Off Grid Building 2

[Mbrooke]

So I'm designing a co-gen power system for an off grid remote research building. I'm not really sure where to begin- I need a few pointers/guidance on where to start...

Does this look like a good design? Should I get creative with how the gensets are interconnected?

Or just keep things really basic?

Anything I've missed? Go/no go? Or am I way off?


Power is normally produced by natural gas and solar with onsite diesel tanks to meet the requirements of article 700.

https://files.engineering.com/getfi...abd-4b3c-8a21-98610d4667c5&file=IMG_0040x.jpg

https://files.engineering.com/getfi...ad1-4634-8dbf-62914832e2a8&file=IMG_0041x.jpg

 

[Mbrooke]

Power will be out for only cycles, UPS for what really can't handle it.

Motors are an issue, I'm thinking a delayed transition ATS to demagnetize the motors before re-energizing them.

 

[Mbrooke]

@Hoxton: The hope is that the another gas gen-set or a diesel genset will take over for maintenance and failure of the main set. Letting the diesel run during scheduled outages will allow the genset to warm up without having to load bank thus preventing wet stacking.

What gensets did you have that shut down on low load?

During the summer with cooling the load will be steady, perhaps near full load for 12 hours a day. Other things like lighting, refrigeration, computers, lab equipment, ect will present a steady load.

There might be absorption cooling using the gen's waste heat if that helps.

 

[cranky108]

There are packages for diesel set that provide cooling from the rejected heat.

Just saying.

 

[Hoxton123]

The sets were MWM (Now CAT)

I think other high tech engine control systems had a similar feature. The problem is lube oil carry over at low loads / low turbo pressure causing knocking and spark plug fouling.

 

[Mbrooke]

I hope CAT isn't the same today. I plan on [highlight #CC0000]**auto correct messed up here**[/highlight] CAT and ONAN gen sets as they seem to be the most reliable.

Do smaller units have the same issue?

 

[davidbeach]

Suing?

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[Mbrooke]

LOL- auto correct strikes again!

I plan on using USING CAT and ONAN sets.

My bad.

 

[itsmoked]

Just so you know there are no more Onan generators they were bought by Cummins in 92 and for a while were called Cummins-Onan generators when they were Cummins prime-movered. Now they seem to have dropped the Onan with the exception of residential and RV where they do still call them Cummins-Onan probably riding on past consumer memory. They seem to have stuck with the Onan pukey green though.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Mbrooke]

And thats why I think of them as ONAN (from the green color).

CAT and Cummins seems to be the best, Ive seen to many Kohler units fail as the worst times.

 

[Mbrooke]

I'm nearing on something like this- 3 gas units 3 diesel units:

I hear smaller ATS are more reliable than large ATSs.

Debating whether I should rotate the feeders around that the total load served by any gas unit has leads to more than one diesel unit but I can't see any technical need for that.

Still not 100% set on number of units or ther size- ie keep all the gas units the same size or have different sizes.

Or if I should combine any of the diesel units since they are physically smaller (and more inrush tolerant).

 

[itsmoked]

Gotta say from a user's perspective I don't like any of this at all. I'd not like any of these proposed non-solutions. If you're going to power off-grid a facility then I'd expect power reliability and operational freedom like the grid supplies. This should be N+1 straight-up. You run the number of generators needed paralleled as more are added or dropped. An extra there for maintenance.

If you use what I'm seeing you will have management cursing you and coming to spit on your grave after they have you bumped-off. Operations does not want pieces of the plant dead nor does maintenance want all these confusing disjointed circuits with double and triple the conduits running everywhere, machines next to each other with one that can work and one that can't. Truly an operational nightmare other companies they compete with won't have and at greater expense.

If you have N+1 there is no logical reason you can't have endless reliable power. Entire towns run this way. This gas/diesel thing is crazy too. If you really can't get a single fuel to do the job then use bi-fuel engines. I live next door to UCSC where they have on-site generation that is a 2.5MW diesel generator that they normally run on 5% diesel 95% NG. If they have NG problems it goes to 100% diesel. Do the same thing. Having a set of diesel and a separate set of NG generators is bonkers!! They'll always run the NG generators and eventually you'll have a bazillion gallons of aged-out crap diesel they'll have to deal with. <shudder>

Just saying.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Mbrooke]

Uhhh... there is N+1. Thats what the ATSs are for. If the gas gen fails or need to be taken out for maintenance the diesel unit starts and load transfers over in cycle to seconds. The diesel fuel gets burned off and the gen is fully loaded preventing wet stack. Because the gas unit needs to cool down, checked over, oil changed, ect you are looking at least 96 hours on the diesel unit.

