you to purchase a spare for that smaller unit.
- Could you clarify "size the bypass line to larger dia"? I was thinking of using a 4" line and yes dumping into the settling chamber.
- VFD will cost roughly 20k (high estimate) to implement on one primary. I plan on running a plant test with a temporary 4" line to see our flow.
- I guess my other question is when sizing this recirculation line, what do I need to do in terms of losses? To ensure the fluid would still get where it needs to go.
 
4" should be ample, the head loss reduction compared to 2" is 2'/4'^5, if you want to be academic about it use physical dimensions, but assume the line is a short rum so probably not necessary.

as for a stand by, why bother - maintain the existing primary in case it's ever needed.
Look at a quality electric submersible as the low flow unit.



It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
 
Well I guess my question would then be if that smaller unit went down, what would you use? The two existing ones ? Because then your running back on the curve again. Or perhaps size a recirculation line for that situation?

Furthermore if a VFD can get the job done and it's a cheap option than a submersible, that may be the way to go no?
 
I would select a good electric submersible as the primary, keep the existing verticals in the case of emergency --- you use one as the duty pump now, so to use it again in an emergency is no big deal.
And why would VFD be cheaper than a submersible, bear in mind VFD is not a panacea to cure all ill's or a poor pump selection.
You need to undertake a proper cost analysis, including the running costs of a VFD running an overcapacity pump compared to a correctly selected pump for the 95% operation.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
 
Also look up ARVs, automatic recirculation valves. You can set those such that flow through the pump is your minimum flow which reduces energy loss but close when you need the flow properly.

If static lift is a big part of your pump output, vfds are often not very variable as head would drop off too fast.

Everything else artisi says I agree with.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
Have you verified that a VFD will work? Just because you need less flow does not necessarily mean that you need less head. For many sumps in our systems, the downstream pressure is somewhat fixed by elevation or pressure from other sources. If the suction pressure was constant (level in the sump) and discharge pressure was constant (slop header pressure, for example), then a VFD might not work well at all. The VFD would have to run at nearly full speed to maintain the head required to get into the downstream system. This depends on the system characteristics and the shape of the pump curve.

For this application I like the automatic recirculation valve. We use these on a number of pumps and have generally had good experience with them. This really gives you the best of both worlds. You get the high capacity during rain events and protect the pumps the rest of the time.


Johnny Pellin
 
If a quality submersible was chose as the primary, you're saying still run the secondary how it does currently run? So when max flow is needed we have it? And then the third is backup?

I'm just saying if the smaller unit it went down for whatever reason, you'd be back down to running a primary in poor condition for whatever duration. I guess this would still happen if there was a VFD on one of the them anyways. Would it then be wise to still go ahead and size a larger recirc line for the other pumps if the submersible is down?
 
Currently you run a major pump at 95% continuously , can't see why you are now concerned if there was a need to run it in an emergency situation. Yes I would certainly review the by-pass line, even if for no other reason than to move it from discharging near to any pump intake.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
 
What would be my plan of attack for finding out IF a VFD would even work? Get my losses for the system and see if at the lowest frequency, would you produce enough head to meet the head required?
 
Get your min head number, and then look at pump curves plotted for various frequencies and see if the head/flow that you need is on one of them sort of near BEP.
My hunch is that you have lots of excess flow, but very little excess head, so a VFD will probably not work.
I like the idea of a smaller pump for standard duty, and a larger recerc line for in between times.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube
 
Generate a system curve. Overlay it with the pump curve at full speed. Use the affinity laws to estimate how low the speed would need to drop in order to drop the flow rate to your minimum requirement. I have attached an example. If I wanted that pump to be able to pump 3000 gpm at full speed, but need to run the flow as low as 500 gpm, a VFD is not going to work well. The angle of intersection between the pump curve and system curve at 3000 rpm and 500 gpm is not going to control well. With a much steeper system curve that has a much lower head at zero flow, it might control just fine.

Johnny Pellin
 
 http://files.engineering.com/getfile.aspx?folder=ecdf3021-861a-4fdf-a090-4893bbd80cf8&file=System_Curve_Example.docx
Recirc lines are a lot cheaper. Sounds like you don't have a lot of head to play with or even much system data. Recirculation back to a better place than you have at the moment doesn't change the basic design.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
When implementing the 4" recirculation line would the valve on that line just always stay open and then that CV at the discharge would still be used to throttle to get the desired flowrate to maintain the level?

 
You can do it that way ( cheap and cheerful), but then that re-circ flow will be there even when you don't need it, so your pump will never pump 100% of flow forward. You really need some sort of manually variable CV control valve to adjust the flow to be the minimum you need rather than rely on a valve CV from an isolation type valve.

Like I said a few posts ago, search for ARV - automatic re-circulation valves. These automatically monitor the flow and without instrumentation and power, only divert flow via the re-circ line when there isn't enough forward flow.

An example is here
Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
As an exercise you need to measure the forward flow and the discharge head to achieve this flow (not including what's going thru' the by-pass) and calculate the water horse power to achieve this flow / head, and then measure the power being used by the drive motor and the running costs per day/ week / month. This should highlight the wastage of throttling oversized pumps.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor