Vacuum - pressure drop calc 1

[gjmorin]

Hi all:

I'm working on a refinery unit that has a vessel at Deck 12 that'll need to have its dry adsorbent pellets removed once or twice per year as it is spent. This is a new feature on this type of unit, so we don't have specific operating experience.

As we can't be assured the material will flow well and a manual unloading operation would be a pain in the --- (plus throw adsorbent down 11 decks and everywhere else if it happens to be windy), we were thinking that providing a dedicated vacuum line up the structure with a couple connections at top for a hose and a connection for s vacuum truck at the bottom would be advisable.

I'm a little concerned with pressure drop in the line and hose however. We'll have the hose from the truck, 200 ft (60m) of pipe, and then a hose on the vessel side. We'll need to lift the material from the height of the bottom manway, through the top manway, and up to the connection on the vacuum header - call it 20 ft (6m).

Presumably once I lift the material the 20ft, I'll have gravity in my favor and the material will move down to the bottom, but I'm unsure of the calculation method I'd use to confirm I can:
a) have adequate suction at the end of the hose to lift the material
b) be able to convey the material in a hose up 20'

Can anyone point me toward a reference or methodology that we could use to confirm that this'll work? Conversely, can anyone see a consideration / flaw I haven't thought of that would make this unworkable?

Thanks

Greg

 

[zdas04]

Normally I would use the Spitzglass formula for vacuum flow, but it doesn't do so well with two phase flow, so I'm not sure how much it would tell you. In this kind of operation, I tend to want to put a drain nozzle into the bottom of the vessel with a hose connection. Then put 5-10 psi of air on the vessel and blow it into a vac truck.

You say the adsorbent pellets are dry (you didn't say a pellet size) so your standpipe idea can work, but 20 ft is higher than you are going to be able to lift much weight at all (any moisture or pellets much bigger than BB's). Coming out the manway can be a challenge.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

 

[gjmorin]

Thanks David.

The adsorbent materal should be dry, light and small, but it seems I'll be imposing a limitation that might affect the flow rate of the material.

Gravity removal is not a good option- even with pressure assist. There is inert bed material at the bottom that would not need to be removed every time, plus there are trays within the vessel. If you're correct, I might need to provide an extra manway & platform such that my max lift is reduced to 10 ft.

Is your skepticism of lifting 20' based upon experience? If so, I'd like to talk though how to calcuate the limit. I can't see where 20' is a physical limit, but there ight be a practical limit regarding how much material can be removed.
- For argument's sake, let's assume the pellets don't adversely affect frictional pressure drop
- I restrict airflow via the hose size at the vessel so that the truck pulls 5 psiv at a given flow condition (whatever that is) & we have ~1 psi pressure drop in the truck hose and pipe. Under these conditions, the high point in my system is at say -4 psiv (don't know if that's realistic or not). Less pressure drop in the 20ft + hose I'll have say 3.5 psiv at the inlet nozzle.
- If the velocity generated by the hose exceeds the terminal velocity of the pellet, then I'll lift it out of the vessel.
Here's where it gets tricky:
- The physical limit to lifting one particle should not be 20'. If the velocity at the hose inlet is adequate, all that is required should be to maintain terminal velocity.
- As we entrain more particles, though, a limit is reached since to remove a realistic amount of material, the fluid density now includes the mass of the adsorbent & I need to make sure I have enough DP to pull the material up to the high point. The calculation becomes a head calc based upon the 2 phase density.

Do you think I'm on the right track?
- do a fluid pressure drop calc based upon air only to figure pressure and flow & calculate the CS dimater of my extraction hose
- confirm that I have terminal velocity for my particles based upon that flow rate - adjust hose diameter accordingly.
- calcuate the removal rate of particles based upon head loss up 20'
- Finally, divide results by 2 to account for friction.

Greg

 

[zdas04]

I always take a worst case as liquid full. In order to be able to siphon water out of a tank, you need to lower the pressure, and the max negative you can be is local atmospheric. I live at a mile high so I take atmospheric to be 12 psia so the highest standpipe I can vacuum is 27 ft which must be reduced by friction losses. I start getting nervous around 20 ft.

You are not dealing with water, so your calculation would be different. I would probably look at the density of the dry material, convert it to a hydrostatic head as though it were a Newtonian fluid to try to get my head around the physical limit. If the density of the adsorbent is less than water, then your 20 ft is probably ok.

You seem to be in a reasonable track. I would probably use a drag model instead of a terminal velocity model.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

 

[gjmorin]

OK. I was thinking water when you mentioned that there is a limit around 20'. That shouldn't be an issue here - no phase limits.

I appreciate the thoughts....

Greg

 

[katmar]

What you are trying to do is pneumatic conveying and there are plenty of texts and information available. However, if you have not done it before it may be better to call in a specialist. None of the usual pressure drop (eg Darcy-Weisbach etc) or terminal velocity calculations will be accurate enough. By using vacuum you will be lowering the carrying capacity of the hose. Sometimes an ejector is used to draw the solids up into the suction of the ejector, and from then on the conveying is done by the pressurized gas (usually air).

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[StoneCold]

We have used a ejector type of equipment from exair. They work but they are slower than you might think it would be. You need to put the ejector close to the suction point of the hose to get the best effect.
To give you an idea of the situation we used a 1-1/2" unit connected to a 1/2" 100 psig air line to vacuum out a 1ft diameter column 9ft deep in about 45 minutes. There was very little work involved just feeding the suction end slowly down into the 1/16th inch molecular sieves. But the feeding has to be done carefully or you plug up the suction side of the ejector. If your unit is very big I would seriously consider a side dump nozzle above the inert bedding, or dumping the whole thing and sieving to recover the inert bedding.

Regards
StoneCold

 

[LittleInch]

The key consideration I would have is that your vac truck is nowhere near absolute vacuum and only being able to lower pressure at one end will always limit your flow velocity and "lift". For that sort of distance you don't have a plan B if the vacuum isn't enough to actually cope with the irregular mass flow rate you get when you suck some of this up. Also some vac trucks will be better than others. If you gat a bad one you're in trouble.

Katmar has it spot on - blow using a piggybacked air line and educator and you can vary the air flow rate / pressure. The agricultural guys have much more experience for delivering feed and grain - look up pneumatic grain feeders or similar and then apply to your issue.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way