Shell/Tube HX & Condeser

[TehMightyEngineer]

The principle at work gave me this project and I fear I'm out of my league but making progress. We have an existing HX/Condenser fed from a steam ejector pushing steam and non-condescensable gases (NCG). This exits the HX/Condeser as condensate, some latent steam, and the NCG. That is fed through a mist eliminator, flame arrester, and finally into a lime kiln and burned.

Now, the client is looking to change this from burning the NCG in the lime kiln to burning the NCG in an oil fired boiler. They're increasing the motive steam and replacing the steam ejector with a larger one doubling the steam flow to the HX/Condenser but keeping the NCG flow the same. Now, my question is simply will the HX/Condenser pull more condensate out or will that be constant? If it's constant then this wont work as the extra un-condensed steam would end up being fed into the boiler significantly raising the oil usage.

I unfortunately don't have any of the manufacturers data on the HX. We're working on getting that but, for now, I don't have it. The cooling water flow to the HX is constant and I have inlet and outlet temperatures for the existing conditions. Assuming a constant heat transfer rate (wrong I know, but I have no other data) and I assumed that the change in temperature of my cooling liquid will be roughly constant (given that my cooling flow is over 12x the NCG/steam flow rate). Thus, I calculated my that my new NCG/steam will have the following inlet and outlet data:

NCG/Steam Inlet:
12,110 lb/hr of steam (H20)
6,257 lb/hr of NCG
269.7 degF temp.
1108 mmHg Abs. press.

NCG/Steam Outlet:
196 degF temp.

I calculated the 196 outlet temperature using the specific heat of the NCG/steam combined gas (0.279 BTU/lb*F). I suspect I did that wrong because I didn't take into account the phase change from vapor to liquid. Is that how I calculate the amount of condensed liquid? A point in the right direction would be great.

Maine EIT, Civil/Structural. Going to take the 1st part of the 16-hour SE test in October, wish me luck!

 

[rmw]

You really haven't given enough information to work with. Unless this Hx is grossly oversized for the current application, it probably won't condense twice the steam flow - with the same water flow that is - unless, again, you have some really cold water at the inlet. You probably will condense some more and even quite a bit more, but twice?????

If you know your water flow, you can measure the temperature rise and then you can calculate the Q and from that calculate the U from the area and the DT. I assume you know the area.

With the U value, you can find out what changing the water flow will give you.

I am curious why you think blowing this steam/NCG stream into the boiler furnace will cause the boiler to use more oil. I'm not saying it will or won't, but I am curious as to why you make that statement. And... where in the boiler furnace will this NCG stream be fed? Describe the boiler.

rmw

 

[TD2K]

What are your current temperatures in and out of the condenser for the steam/NCG as well as pressures? What are your current CW temperatures in and out?

The majority of your duty is due to the steam condensing since it's latent heat is close to 1000 BTU/lb. You're essentially doubling the duty from what you've said. You should have enough information to estimate the duty today. With the temperatures you have the temperature difference. That gives you at least the UA today.

Do you know anything about the exchanger? I'm surprised the maintenance people don't even have a GA drawing with a tube count/size.

 

[TehMightyEngineer]

@rmw:

Thanks. I don't know where the NCG and vapor is to be injected into the boiler. I will get this information from the principle.

I assumed that, unless more water is condensed out, we're going to be injecting over 5,000 lb/hr of water vapor into the boiler. I assumed that this would be literally like putting water on a fire and would require the boiler to run more fuel to maintain it's current temperature. Obviously we're pouring hot water on a fire but I still imagine there will be significant losses.

Here's all the pertinent information I have right now. The equipment file is missing due to a bankruptcy and restructuring during the first phase of the project.

Original cooler configuration:

Tube (hot) inlet:
H20: 7191 lb/hr @ 226.5 deg. F and 797 mmHg Abs.
NCG: 6460 lb/hr @ same temp and press.

Tube (hot) outlet:
H20: 509 lb/hr @ 120 deg. F
NCG: 6460 lb/hr @ same temp.

Condensate outlet:
H20: 6682 lb/hr

Shell (cold) inlet:
H20: 465 GPM (232,379 lb/hr) @ 85 deg. F

Shell (cold) outlet:
H20: Same flow @ 120 deg. F

Cooler data known:
Cooling surface area: 1650 sq. ft.
28" dia. shell
300" overall length
Single pass, both sides, counter flow.

From there I calculated my LMTD = 64.25 deg. F
I calculated a Q = 8,697,413 BTU/hr based on the specific heats (adjusted for proportions of mass)
I then back calculated a U = 82.04 BTU/(hr*ft^2*F)

@TD2K:

See above for existing.

I calculated my existing Q value based on specific heats alone. I suspect this is wrong and I have to take into account heat from the phase change of vapor to liquid in the water, correct?

I do not have a tube count but the sizes are listed above.

Maine EIT, Civil/Structural. Going to take the 1st part of the 16-hour SE test in October, wish me luck!

 

[TehMightyEngineer]

Oh, and the cooler has a maximum pressure loss of 5.6 mmHg on the tube side and less than 10 psi on the shell side.

Maine EIT, Civil/Structural. Going to take the 1st part of the 16-hour SE test in October, wish me luck!

 

[TD2K]

Your numbers are pretty close to mine.

8.14e6 BTU/hr based on heating 465 gpm of CW from 85F to 120F.

dT is the same as yours and I get an overall U of 76.7 BTU/hrft2F.

I'll have to do some more work to estimate how it's going to work with the additional steam. Some of the additional steam will condense (the CW outlet temperature will increase) and it will be a trial and error solution (well, it wouldn't be if I had Hysys on this computer but I don't).

 

[TehMightyEngineer]

Ah, so more water does condense out, that makes sense.

What are you using to calculate the three variables? (Cold water temp. out, hot temp. out, condensed water out) If it's a program or book I can buy I imagine my work can purchase it for me.

Thanks for all the help.

Maine EIT, Civil/Structural. Going to take the 1st part of the 16-hour SE test in October, wish me luck!

 

[TehMightyEngineer]

I was just told that we're going to be able to get the equipment file tomorrow which will obviously simplify this problem greatly.

@rmw: The NCG and remaining steam from the cooler will be injected directly into boiler to burn the NCG gasses. As the steam will be less than it's saturation temperature it should absorb the heat required to bring it up to saturation, right?

Maine EIT, Civil/Structural. Going to take the 1st part of the 16-hour SE test in October, wish me luck!

 

[rmw]

Well, actually, it will not only bring it up to saturation, but will superheat it to the furnace gas exit temperature. But then, that vapor combined with the rest of the products of combustion will enter the tube banks and start transferring heat to the water in the tubes and generate steam. Some of the products of combustion is water both from fuel generated reactions, water in the fuel, and water in the combustion air. This little bit of steam will get lost in that big soup of constituents. It won't really take any more fuel heat to heat the steam than it will the combustion air.

Now you purists don't start posting about the difference in specific heat of water vapor and air. This is said to "pirate" in general terms to allay his/her fears that the steam injected is going to penalize the boiler performance.

rmw