Micro CoGen.

Quote from: Number21 on June 11, 2016, 09:52:28 PM
If you want to make any further replies here:
The subject is highly efficient variable speed 30hp natural gas engines. Nothing else is acceptable to talk about here.

Quote from: glort on June 12, 2016, 12:49:06 AM
then why aren't you looking at engines that run on waste oil? You are not going to find a more economic fuel than free now matter what your efficiency.

In any case, if the arrow engines are so good on NG, just go buy one of them.

Waste oil is not free in industrial quantities anywhere in the world, never has been, and never will be. If it was somebody would buy it. Arrow engines do not meet the qualification of low cost and readily available. Does not mean a similar engine could not be built from automotive stock. Arrow engines are great examples because they publish a lot of really accurate, highly detailed fuel consumption information focusing on natural gas.

As for your other comments, I'm not even going to begin to have a conversation here with you about real estate, and you definitely have no idea what I'm doing, and no knowledge of construction prices in my area. If you can't stay on topic find somewhere else to post.

Quote from: mobile_bob on June 11, 2016, 02:37:05 PM
it has probably already been mentioned, but how about a dual fuel diesel engine?

the thermoking/isuzu c201 is a 4 cyl engine, that runs forever it seems
my bet is you could either dual fuel one of those or convert one to spark ignition
to burn nat gas only.

or use any number of other 4 cyl diesel engines

those engines are generally very heavy built, relative to a gas engine, the compression ratio is already high enough, and they are not that expensive?

alternatively use 2 or more smaller 1 or 2 cyl diesel engines, that way you can tailor the power to the load more effectively, given your need to have output over a fairly wide range?

without going back and rereading all this, here are a couple more thoughts

it seems to me i read you were asking whether or not you could increase the air to the cylinder and decrease the nat gas and still get it to run... the answer is generally no.

there is a fairly narrow fuel/air ratio where the engine will reliably start and run, getting outside this range will be met with hard starting, loss of power, stalling, misfiring, and possible engine damage, let alone having issues with EPA limits on nitrogen compounds in the exhaust.

if it were me, given what you want to do, i would go with multiple smaller displacement engine's or find a large single cylinder engine from the oil patch that is already setup for running nat gas.

30hp seems like you are looking at pumping a massive amount of water, relative to what most offgridders would have in mind.

bob g

Thank you, Bob, for posting something extremely useful, as always. If only we had 10 Bob's around here.

I have considered duel fuel engines, but, I need to calculate the total fuel consumption, including diesel, and decide whether it's worth the increased cost of diesel. Diesel cost more but I also think it makes the natural gas side burn more efficiently, so it might equal out. My gas rate is only $0.52 per therm, and offroad diesel is around $1.60/gallon right now. Diesel will likely rise, while gas will probably stay the same for the foreseeable future. It's also nice because my engines would continue to function on full diesel in the event that natural gas supply is disrupted, however unlikely that is. I also might be able to burn some of the waste oil I produce.

And you are absolutely right, I need several smaller engines working together, that can be turned on and off independently, not one giant machine. That has been the entire subject of this post, and a few people can't seem to get it through their heads. Too bad there are only 4 people on this forum....

Quote from: mobile_bob on June 11, 2016, 02:37:05 PM
let alone having issues with EPA limits on nitrogen compounds in the exhaust.

Just to be clear - if I could find an engine that ran economically on old tires and ground up baby whales billowing black smoke, I would. I have no EPA concerns of any kind.

Quote from: SteveU. on June 11, 2016, 12:37:08 PM
Sorry Charlie, you are wrong.

You know Steve, as long as I've been a member here, I've always thought your posts were generally rude and off topic. This one is no different. I'm trying to have a conversation about the most efficient engine in a specific size, that's it. I should have never told you what I was using it for, I was just being nice, and thought some of you might be interested. There is no "wrong" when I am simply posting thoughts and asking questions. If you think there are wrong questions you should not be here.

Quote from: SteveU. on June 11, 2016, 12:37:08 PM
Ass-sumptions.

