Pressure drop and velocity for gas application 6

[takiyasamsama]

Hi,

Okay this seems like same question asked so many times but I'm still new in heat transfer field and wanted to know between these two parameters. So when sizing for HE, the pressure drop shall be kept below the allowable pressure drop as specified for example allowable is 50 kPa then when design I get somewhere around 41 kPa so I know that my pressure drop is under the allowable limit however my velocity gets too low and I know that the velocity of gas should be somewhere around 10 m/s.

What I would know is there any guideline when sizing for gas application mainly on the pressure drop and velocity? If no then what are the criteria I should consider when sizing for gas application?

 

[zdas04]

I can understand the dP limit (any pressure you lose in the heat exchanger has to be replaced in a compressor somewhere), but the only reasons that I know of to limit gas velocity is: (1) to minimize dP on the high end; and (2) to minimize liquid drop-out on the low end. Target velocity inside gas/liquid separators is generally around 1 m/s, and liquids accumulation in piping seems to increase dramatically below about 3 m/s. Those numbers are pretty well documented in a bunch of places. Your 10 m/s number is probably a safety factor above 3 m/s so if you are between 3 m/s and 10 m/s you are likely OK, but you have a lower safety factor that people typically want. As to other criteria, mostly I see dP and velocity in specifications.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[EmmanuelTop]

Velocities in the range 15-30 m/sec are common for heat exchangers in gas service (tube side). On the shell-side, gas velocity should be as high as possible (constrained by allowable pressure drop) in order to ensure good distribution and particularly to minimize deposition of solid particles, if they are present in the gas.

Gases have notoriously low heat transfer coefficients so maximizing velocities within given pressure drop should be utilized to boost the performance. Dave Gulley has published some empirical equations that correlate between gas heat transfer coefficient, exchanger pressure drop, and tube length - the results are claimed to be within 25% of the real value. You can play a bit and see how increasing velocity and pressure drop affects the gas (and the overall) heat transfer coefficient, and what kind of performance difference is between 41 and 50 kPa pressure drop.

Ho = 430.Cp(ΔP/L x ρ)1/3

where
Cp = specific heat (Btu/lb-F)
L = tube length (ft)
ΔP = shell side pressure drop (Psi)


Dejan IVANOVIC
Process Engineer, MSChE

 

[takiyasamsama]

Thanks zdas04 and EmmanuelTop! I appreciate your kindness in such detailed explanation.

 

[georgeverghese]

As suggested, you may need to change the configuration and / or type of HX to arrive at an optimum design solution. This may however, not be the optimum cost solution.

 

[takiyasamsama]

georgeverghese,

actually I am sizing for plate and shell heat exchanger (PSHE) and this kind usually gives low calculated pressure drop compared to shell an tube however the limiting factor here is the velocity.

PSHE have two different velocity where one is on the connection side before reaching the plate packs and the other one is between the plate packs so that is what I am trying to figure out for gas application case what usually the optimum velocity.

 

[georgeverghese]

The first post said you have almost used up the available pressure drop ( 41kpa vs 50 allowable) and your concern was low gas velocity.

Your recent post says you have low pressure drop utilisation..

Anyway, the other key factor is to ensure that the design case duty Reynold's number is above the transition flow regime as fas as heat transfer is concerned. This transition occurs at approx Nre 4000 for shell and tube type, but could be lower for plate type HX with corrugations of some sort or other.

 

[davefitz]

The design needs to meet the commercial, economic and functional requirements of the project.

Commercially, if you are bidding from a spec that was distributed to other bidders, then the specified DP cannot be exceeded , and the technology used to beat the competition needs to be better ( in terms of surface area per ft of HX length) or the manufacturing techniques changed to provide a lower cost of construction. Extended finned surfaces or surface treatments can be used to improve the effective surface area per ft of HX length .

manufacturing techniques that reduce the fabrication cost could include use of robotic welders or outsourcing, or just a smart review of the fabrication sequence.

Economically, some process engineer had to decide on the value of pressure drop vs operating and capital cost. That economic analysis can always be re-calculated and reviewed based on changed economic circumstances, such as the changing cost of power, fuel, materials, capital ( interest rates), labor. For example, the rule of thumb of 10 fps as the economic pipe size for typical liquid piping, valves, pumps seems to be longstanding but can be recalculated for each job. Erosion/corrosion is often a reason for velocity limits in liquid pipe sizing.

Maximum gas velocities are often proportioned to the soundspeed or mach number , or are presented as a percentage of the inlet pressure. A spec may ask that the HX DP be less than 2% of the gas inlet pressure, or that the max gas speed not exceed 15% of soundpeed over the range of process conditions. Above some mach number ( 0.2?) the noise and vibration become excessive and damaging.

Overall, the bidder that consistently wins is using a technical approach that has been optimized, so studying those past winning bids will give some direction.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick