Cooling solution for enclosed stationary engine

[motorambler]

I am trying to figure out a cooling solution for a stationary generator set that will be housed (relatively tightly) by a steel enclosure. The engine is ~8L in size, outputs about 150kW, and is fueled by an exotic gas which means it burns very very hot.

My only concern is the engine block not receiving enough airflow/cooling.

The enclosure is simple in that it's basically a rectangular box with louvers on each long end. The louvers are 8' tall x 4' wide and also act as walls.

Right now my design is such that the engine driven radiator fan sits right next to one louver and "pulls" air through the louver on the other end of the enclosure (8' away). This air would travel over the alternator, the engine, through the radiator, and exhausted through the other louver.

Right now I am having issues with the engine getting way too hot. The air seems to be taking the path of least resistance by mostly flowing up and over the generator set instead of in an even manner which would also keep the block (which is obviously down in the lower area of the enclosure) cooled.

I thought of implementing a fan on the other end in a push-pull configuration but don't know if this will solve my problem or just give me more airflow where I don't need it.

Do you guys think wings (spoliers/airfoils) attached the the side walls (or floor) and directed towards the block and oil-pan area would be my best solution? I don't know if differences in pressure would make this a great or terrible idea.

Or should I look at installing additional fans on the longer side walls and direct the airflow directly onto the engine/block?

I am open to any suggestion that works best at solving this heat issue. Cost is to a certain degree not an issue if it means the block is optimally cooled. I am willing to sacrifice other factors (noise, space requirements, etc) to satisfy this singular goal.

 

[MikeHalloran]

You need to be sure there's a shroud around the radiator fan so that all the air going through it is forced through the radiator and out through the louver, and doesn't recirculate through the interior of the box.

Consider baffles, walls or shrubbery outside the enclosure to suppress any recirculation external to the enclosure proper.

Add thermal blankets to the exhaust manifold and all of the exhaust system downstream within the enclosure to limit heat gain from that source.

Check the engine's cooling system, especially that the thermostat is installed properly and operating properly.

Before doing any of that, ask the engine manufacturer for guidance. They should have extensive and specific data on how much heat goes where, at least for more conventional fuels.



Mike Halloran
Pembroke Pines, FL, USA

 

[zekeman]

I would go to the mfr and find out the maximum ambient temperature and cooling airflow and block temperature spec for the unit It's his job, not yours.

If I buy this generator set, it is implied that the mfr should be giving me minimum cooling data. They can usually assist you with your design basd on your containment requirements.

After you get that data, then you can configure the cooling system intelligently instead of trial and error.

After that , if necessary we can offer assistance for the ventilation system.

 

[motorambler]

Hi guys,

thanks for offering to help. There is no manufacturer, per se. This entire piece (engine, alternator, enclosure, everything else) is a custom designed unit. The solution will have to come from me.

It's simply airflow basics that I need help on.

Picture a birds eye view of a 8' x 4' box with louvered ends & housing something that closely resembles this:

http://www.dieselgeneratorco.com/threephasedieselgenerators.jpg

The radiator fan is the only active cooling device. The fan exhausts air directly out the enclosure via the louver right next to it, thus air is drawn through the opposite louver and into the enclosure and over the genset.

Should I add a "pusher" fan on the opposite end? Should I add fans on the long ends of the enclosure & aim them right at the block? Spoilers/airfoils? Something else?

 

[MintJulep]

You are using the term radiator. This is colloquially understood to be a liquid-to-air heat exchanger with engine jacket cooling liquid on the liquid side and air on the air side.

Is our understanding correct?

If so, then you engine would seem to be designed as a liquid cooled engine. Focus your efforts on improving the efficacy of the primary cooling system.

Consider adding an oil cooler.

Consider segregating the radiator cooling air and the block cooling air.

Consider adding fins to the block to enhance heat transfer.

 

[MikeHalloran]

>>>This entire piece (engine, alternator, enclosure, everything else) is a custom designed unit. The solution will have to come from me.<<<

If you are designing an alternator from scratch, and designing an engine from scratch, you will be busy indeed.

On my planet, we _buy_ items like that from someone who mass produces them. I even know of a shop that produces custom enclosures in substantial quantity, and another shop that produces custom skids. I'm sure there are many more of both.

You used a Cat skid-mounted genset as an example. Cat also sells loose genset engines. In either case, Cat can _bury_ you in information about any of their engines, including the numbers you would need to predict temperatures in an enclosure. So can all the other engine manufacturers. In most cases, you need merely ask their application engineers for help in selecting a particular engine model, or for specific information on the engine model you have selected.

Given the heat rejection to coolant and the heat rejection to ambient, you can call up a radiator manufacturer and get help in picking a proper radiator and fan. For Cat engines, you can get help selecting parts from their catalog, including radiators and fans, or you can use third parties.

Again, all you need is the engineering numbers, which you can get easily, and which you can use, if you are an engineer.

If you are not an engineer, this would be a really good time to rent one for a while.


Mike Halloran
Pembroke Pines, FL, USA

 

[motorambler]

Thanks guys, judging from the replies I'm seeing here it's plain to see that my post was entirely lost in translation. This is likely my fault.

Anyway, we have come up with a suitable solution is both simple and inexpensive.

Cheers.

P.S. Mike, yes the one-off alternator was built for us because there was not one in existence that met our spec. As were the heads, CGI block, and custom intakes for a hydrogen fueled engine.

 

[MiketheEngineer]

Care to share the solution??