There are no double or tripple conduits*. 2000 amps will require parellel sets no matter what, so you could either combine all of them into a single circuit or multiple smaller circuits totaling 2000amp. You'll have more panelboards and gear, I'll give you that, but at the same time I'd rather use smaller ATSs than one big ATSs. Also a bigger breaker is less likely to trip before the gen stalls out.

*Some parts of the facility will have extra conduit via normal power besides life safety, COPs and so forth. The idea is that if the COPs goes down, equipment can be unplugged from the red outlets and plugged into the white ones like done in a hospital. Same with the stairwells, life safety runs up the shaft, but a normal conduit from each floor hits another supplementary luminary on the landing. All electrical rooms and critical areas have two switches, a lighted toggle off the life safety branch running 50% illumination a white toggle off the normal branch running the other 50%. Gen rooms, e;electrical rooms, telephone closets, tunnels, ect may also have bug eyes or a Bodine on the fixtures fed from the life safety/critical operations circuit.

 

[FreddyNurk]

So I'm curious, what alternators have the OEMs advised that you can use for your proposed voltage and power ratings?
Whilst I'm on the other side of the world with different voltage and frequency, I've not seen a datasheet from the alternator OEMs that provides the power level you've shown with the current capability on your drawings.





EDMS Australia

 

[Mbrooke]

Here is a "standard" 1000kw diesel data sheet:

http://resources.kohler.com/power/kohler/industrial/pdf/g5580.pdf

I can get up to 3470 amps at 120/208Y, at 125% that gives me 4337 amps rounding up to a 5000 amp breaker.

127/220Y, 139/240Y, and 230/400Y for anything international. 277/480 if needed.

A gen set in a prime power application may take the amp rating for that instead of the standby rating.

This why gen's breaker may appear slightly undersized or grossly over sized relative to the kw rating.

For anything none standard or exotic I can turn to Stamford alternators:

https://files.engineering.com/getfi...63-ab41-3474a49973e2&file=ratings_book_08.pdf

https://files.engineering.com/getfi...86-b4ea-b7794db70cef&file=ratings_book_11.pdf

 

[catserveng]

Have you actually been to an island mode cogeneration site in the size range you're looking at? When you add heat recovery and the associated systems then you have to manage engine loads to that all the balance of plant works like its supposed to. So along with many of the issues already raised, what about factors like environmental, such as ambient temperatures, humidity, sea air, etc. Also will these units be operating in an area with air quality regulations?

What kinds of loads? You said research facility somewhere in your post, so where at, what kind of research? You looking at something like an agricultural research station in a fairly friendly climate, or like an arctic research station?

Mixed engine plants aren't all that uncommon, especially in systems like mining and oil and gas sites, sizing and operation depends on a lot of factors, like cost of liquid fuel, quality of gas fuel, operating cycles and so on.

Bi-fuel engines can be a good solution depending on sizing, load factor and fuel quality. And designs range from units like Keith mentions with very high replacement rates, to more common systems that use lower replacement rates but are usually bolt on additions to existing diesel engines.

Not sure what really is your heartache with a paralleling system for what you're proposing, ATS's have their place, but in trying to use them as you've described you may need to spec "dual prime" switches, as the typically have more robust contacts and controllers.

You also mentioned at some integrating solar, which could be fairly simple or a real PIA depending on a number of factors, so unless you can provide a lot more info about what you are really trying to you.

If nothing else, this post has been quite entertaining for me, good luck in your venture.

MikeL.

 

[itsmoked]

Thanks Mbrooke for putting up with my pained sensibilities.

I'm still saying you don't really have N+1 capacity it seems more like N+3, too many excess generators.
You're still working the emergency generators, COPs (whatever they are), and life safety. The whole premise; its operational and cost complexity, can be dumped if you have a flat N+1 deployment.

There is no need for emergency stuff in this case because the system isn't hooked to a POCO grid that they have no control over and can be lost for whatever reason. As an OFF-Grid facility there is no reason to ever have it all go down unexpectedly. You have hot standby that can come online faster than an emergency generator. Hence no emergency circuits, no life safety, no COP.

