Multiple Small Boilers vs One Large Boiler 5

[RamWreck]

I work for a company that is designing a novel process at a large scale but we are currently in the early stages of pilot-scale development. When thinking about the scale-up I am not sure at what point the system needs to be split into two plants. The process is very steam intensive and thus I believe the boiler may be the limiting factor. The largest scale we are currently looking at requires ~100kpph steam at 250 psi but in theory, scales could be 10x this.

My question is:
When sizing a boiler is it always the case that a single large boiler is a better choice than two smaller boilers? When does it become more reasonable to break this process into multiple plants with their own boiler or multiple plants run from a single boiler? Obviously having two boilers gives you some redundancy if one "plant" goes down or you need to do maintenance or repair on the boiler. On the other hand, I imagine if I double the capacity of a single boiler, the upfront capital cost may only go up by 1.6-1.8 times instead of doubling like it would if I buy two smaller ones. Any thoughts?

 

[mk3223]

It seems that for the new project, you have listed some factors and options on the post.
Suggest to group your items on a list, provide the necessary data, pro & con, etc., for comparison. And then, review it with the Project and Mgmt for a agreement of the forward plan and decision.
Of course, you may also seek a proposal from an EPC Company to have a feasibility study to fit your plan and conditions.

 

[MJCronin]

First, I do not understand: The process is very steam intensive and thus I believe the boiler may be the limiting factor.

Could you explain this ?

Second: Many Process Plants using saturated steam have multiple boilers available because without steam, the process stops and costs skyrocket

Usually there are two boilers, a single DA and multiple Boiler feed pumps. There may be a condensate collection/forwarding system, condensate storage tank and, of course boiler water treatment system. Sizing and selection of all of this auxiliary equipment depends on how you want to run your steam system

Some boiler systems are designed to burn multiple fuels (usually NG and fuel oil) ... some not. Your emissions permit will control things here

Third: Your required Process Turndown is EXTREMELY IMPORTANT in selecting the sizing and number of boilers. Do you need to run your steam system at 10-15% of maximum capacity ?... This fact then pushes your decision to two (or possibly three) boilers.

Boiler system maintenance is another factor .... Sooner or later, things must be fixed.

Fourth, and finally (to complicate things) .... Some process plant managers install provisions for a temporary rental/mobile boiler to be placed and run during times of crisis. I have personally done studies and specified steam plant mods to permit rentals to be installed. It is important to consider rental boilers during the plant/equipment layout stage of engineering

https://www.wabashpower.com/rental-...gc7FQZsC9a7L-v_Qmyiwij_p09fQnPMwaAv3fEALw_wcB

https://www.nationwideboiler.com/mo...jlcaQell7yOg2y27d4WC5HkhWTvUaj5AaAu0HEALw_wcB

I suggest that you spend the modest amount of money for an experienced professional to perform a study.

MJCronin
Sr. Process Engineer

 

[LittleInch]

" a single large boiler is a better choice than two smaller boilers?"

In all these case my first point is Define "better". This is a summary of MJC similar post above.

Lower CAPEX?
Lower OPEX?
Higher Reliability?
Larger Turndown?
Smaller Space?
Less Maintenance?
Capacity?

The means by how you rank each of these competing factors will give you a different answer to the one which went before.

If you can't produce your substance without steam then reliability of supply would be very high on my list. Work out the cost of the plant not working for 1 day, 5 days, one month without steam and your $ cost equation can get rapidly impacted.

Similarly turndown. My normal ROT is that anything more than 4 or 5:1 becomes a real practical issue. I'm sure you'll find a salesman who tells you their boiler can do 10:1 or 20:1 turndown, but you won't find many operators saying the same thing.

Ditto how fast does this process speed up or slow down. A Big boiler will be slower to respond than multiple smaller ones.

BTW what is 100 kpph? Thousands of pounds per hour? So 45 tonnes/h? Decent size , but not that big.

SO another factor is vendor supply and capacity. As you get bigger the number of vendors gets smaller and you would get less pre packaged units. Each vendor will have different ranges of supply.

But this is why design engineers exist to actually design things for each specific use and deliver the most efficient design, once you work out what is more important to the specific company or process.

SO yes, one larger unit will normally cost less CAPEX than two smaller ones, but if the one stops working then you have no steam.

Similarly if you have say 5 units and one stops working, the others can probably cover the gap or at worst still supply 80% of the need.

Very few plants work with a single critical item so it's usually at least 2 x 50% or better.