You and Glort are the only ones making ASSumptions. I have not assumed anything, only asking questions. You have certainly made an ass of yourself here. I have not given you enough information about my project to make any assumptions what so ever, but that hasn't stopped you from trying.

The problem here, is that MICRO cogeneration is a dead end, always has been, and this forum is dead, no more than about 4 people ever respond anymore. It's great if you want to share your opinion, Steve, but I'm not going to bother arguing with you about what engine I want to use or why, and I'm certainly not asking your permission, nor do I want your singular advice. Large scale cogeneration, on the other hand, is something that will continue to grow and grow as the years go by, and is an increasingly common subject.

If you had bothered to take a look at the info I posted from Arrow engines, you'd see there is a huge difference in fuel consumption among NG engines. They rely heavily on load, speed, and air flow. Simple things like a crossflow head, turbochargers, compression ratios, and load ratios, have a big impact on the BTU per horsepower consumed. The information is easy to see on a graph from just one engine with varying loads. Your suggestion of a big six cylinder running slow is a bad one, that is not an efficient way to run any natural gas engine. You will find if you read the fuel consumption graph on a 6 or 8 cylinder engine, at 25% load, it burns more fuel per horsepower than at 75% or 100% load. Thus a more heavily loaded 3 cylinder will use less fuel under the same load. Slower engines also have worse BSFC numbers, because the cylinders have more time to absorb heat. And Arrow published some very specific fuel consumption information comparing an inline 4 and inline 6 without crossflow head, and with crossflow heads, the crossflow heads make a substantial difference on the same engine in the same application. Same with turbos. A company like Arrow would not be experimenting with crossflow heads and turbos if they didn't actually do something important. It's not rocket science, the information is highly published and easy to read. I would have to be really stupid to chose an engine that burns 12,000 BTU/HP when there are engines available that burn less than 7500 BTU/HP. That is an INCREDIBLE difference in efficiency! Yes, inline 6's are great motors, and I love them, but they are not efficient on natural gas without crossflow heads, and they are also not efficient at partial load. An inline 6, pushing 30hp or less, will be in the 12,000 BTU/HP range.

Quote from: SteveU. on June 11, 2016, 12:37:08 PM
So stop talking and start Doing like all of the active members here.

What an incredibly stupid thing to say. I only asked this question 8 days ago. My timeline for getting this done is absolutely none of your business. And the vast majority of what is talked about on this forum is theory that is never actually put into use. With that kind of attitude I'm not going to bother to come back and share my results. This kind of attitude is why this forum is dying...or pretty much already dead. "All the active members here"...what "active members"? Have you seen how dead this forum is? Most of the people that still post here don't even have an operational cogen system!

And I'm willing to bet a thousand dollars that those pinto/Onan engines you speak of were never optimized for natural gas, and wasted a shit ton of fuel. Fuel efficiency was likely one of the last things they had on their mind. Even today, small automotive engines used in generators for natural gas are not usually optimized for the NG, they are simply low compression gasoline engines fitted with a mixer. If you are a telecom company, gas is cheap, waste away. I am not a telecom company.

If you want to make any further replies here:
The subject is highly efficient variable speed 30hp natural gas engines. Nothing else is acceptable to talk about here.

Quote from: glort on June 10, 2016, 07:34:49 PM
You are going to be needing a lot more than pissy little single/twin industrial engines here with the loads/ functions and most of all RELIABILITY this is going to require when you have 20 families/ businesses relying on you.

I think it all needs to be mapped out NOW and you are going to need an expert to work out the loads and requirements of what you are going to need.
That said, worrying about costs on this " toy" engine seems bit of a moot point. What you are talking about is a multi million dollar development so if you are worrying about the $500 cost of a governor for an engine and one that could be far more in line with your ultimate needs than immediate ones, I have to wonder about the likelihood of this all coming off.

Way, way off. This is why I tend to keep my posts vague, otherwise I get people answering questions I didn't ask and taking things way out of context. The topic here is I want the the most efficient 30hp engine I can find, that's it. I shouldn't have said why, it doesn't really matter. Let's just pretend I entered a "most efficient engine" contest.

Now, as for why you are incorrect:
Nobody is relying on this system. Like I already said, the units will be built with standard hydronic heating and cooling. This is simple stuff they have been doing for more than 100 years. Common, cheap equipment will be used to heat and cool the water on a constant reliable basis, and the grid will supply the same type of power. No, I won't be spending millions on this project, I find that quite laughable. I suppose it might approach near a million when I'm all done. (Appraised value, not what I actually spent) This system will use regular grid power and be billed to tenants at whatever that ends up costing. (via BTU meter measuring water delivered)

The cogenerator(trigenerator) is simply an add on. If it doesn't work, then the primary grid powered system continues to function as a normal, highly inefficient heating/cooling system. Given the cost of electricity, compared to the cost of natural gas, I can generate electricity for about the same cost as the power company, including engine maintenance. I can then sell this energy to tenants at the same price they would pay the utility. Now I have thousands of free BTUs to use as I please. I can use those free BTUs to make domestic hot water and heat living units, which again I can charge for. I can also spend about $40,000 on an absorption chiller to convert that hot water into cold water in the summer. All of my heating and air conditioning needs can be met with free waste energy, and, can be sold for a lot of money every month.

Having a completely hydronic system allows me a lot of flexibility, I can feed it with many different source of heating and cooling, with reliable backups. I could also supplement with a waste oil or wood burner, which I could even use to make air conditioning with an absorption chiller.

Yes, the equipment will be expensive, but I'll also get paid a few grand a month to have it. As for those "pissy little single/twin industrial engines", well, I never asked about anything smaller than 30hp, I don't consider that "a little pissy engine". Concerning efficiency and reliability, there needs to be several smaller individual generators, which can either work alone, or together. If one is down for maintenance I'll have a backup. If the power consumption at night is low I only need to run one generator, and during peak periods I can run two or more. As many as I need. Having one common engine means it's easy to keep parts and whole long blocks sitting around ready to go.

And yes, efficiency is a big, BIG deal. EVERY single BTU matters when you are running 30, 60, or 90 horsepower 24/7. Just a difference of 500 BTU per horsepower adds up really fast. The fewer BTUs I use the more money I pocket.

But let me repeat: I never asked about any of that. I simply wanted to explain why I am asking. The only topic here is high efficiency 30-ish horsepower engines. The simple economic problem is that the type of engine I need is mostly only available in very large power units, which I do not need. Supply and demand makes bigger engines and generators cheaper than the small-ish ones I am looking for. The ideal engine that I want, a 3 cylinder iron block with a crossflow head and easy availability of short blocks simply doesn't exist, so I'll have to come up with a compromise.

Quote from: glort on June 10, 2016, 07:34:49 PM
My advise would be go with what you will ultimately use NOW so you have the development stage of the rest of the project to sort out the bugs and learn it all.

That is exactly what I am asking for. At this time I do not need any cogenerator engines or generators. I do have an immediate need for a 20-30hp water pump (it will be variable speed) and since I need that anyway I want the same engine here that I'm going to be using over and over everywhere else. I'm not going to rely on what I read on the internet, I'm going to find a good engine, modify it the way I like, and then actually study it to see how that effects fuel economy and engine life. We can argue about theories all day but nobody really knows until you get out there, do it, run it, and measure it.

So, to sum up:
- Kubotas are nice but big money and not easy to find
- Briggs and other aluminum engines are cheaper, probably less reliable, but come with a 2-3 year warranty
- Most NG engines converted from gasoline are not as efficient as they could be
- Converted car engines have a few issues that must be solved, not impossible, but, also few car engines are available with cast iron cross flow blocks in smaller displacement

I said I would prefer a car engine because even though I would have to spend a lot on a governor and shaft adapters, those parts would last indefinitely and automotive long blocks (all the wear parts) are usually cheap to replace.

Quote from: Tom on June 10, 2016, 03:59:48 PM
Thinking outside the box here, but how about a power plant from a hybrid car converted to NG? Start with a junk yard engine/alternator combo, shave the deck to  up the compression and fit it with an impco carb? I do believe they are in the correct KW range for your needs.

I've thought about that - the Prius engine being a modified Atkinson cycle is supposedly 40% thermally efficient on gasoline. My concern is the actual conversion to NG, I think it would be very difficult if I did not completely eliminate the OEM computers. And if I did that, will the engine even run? Something needs to control the VVT, and I really have no idea how that works.

When I do a google search for CNG Prius, all they have are some very expensive factory converted examples simply designed to make an environmental point...I don't think anybody is actually driving one or converting them on any sort of commercial scale. There must be a reason? Surely it's possible I'm just not sure how difficult...

With that said I have seen some youtube videos of Priuses converted into gasoline generators, and I suspect a used Prius with bad batteries doesn't cost very much. If the Atkinson style valve timing could be kept, it might be a great engine to use.

Quote from: Tom on June 09, 2016, 10:29:43 AM
Look to an large truck alternator for a variable rpm unit.

I'm looking to build something in the 15KW range, so I need a BIG alternator.

I assume if you spin them slower than 1800 you would get less than 60v?

The inverters I want to use accept 45 to 90 volts, so I could try to belt drive the gen head within that range, without going much over 1800 max. A 3 phase head is probably the best idea.

I'm looking at various options for the most fuel efficient natural gas engines I can find. My conclusion is that all of the modern and most fuel efficient engines are very expensive and use some sort of lean burn technology, which is highly computer controlled.

It is possible for me to convert a standard, old school spark ignited natural gas engine with a simple carb/mixer and low pressure gas to run on a highly lean mixture? Lets say I take a naturally aspirated engine and add a turbocharger, doubling the air intake, but do not increase the fuel rate. Would I have problems running the engine? Can this be done without a modern computer system to control everything?

The idea/hope is to use an engine that is larger than necessary with a small fixed load for longevity, while still keeping the throttle near wide open to reduce pumping losses, just like a diesel.

Can this be done by a backyard mechanic or is this just for university studies and high dollar computer controlled gensets?

I am interested in putting together a large-ish inverter generator that would generate 60v using an ST head modified as described elsewhere on the forum, and then feed the 60v into an inverter. The engine speed would vary depending on load.

What I am curious about figuring out, is what happens when you over speed or under speed an ST generator head modified for 60v output? Obviously the frequency would be all over the place, but that's not a problem if I rectify into DC. What happens to the voltage when you spin an 1800 RPM head faster than 1800 RPM? Or slower than 1800 RPM? Does the head still maintain the same or similar voltage?

And will an 1800 RPM head survive if you ever spin it much faster than that?

Quote from: SteveU. on June 07, 2016, 04:45:48 PM
Yes absolutely in a narrow speed load range of operation a cast iron only four bearing,cam-in-block, crossflow head,  inline six could beat a three cylinder GM/Suzuki's fuel gas use butt.

Are there any readily available automotive inline 6s with a crossflow head? That's the big sticking point with the inline 6, most of them are not crossflow. The Arrow engine I was talking about above would be great but I'm sure far out of my price range and dealer only. A CNG converted Cummins 4BT or 6BT engine from a bus on ebay would be fantastic, but none available right now. I've seen them in the past.

I wonder if the same would be true for a V6 with a cross flow head, for example, the 4.3L GM Vortec engine? Could I run that engine super lean on natural gas with the throttle near wide open?

I can get a reman 4.3L for under $2000. If I spent some money on a stub shaft and governor system, it can easily be transferred from engine to engine and probably last forever. With that in mind an automotive engine might be a better idea because long blocks are usually available for cheaper than industrial engines.

Quote from: SteveU. on June 07, 2016, 04:45:48 PM
So why do it?
Lots of already set up natural gas generator sets used/new out there that would fit the end result of your overall project much off-the-shelf better.
www.centralmainediesel.com their NG gensets are as affordable as you will find new. Look for these systems used, locally. You just add the heats harvesting.

My first goal is to settle on an engine I can use multiples of in the future. I want one engine I can use for a variety of things, from my water pump test to later trigeneration needs. They won't strictly be electrical generators, they will also have belt driven A/C compressors. The first one I buy for my water pump will be a test of what is possible in the future.