I hear your concern about gear size, but. Every other non-off grid place has one incoming buss from the POC and it gets distributed in the standard every-one-does-it way. They wouldn't have a butt-load of ATS, Probably not a single one.
This place should be built the same way. Plan it like the power is from a POCO then build a power plant and connect it to the standard industrial structure.

Doing this allows complete freedom in the future to supply power anyway they want it. Co-Gen, solar, wind, walkaway nuke, maybe the grid even shows up. It all comes in thru the same portal. The existing generation can augment whatever's available. Maybe solar can supply 70% during the day. Drop off 2 prime-movers. Add one when the sun goes down. Maximum flexibility from the get-go. Consider these contingencies in your N+3 design, they'd be so painful as to be non-starters.

Having a flat design means you could have one generator running at 3am. The above design can never have less than 3 generators running 24 hours a day. That can't be as cost effective.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Mbrooke]

Have you actually been to an island mode cogeneration site in the size range you're looking at?

Honestly- I have not been or ever seen an off grid system of such size in person.

When you add heat recovery and the associated systems then you have to manage engine loads to that all the balance of plant works like its supposed to. So along with many of the issues already raised, what about factors like environmental, such as ambient temperatures, humidity, sea air, etc. Also will these units be operating in an area with air quality regulations?

I agree, however power will take priority. If absorption, heating or domestic hot water calls for less heat than the engine is outputing the extra heat is released outside.

What kinds of loads? You said research facility somewhere in your post, so where at, what kind of research? You looking at something like an agricultural research station in a fairly friendly climate, or like an arctic research station?

Computers, server room, office equipment, lab equipment, things of that nature.

Mixed engine plants aren't all that uncommon, especially in systems like mining and oil and gas sites, sizing and operation depends on a lot of factors, like cost of liquid fuel, quality of gas fuel, operating cycles and so on.

Bi-fuel engines can be a good solution depending on sizing, load factor and fuel quality. And designs range from units like Keith mentions with very high replacement rates, to more common systems that use lower replacement rates but are usually bolt on additions to existing diesel engines.

I guess this is a moment of learning for me. How are Bi-fuel engines better? Are they more efficient? Do they last longer?

Not sure what really is your heartache with a paralleling system for what you're proposing, ATS's have their place, but in trying to use them as you've described you may need to spec "dual prime" switches, as the typically have more robust contacts and controllers.

Paralleling has multiple single failure points and presents knowledge/experienced based complexity that not every technician has or understands. Second it tends to force all the gen sets into one room vs diversity of sources- typically this will never be a problem until fire or some other disaster makes itself known. Further it is difficult if not impossible to parallel a mobile unit should one or more end up being brought in to feed the gear during an emergency. I can get just about any mobile unit of any size on wheels, but what guarantee do I have that the paralleling gear will connect to it control circuit wise, recognize it, and be of identical parameters to the existing onsite sets or the other mobile sets?

You also mentioned at some integrating solar, which could be fairly simple or a real PIA depending on a number of factors, so unless you can provide a lot more info about what you are really trying to you.

If nothing else, this post has been quite entertaining for me, good luck in your venture.

Why is this entertaining if I may ask? What do you see me doing wrong here?

 

[Mbrooke]

Thanks Mbrooke for putting up with my pained sensibilities.

Nahh- I need someone to put this through the wringer so to speak. More minds are better than one

I'm still saying you don't really have N+1 capacity it seems more like N+3, too many excess generators.

KW capacity wise there is a lot more excess than if the gens were paralleled and of the duel fuel type. Sure. I will recognize this part.

You're still working the emergency generators, COPs (whatever they are), and life safety. The whole premise; its operational and cost complexity, can be dumped if you have a flat N+1 deployment.

Critical Operations Power System- see NEC article 708:

https://ewh.ieee.org/r3/atlanta/ias/Article 708 - What Hat Are You Wearing.pdf

There is no need for emergency stuff in this case because the system isn't hooked to a POCO grid that they have no control over and can be lost for whatever reason.

Huh!?! Honestly, its statements like this which make me question the knowledge/experience/understanding of engineering in some of the members posting here. I only make this rebuttal on the basis that another member said I was entertaining- so I figure the discussion is open for tit for tat on both sides going both ways

But, I'm glad you said this. I think this where a lot of the disputes in this thread are originating from. A lack of understanding in regards to code, critical power and not having all your eggs in one basket.

