Optimizing Chemical Injection for Viscous Polymer in Winter Conditions

[spectastic]

see attached. the main 10" header is water, and the 1/2" chemical injection is polymer feed (not sure how much by volume, we go by pump strokes). problem is during winter, the polymer becomes viscous, and the pump has a hard time pumping the stuff all the way to the header. So we're thinking about putting in a small header that mixes the polymer feed. this 1 1/2" line would run off the supply header pressure from the 10" header. However, because the two 1 1/2" nozzles are so close to each other (like 1 ft), I'm worried that I won't get any flow through the 1 1/2" pipe. I don't want to have to put in a pump if I don't have to.

 

[spectastic]

nevermind. I'll need a pump for pipe "B", otherwise it wouldn't get any flow... too much friction loss.

 

[Travis_King]

You are right to be worried. After the losses in the 1.5" pipe, the mixed fluid won't have sufficient pressure to break into the flow of the header again (speaking colloquially). You'd need a pump, or alternatively a control valve or PRV that controls the flow in the header so that the pressure regulates with the pressure in the small header. This will alter your header conditions downstream and may negatively effect your process.

Instead, you could consider installing heat tracing and insulation before and after the polymer pump to help maintain the polymer's fluidity.

 

[spectastic]

You are right to be worried. After the losses in the 1.5" pipe, the mixed fluid won't have sufficient pressure to break into the flow of the header again (speaking colloquially). You'd need a pump, or alternatively a control valve or PRV that controls the flow in the header so that the pressure regulates with the pressure in the small header. This will alter your header conditions downstream and may negatively effect your process.

Instead, you could consider installing heat tracing and insulation before and after the polymer pump to help maintain the polymer's fluidity.

yea that was another idea. I have no experience with heat tracing tape, but now that I think of it, it might be cheaper to both maintain and implement.

 

[Doug Huffman]

Put a nozzle in the 10" pipe or merely an orifice to create the flow inducing differential pressure. Move the injection pump.

My facility wrapped heat tape under insulation for critical systems that weren't environmentally enclosed.

 

[spectastic]

Put a nozzle in the 10" pipe or merely an orifice to create the flow inducing differential pressure. Move the injection pump.

My facility wrapped heat tape under insulation for critical systems that weren't environmentally enclosed.

I see what you're saying.. is that something that you've seen work? I'm skeptical because even though it might change the fluid pressure, it will also increase head loss. without the pump, the flow through the 10" pipe is already picking the path of least resistance, which is the 1 ft long 10" orifice, vs the 1.5" 100 ft long pipe... so we're already barely getting any flow through there. putting a flow restriction on that pipe would make it worse

 

[Travis_King]

He was suggesting that you put an orifice in the 10" pipe, to increase it's pressure loss between the inlet and outlet of the mixing header.

An orifice does the same as a control valve would, it restricts the flow path to increase head loss; except that it isn't variable. Ideally, you'd match the loss over the orifice to the expected flow rate (based on orifice size and header pressure and current demand flow) with the expected flow rate and pressure drop along the 1.5" header. You'd have a bear of a time getting any specific flow rate, but with a couple valves that you can hand-modulate, you could probably get it to work assuming upstream and downstream pressures remain relatively constant.

The problem is that you generally don't want to make unnecessary drops in line pressure, especially on mains. Your downstream users will not be happy about it and if this is an operating facility, you'll have to make sure that all equipment downstream of the orifice will be operating properly (maybe they'd have to be adjusted) with a lower feed pressure and (likely) flow rate. How drastic this would be depends on how much flow you need in that pipe and what the expected head loss there is, if all you need is like 20 gpm, then you could install an orifice that, at your expected mains flow and pressure, would dissipate something like 2 psi.

Whether it's more economical to leave what you've got an install some heat tape and insulation (and have somewhat higher upfront costs plus some additional operating expense in the winter for the electricity) or have lower upfront costs, but have to deal with reduced line pressure and likely line flows (slightly), or do something else; well, that's something you'll have to look into.

**btw, by explaining everything as I am I don't mean to seem like I'm assuming you don't know anything. It's just that I tend to answer questions assuming someone else will look at it later and potentially learn something.

 

[Dutchwayne]

Your system is designed to function at a minimum temperature which is why you have problems during the winter. You need to determine the best temp range for service and install heat tracing and insulation. If the range is tight, you need to use an active controlled heat tape. If it is not important, you could use self regulating, but I would not recommend it. Do not use cheap heat tape or you will be replacing it in 2 years. You can also get hot spots which would affect the polymer as well. Usually you need to calculate the heat loss both atmospheric and to the polymer to determine the watts per foot. Make sure the tape is installed on the bottom of the pipe and it is attached directly to it. Also, do the insulation correctly all the way to the connections on the main. I have seen a number of heat tape installation that were useless because of bad insulation.

 

[spectastic]

we're going with chase water

with a pump..

 

[Travis_King]

Be sure the pump is spec'd to be able to operate acceptably over all viscosity and density changes due to temperature.