Also, most of the "cheap off the shelf" natural gas gensets are simply gasoline engines converted to NG. They are not efficient. I have researched a number of generators based on the 3.0L GM 4 cylinder, and without a crossflow head or high compression, it is not very efficient on natural gas. Companies like Arrow offer more efficient natural gas optimized gensets, but they cost a fortune. I'd rather put together my own system.

Quote from: SteveU. on June 07, 2016, 04:45:48 PM
In addition to your whole system projects of an HVAC, refrigeration, pipefitter, sheetmetal specialist. Only the Gov'Mint, or a pocket billionaire can afford to hire out for all of these.

The future rental units will all be setup for hydronic heating and cooling. There will be various thermal storage tanks in the system. Each heating and cooling system will have a standard backup like a natural gas fired boiler and an electrical hydronic heat pump in case any of my trigenerator systems fail. I can handle most of the necessary work and engineering for compressor refrigeration systems, plumbing, hydronic HVAC, ect. I have fun doing this kind of stuff, especially if I can make money at it. 

This is some fuel consumption information I found on the Arrow engine website. Their engines are all optimized for natural gas, not gasoline.

K series - big single cylinder that looks like a converted imported diesel:
At 4hp and 650 RPM, it burns about 8900 BTU/HP.

VR330 330ci inline 6:
Roughly 8200 BTU/HP at 1600 RPM and 60HP

VR330 with crossflow head:
7717 BTU/HP at 1400 RPM and 55HP

VR220 220ci inline 4:
8100 BTU/HP at 1800 RPM and 45HP

Compared to other gensets, using a 3L 4 cylinder gasoline/natural gas engine at 25% load, burns closer to 11,000 BTU/HP. I don't think using an inline 6 at a fraction of it's rated horsepower would be very efficient.

The best fuel consumption was with the cross flow head in a 330CI inline 6, but 55hp is too much. Cut that in half (inline 3) and it would probably be just right. If I could get anywhere near 7700 BTU/HP, I would be very, VERY happy!

Quote from: Number21 on June 07, 2016, 07:14:37 AM
Do you really think it is possible for an engine as large as an inline 6 to be as efficient while running at say, 1800 RPM, with a 15kw load, than a 3 cylinder running twice as fast with half (or less) the displacement?

Could I lean the A/F ratio on big engine way out to reduce the horsepower, and run the throttle wide open to reduce pumping losses, without burning the engine up?

I wrote a couple big long posts, although this was the main information I wanted to focus on. Anyone know? I'm kind of an inline 6 geek, if I could use one of those efficiently, I would be very happy.

I appreciate all the detailed replies!

Quote from: glort on June 05, 2016, 06:56:44 AM
As for the the thermo electric, what sort of power are you looking to generate? 200w Or 20 KW? I'm not aware of any practical setups to make power in the KW range with thermo but if there is, I'd be interested to find out about it.

I really do not know what is possible, other than one time many years ago I remember reading an article where they built a 1KW thermoelectric generator powered by exhaust heat to replace the alternator on a semi truck. If I could make an extra 1KW I would be very happy with that. (in addition to a small turbine generator)

I am interested in modules like this:
http://www.ebay.com/itm/172159943602
They claim with a 100 degree difference, just one of those modules produces 3.2 watts.

Say the creek is running at 70 degrees in the summer. And I have 220 degree engine water. That's a 150 degree difference. Even better in the winter. Whether it is actually economical to do or not I am not sure, just want to play with it. If the idea actually worked, I could make more hot water with wood, or waste motor oil, and turn that into power!

It has been my dream to own property with a creek since I was a small kid, and I'm going to do everything I can to extract every last bit of energy and geothermal heating/cooling out of this creek. It makes a lot of my land unbuildable, and therefore, I expect the creek to pay rent.   I have not had time yet to measure flow rates and temperatures of the creek.

Quote from: glort on June 05, 2016, 08:58:35 PM
I also agree with the idea of the inline 225 sixes which was what I was thinking in theroy as well.  A large underdriven engine in stationary use can be just as fuel efficent when set up properly as a smaller engine and of course far less stressed aiding longevity.