There is very much a need for emergency stuff:

1. Article 700 of the NEC (along with NFPA 101 and driven by pother applicable building codes / AHJs) still require life safety systems in buildings even when supplied by an off grid system.

2. The Normal source of power can fail for a variety of reasons, in a variety of ways even when fed from a co-gen.

3. Article 700 requires separate circuits reasonably protected from fire in regards to emergency egress lighting where as normal power does not have this requirement.

Granted in remote construction or government project you might be able to pick and chose what codes apply, if any, but do you really want to put people at risk? Do you really want a fire taking out the normal power system leaving an 8 story building pitch black to the point people can't find stairwells let alone be able to walk down them to get out?

As an OFF-Grid facility there is no reason to ever have it all go down unexpectedly.

Not when you are paralleling everything into a single piece of gear.

You have hot standby that can come online faster than an emergency generator. Hence no emergency circuits, no life safety, no COP.

I know plenty of places with parallel power production that can take over when the utility drops out without so much as a blink. Does that mean hospitals in these cases can ditch all the extra conduit, wiring, runs, isolation, dedicated rooms, dedicated shafts, panels, ATSs, ect for the separate branches? Required by code as a means to create redundant diversity and so a fault on one branch does not take out the other?

I hear your concern about gear size, but. Every other non-off grid place has one incoming buss from the POC and it gets distributed in the standard every-one-does-it way. They wouldn't have a butt-load of ATS, Probably not a single one.

So based on what you said:

...there is no reason to ever have it all go down unexpectedly.

The main bus will never fault? And be protected from fire? As well as all the other unprotected runs through out the facility?

This place should be built the same way. Plan it like the power is from a POCO then build a power plant and connect it to the standard industrial structure.

Ummm- if I was to plan this as if from the POCO I still need at least 3 ATS for life safety, legally required and COPs load.

Re-read what you wrote, at first you said I would need emergency circuits if only fed from a POCO, but are now telling me to design as coming from a POCO.

Doing this allows complete freedom in the future to supply power anyway they want it. Co-Gen, solar, wind, walkaway nuke, maybe the grid even shows up. It all comes in thru the same portal. The existing generation can augment whatever's available. Maybe solar can supply 70% during the day. Drop off 2 prime-movers. Add one when the sun goes down. Maximum flexibility from the get-go. Consider these contingencies in your N+3 design, they'd be so painful as to be non-starters.

Having a flat design means you could have one generator running at 3am. The above design can never have less than 3 generators running 24 hours a day. That can't be as cost effective.

I'll give you that- the gas units will see low load levels at night during cool weather periods. This is where I need to really think about paralleling or just accept poor fuel economy.

 

[davidbeach]

Yes on emergency generation. And it should have sufficient capacity to provide what ever blackstart resources the main power plant needs. One ATS for the whole kit and kaboodle. Enough paralleled generation in the main power plant to allow any single unit to be out of service at any given time and paralleled so that the building occupants never know that generation sources have been redispatched.

I’ll see your silver lining and raise you two black clouds. - Protection Operations

 

[FreddyNurk]

Honestly- I have not been or ever seen an off grid system of such size in person.

Honestly, its statements like this which make me question the knowledge/experience/understanding of engineering in some of the members posting here. I only make this rebuttal on the basis that another member said I was entertaining- so I figure the discussion is open for tit for tat on both sides going both ways

Not really, the reason for the pushback is that its obvious that the first statement is correct, and the assertions that people who do have experience in such off-grid facilities aren't following code are not likely to be accurate.

Paralleling equipment does not require everything to be in a single switchboard, or a single point of failure. Its done on maritime installations with paralleling to shore (grid...) as well as multiple sets and the capability to have additional emergency sets all the time.

There's no forcing all generators into a single room to allow for paralleling, nor is it impossible to parallel mobile or rental sets, politely, you're inventing reasons that do not exist as an argument to continue your designated solution.

As for the code compliance, what loads are designated as emergency? If there's secondary fire pump systems or the like, then they can be treated appropriately. If there's other critical loads then the overall design can cater for that as well, whether it be by a dual distribution system with an intertie arrangement or similar.