Modifying a boiler system - switch to primary/secondary?

[BronYrAur]

I have run into a situation that I haven't seen before. I have been asked to make recommendations on a boiler system. The system has 10 Raypack boilers at 2,000 MBH each. The piping is weird. It looks like it may have been intended to be some form of quasi-primary/secondary with a 3-way valve in what would have been the bridge. The valve is not operating and is always in the position where the bridge (bypass) leg is closed. So this puts the main pumps in series with the individual boiler pumps.

They have been running this system for years. Flow from the main pumps is pushed through all boiler all the time. The non-operating boiler pumps are "free-wheeling" as the system pumps push through them. The water in those non-operating boiler is of course not heated and blends in with the heat water. There is a temperature sensor on the line leaving the boiler room that is used to attempt to maintain a setpoint.

These are non-condensing boilers that run year round. They provide not only building heat but also domestic water heating. I think someone had the bright idea to drop the setpoint as low as possible in the summer, which brought back water that was too cold. Their heat exchangers have been rotting away.

The suspected load is only about 10,000 MBH, so they only need about half of the boilers. Some are beyond repair, and I think right now they only have 6 boilers in service. The others are valved off. So they are pumping through 6 boiler all the time and running boilers as needed to make setpoint. The main system pumps have VFDs but they are in a fixed position. They have suggested removing the boiler pumps and replacing them with 2-position valves. That way, there would be no pumping through non-operating boilers. I have concerns about this because the system flow would then be limited to the the number of operating boilers. Right now, nothing looks at the loads on the other end. I can foresee the loads not getting enough flow, which will not only be unsatisfactory, it may also lead to too low return water temp. The load consists of 4 heat exchangers, 3 of which have 3-way valves. So this will lead to highly variable flow (depending on the number of boilers), which doesn't seem like it will work. Also, the min/max range on the boilers is not very large. I am struggling with whether or not this "primary-only" type of system would be a good idea.

My first thought was to remove the 3-way valve and create a true primary/secondary bridge. That way, the boiler pumps and the system pumps would have hydraulic separation. I don't think the boiler pumps have enough head, but that can be fixed. The temperature sensor would also need to be moved so that it is on the other side of the bridge.

A sketch is attached. Any thoughts?

 

[nglty]

Bron

Quite a situation you have there. Got to break the eggs to make the omelette.
My few thoughts if I may:
1. I'd Keep dedicated pump for each boiler, much simpler to control.
2. If you provide Control valves at each boiler with common pumps, you will have problems during staging, the sooner you bring a new boiler on, the boiler
already in operation will see a quick drop in flow and may trip on low water flow.
3. Consider using PICV valves with Primary/secondary. PICV's have proven to eliminate the low Delta Syndrome.
4. the setpoint is an ASHRAE 90.1 requirement. but it should be limited to the boiler operation, i.e 140 Min for non-condensing.
5. Agree with temperature sensor moving to other side of bridge but you should have a flow meter going with this to stage boilers based on building load.
6. I would provide a dedicated pump for DHW and shut-down the remainder of the plant in summer.

See attached sketch and let me know what you think

best

 

[willard3]

If these boilers are at or about 10 yrs old, it is probably time to replace them.

 

[EnergyProfessional]

that 3-way valve seems to be some valve to provide minimum flow to secondary pumps. Raypak boilers are horrible. No support, low quality. We removed many of those from our buildings. If you have 10,000 kbtu load, you could get by with 2-3 boilers and do a variable primary flow. that sure will simplify things.
Raypak couldn't even give me the manual of a 7 year old boiler. nada. For any normal manufacturer I can download old manuals from the website. Raypak has nothing and support doesn't either.

I'd just remove all piping and start with a clean slate. And get condensing boilers. Even if you don't operate at condensing temps (you may at shoulder season), you don't have to deal with the prevention of condensation. The non-condensing boilers need to have a thermostatically controlled return flow bypass to ensure a minmimum return temps. Yours don't seem to have that. That alone makes condensing boilers a no-brainer.

 

[nglty]

EP

From what I can tell, condensing boilers may not work in the application, they use heat exchangers at the end of the line, which are sized for a hot side temperature coming from a non-condensing boiler. Meaning you would need to replace the HX's with much larger HX's, as condensing boilers will not give out the same capacity due to their low-grade heat. Even worse when they go on Outdoor temperature reset.
Retrofit is often painful, Can't just replace a piece without affecting the entire system. And more often than not, retrofit means retrofit, i.e comes with a retrofit budget, not a capital improvement budget.

But we're in luck, Bron will shed some light !!!!

Cheers.

 

[EnergyProfessional]

The only limit modern condensing boilers have is they usually can't be set to above 180F, but there may be models that can. Everything else is just a matter of control.
They also have stainless HX, so they last much longer. High turn-down also will eliminate short-cycling, which will make them last longer. Look at the boilers they have now, half of them don't work anymore. I don't call that an overall saving. I bet 2 condensing boilers with variable primary are cheaper then 10 boilers with 10 primary pumps and proper return temperature control (which theirs don't have, but should). Not even to mention the maintenance cost, boiler inspection cost etc.

A bad design is not necessarily cheaper, it is just bad. The owner paid for unneeded 10 boilers twice the needed capacity, needed a large space to house the 10 boilers, paid for all the piping, valves, pumps, insulation, maintenance on 10 boilers etc. A properly designed plant would have been cheaper upfront, last longer, and would be cheaper to operate. I really don't see where they saved money, every step in the life cycle was more expensive. If you retrofit a boiler plant, the boilers themselves are the cheap parts. All the work around them is what costs money, and more complexity costs much more. And installing a cheap crappy boiler costs exactly the same as installing a good boiler in a given design. I designed and managed a few boiler plant upgrades and purchased all major parts and hired the sub contractors, trust me, the boiler cost differential between good and bad boiler is a very small % of the overall cost. Adn that is replacing 2 boilers with 2 boilers. Replacing 10 boilers with 2 instead of 10 must be much cheaper. Make it as simple as possible, and as complex as necessary.