I did know one company whose attitude was the 2 x 50% solution on the basis than in reality one unit would probably be able to supply ~60-70% of the requirement and then they would just throw money at the repairs or hold an uninstalled spare or major components and then fly people or bits in to fix it. They worked out over the course of years and multiple locations that was "better" for them than installing 2 x 100% or 3 x 50%. Not many others think this way.

The majority of systems I know go with a minimum of N+1 for spare equipment.

Sometimes it helps to think of this in terms of what would you do if you had to generate your own electricity? Same set of issues. One big generator, but if it stops the lights go out other than the emergency ones until someone plugs in a rental generator or fixes the old one. Same issue about size - one big unit is not good running a <30% load and can't respond very fast. Steam is the same.

Whatever you decide, someone will say it is wrong or a bad decision.... just make sure your boss thinks it's a good decision.






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

 

[RamWreck]

Thanks for the responses. This is a lot of good information and will be very helpful!

 

[OldDuffus]

I had a chance to visit some utilities "here", East of the Mississippi. Our company does a lot of replacement parts work, even offering to replace part or all of the section tube heat absorption surfaces, i.e. Economizer, Reheater, Primary and Secondary Superheat. We replaced entire furnace hoppers, many of these were on competitor-built units and they did the same to some of our fleet.

I got to see a few strategies, not as many as a service engineer since most of my career was back in the office behind a desk performing stress calculations.

At one site, the customer appeared to have (2) 100% capacity units and maybe a 50% capacity backup unit. It appeared they operated one of the 100% units while the other was in down time and perhaps out for an extended time undergoing repairs. They may have "harvested" some parts from the downed boiler to keep the active unit running and available, using the down time to order replacements for these parts for the downed boiler to ready the downed boiler when it was placed back in service. It seems logical that they used the 50% capacity boiler to allow the larger units to meet peak demands without running full tilt by filling in the peaks.

Another customer seemed very successful running, perhaps (2) larger, devoted boilers or maybe had one for backup like the first customer. They met peak demand by employing a combination of natural gas turbines and heat recovery steam boilers. The gas turbines I imagine were hooked directly to electric generators, supplying additional energy in that form. But the exhaust heat from these turbines started at over 1000 degrees F, plenty for attaching an HRSG (heat recovery steam generator) down stream in the flue work. Main steam lines from these small boilers fed back to the old steam turbine "house". I don't know if they introduced their steam at some intermediate pressure turbine or had a low pressure, devoted turbine. I believe the steam generators were plumbed to two gas turbines in dual flue feeds, two gas turbines feeding one HRSG. In total, they had (4) gas turbines and (2) steam generators (HRSG's).

We knew our large boilers and the competitors had a common weakness: The boilers, especially the pulverized coal burning type could not be run at loads less than, say, 30% capacity. This means the customer had to run the units at least at this load or higher and, I guess vent off or condense the excess steam at a significant loss of energy. The main cause of this limitation is the problem of cold end corrosion. If you run a large boiler at very low loads, the temperature in the convection pass drops low to prevent overabsorption in the convection pass tubes, you just don't need that much heat for a low steam demand. Down stream components of the exhaust flue suffer from the phenomenon of cold end corrosion. We saw entire scrubbers or precipitators reduced to "swiss cheese" in an alarmingly short time.

We and our competitors developed at least two solutions or methods to modify or retrofit the large boilers to keep their exhaust gases at a non-corrosive temperature, higher, of course during low load periods. Granted these "solutions" also waste energy by passing unabsorbed heat downstream of the convection pass tubes. In one method, we bypassed some of the tube sections or their portions by routing the flue gas so as to "miss" some of the heating surface. You can imagine this is not an easy retrofit, more of a rebuild of the unit's back end. The second solution is to modify, through valves, the circulation in some of the tube circuits so that not as much heat transfers to the water/steam even though the tube section remains fully wetted by the flue gas and the flue gas flow rate past the tubes is not reduced. In either solution, the flue gas leaving the boiler remains hot and non-threating to downstream components.

 

[FacEngrPE]

Consider your outage plans when designing your equipment arrangement. Boilers and deareators have periodic inspection and maintenance requirements. If the process can not be off line while the boiler is off line for maintenance you may need to consider N+1 redundancy for all boiler house components. N+1 might also be applied to the inter connecting piping, control systems, etc.

Your choices should consider the cost of being off line vs the extra cost of N+1 design.

 

[MedicineEng]

To me, I would say 2 small boilers are almost always better than a single boiler unless you have such slack in production lines that you can afford to be down an extended amount of time.

It is not only the unforecasted corrective maintenance, but also the big annual maintenance that typically a boiler goes through.
And the dirtier the fuel is, the more maintenance the equipment will require.