Feed pipe design for gentle flow down the wall of a tank

[JKoenders]

Related to thread378-239797, I'm interested in design criteria for designing a feed pipe that directs a top-fed fluid onto the inside wall of a tank. If the velocity in the feed pipe is at most 1 m/s to limit static charge generation, how do I ensure that the fluid then flows gently down the wall of the tank without splashing or spraying? I'm going to help things along by angling the feed pipe downward at a 45-degree angle to the tank wall and cut the end of the pipe parallel to the tank wall, but how much gap should I leave? I want to keep it close to minimize the free span that the liquid has to travel, but don't want to restrict it so much that it sprays out - I feel like there's a Froude number and/or momentum calculation at play here.

I understand that part of this solution could involve reducing the velocity out of the feed pipe even further, but as the flow out of this feed pipe decreases, when is the end of the pipe no longer uniformly liquid-full? At some point gravity will take over, and I'll be unable to control the velocity at which it 'pours' out the end of the pipe (i.e. I need to check that I'm designing to a realistic flow regime at the end of the pipe).

A dip tube is not desirable because of the mixing that needs to occur in the tank (high-precision recipe that can't have ingredients remaining within the dip tube that don't get mixed in). Also, we're dealing with combustible fluids (not flammable), which, as I understand it, are only a concern with regard to static charge if they become free droplets - hence the desire for a gentle feed down the wall of the tank.

On a related note, regarding the limit of 1 m/s fluid velocity entering a vessel through an open (not-submerged) feed pipe - is it acceptable to have a higher velocity in a smaller line leading up to the tank, and then increase the pipe/nozzle diameter right as it enters the tank to slow the fluid down? From what I've seen, the velocity limit is as it enters the tank, so, it's ostensibly acceptable, but it's also my understanding that a non-conductive fluid generates charge as it flows through a piping system; so, if it's approaching the tank at high velocity, has it not already accumulated the charge? Does the low-velocity segment of piping need to be long enough to provide for relaxation time?

Similarly, can you use a control valve to throttle the flow down to below the 1 m/s limit at the exit, or does the increased velocity through the valve serve to generate charge counter to your efforts to control it?

 

[georgeverghese]

Presume the concern here is for this liquid with electrical conductivity < 50pS/m ?

Re other concerns, see if this helps (section 2.5).

https://www.matec-conferences.org/articles/matecconf/pdf/2021/12/matecconf_mse21_10013.pdf

 

[pierreick]

Hi,
Same vein.
Pierre

 

[JKoenders]

The first part of my question is independent of the static charge generation concern, or of the conductivity or flammability of the material. The question is entirely related to the 'smooth' dispensing of any liquid onto a tank wall without splashing or spraying.

Regarding the "related note" part of my post, the article that georgeverghese linked to does explain that the velocity-limit associated with charge generation in non-conductive flammable materials does apply to at least 30s worth of piping upstream of the vessel. That makes sense, but in the multitude of other articles and texts I've read on this subject, that point was never clearly made - they were all just ambiguous enough that one could reasonably interpret the limit to only apply as the fluid was entering the vessel at the feed pipe opening.

Regarding my last "Similarly" question, the CEP article that pierreick provided does state that control valves can be used to reduce flow for the purpose of limiting charge generation. This may seem like an unwarranted concern, but it stems from the consideration that filters are flow restrictions that cause charge generation, so my concern was whether other flow restrictions, like orifices or control valves, would similarly cause more charge generation than what they're preventing by reducing velocity through the adjacent piping. Apparently not, and it's the large surface area in filters that's leading to charge, whereas valves, while having high local velocity, don't have a lot of surface area.

 

[LittleInch]

Agree with others here. If you're that worried about static then you need sufficient time for the fluid to "relax" and for the charge to find a way to the earthed wall of the pipe.

I've never seen one, but don't see why a simple earthed mesh say 1cm square holes wouldn't conduct more static out of the fluid.

But to get the liquid to flow evenly down the wall you really need a distributor pipe with multipl small holes spraying the fluid against the wall from a small distance, but just make sure the pipe and wall are bonded / earthed together.

So kind of similar to an firewater cooling spray ring but internal and a bit less of the spraying....

or weld on a square gutter and flow into that, but leave a small gap on the inside, but with a few bits left to hold the gutter on.

But they are a bit fancy - Whats wrong with a stilling tube / submerged fill line?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[JKoenders]

LittleInch,

I'm not "that worried" about static - I'm just trying to better understand and properly apply the criteria for controlling it. In the case of the feed pipe, the velocity will be controlled to a level below the limit for a non-submerged fill pipe, so I shouldn't have much if any charge that needs to be relaxed. But, as the fluid flows out of the feed pipe and onto the wall, it still needs to do so without splashing or spraying (free droplets could generate charge). "Small holes spraying the fluid" is the opposite of what I'm trying to accomplish. It's not about the flow being "even", but rather, "gentle". I want to "pour" the fluid as a single stream "gently" onto the wall - but I'm wondering how I quantify a "gentle pour" - I need enough velocity for the fluid to strike the wall and not drip off the pipe, but I need a low enough velocity and enough of a gap that it doesn't splash off the wall or "squirt" out the gap. This is all relative to the viscosities of a variety of ingredients. I imagine there's a momentum & viscosity relationship to be considered, along with the angle of attack and some threshold value to remain under. As the feed velocity is reduced, eventually gravity will dominate, and the fluid will drain out freely from a partially-filled pipe in an uncontrollable manner.

A submerged fill line will retain a stagnant column of the last ingredient added that won't get mixed into the batch - the recipe requires high precision. The 'gutter' distributor is a solid idea, but I'm thinking the right shape & size of feed pipe can accomplish the same thing without the gutter. The gutter might present a cleanability issue. We considered purging the dip pipe out with air/nitrogen pressure, but you'd have to hold that pressure to keep the pipe from flooding again, which would preclude the addition of a weep hole in the dip pipe, and we also don't want to "blow bubbles" through the batch either.

 

[JKoenders]

I see now that the article georgeverghese linked to makes a distinction between loading speed for "splash loading", and the flow limit that applies to charge generated by the flow of liquids through pipes. The 1 m/s velocity limit applies to the charge generated from "splashing", while the 30-seconds worth of pipe-volume applies to a vd limit of 0.38 or 0.5 m^2/s (max 7 m/s), depending on the material, for limiting charge generated within the pipes. So, as long as you maintain the vd limit leading up to the tank (which is a much higher limit than 1 m/s), you can then slow the fluid velocity to 1 m/s as it enters the tank to control the splashing concern. I just wonder how broadly a 1 m/s velocity limit applies to flow that's directed directly onto the wall of the tank.

 

[1503-44]

Why not direct the outlets to align with the a horizontal tangent to the tank wall curvature, assuming its a cylindrical pipe. It will flow along the wall, first horizontally, then curving downwards to the bottom. Should get minimal splashing.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

How about a cone and just connect the pipe over the top with a space between pipe and cone?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.