Do you really think it is possible for an engine as large as an inline 6 to be as efficient while running at say, 1800 RPM, with a 15kw load, than a 3 cylinder running twice as fast with half (or less) the displacement?

Could I lean the A/F ratio on big engine way out to reduce the horsepower, and run the throttle wide open to reduce pumping losses, without burning the engine up?

I've seen a range of fuel consumption on NG engines from 8,500 BTU/HP to as high as 10,000 BTU/HP. At 30hp, that is a difference of up to 45,000 BTU/hour that could be saved!

I've thought long and hard about converted car engines but I always run into a few major issues -
RPM control - I can add a governor but for a good one you're looking at $500 or more. (Somebody please prove me wrong)
Size - they are always way bigger than needed, the best I could find was a 1.0L Geo.
Output shaft - Need custom adapter to bolt anything to the flywheel, might be able to use the front pulley.

Car engines are always cheaper - I keep coming back to the 350 Chevy, I can buy a brand new one, not rebuilt, with a 3 year warranty, no core, for $1400 shipped, but it's way bigger than I need. I could also get a remanufactured Geo 1.0 for about $1600 if I had a core. But when you add the parts listed above, and consider they are less fuel efficient, they might start costing the same as an industrial engine.

I've considered looking for CNG powered junkyard engines, but, they aren't very common here, and I don't think any of the brains would be applicable to me in that situation. With low pressure natural gas, there is no fuel injection, so I can't see any need for a brain, just need a carburetor/mixer and RPM control. Most of them are low compression dual fuel capable anyway, not really optimized for NG.

My other option is to simply use a diesel engine with a diesel pilot ignition and the rest natural gas. This way the engine is already high compression, and maybe turbocharged, easier to find, and, built in speed control. No spark plugs either. I have to calculate if the increase in efficiency outweighs the higher cost of diesel fuel.

A little bit longer explanation about what I am doing:
I recently bought several acres with a creek where the natural gas is just $0.52 per therm. Immediately I only need an engine that will pump water, at varying RPMs. Over the next few years, I am going to be developing this property into a mixed use residential/commercial development. There will be perhaps 20 different rental units and I'll be responsible for supplying them all with electricity, hot water, hot air, and cold air. Long term plans include a dedicated trigeneration building. There will be multiple individual generators, for redundancy, and in the summer the waste heat will be fed into an absorption chiller to create more air conditioning. (North Carolina, hot at least 6 months of the year) I intend on investing a great deal of money into this system, but expect monthly paybacks.

This system will not be built for a couple of years, but, in the mean time, I want to settle on an engine I can use for that project, and do some experimenting with it. I want to find out under what circumstances I can get the best BSFC, and, what does it take to get the engine there. I will be doing careful monitoring for both fuel consumption and any other hourly running costs. Also somewhat of a reliability test. I need to find out for myself what really makes the best BSFC under my circumstances, not just some theory I read somewhere.

Simply increasing the compression on an engine is really easy, has little effect on engine life, and could substantially increase fuel efficiency for a dedicated NG engine. Any dual fuel gasoline/NG engine running on natural gas is not doing so efficiently because of the low compression ratio needed for gasoline. Thermal efficiency relies heavily on compression ratio. It's not very hard to take the head off, have it shaved a little, and bolt it back on, though it will void a warranty. I'd also like to go through the engine and do a little porting and polishing, mass produced engines always have rough spots inside and efficiency can be gained spending some time smoothing everything out.

Also, a turbo with low boost levels could basically raise the compression ratio without actually cracking the engine open. They are only a few hundred bucks and not very hard to install. Most of the engines I am considering do have a factory turbo diesel as an option, so I know they can handle it...and I can probably find OEM parts.

Thanks for the information. I hadn't noticed the aluminum head on the Briggs, (Diahatsu) I don't like that. Is there anywhere besides a Kubota dealer at full MSRP to buy a new 3 cylinder engine setup for spark ignition? There are none that I can find on any of the surplus markets. Only ones on ebay are beat up used engines. I might consider a diesel Kubota with NG fumigation, but I need to do some math on that.

The engine will be used to pump water out of a creek, and I'm going to use a heat exchanger in the creek to get rid of engine heat. This way there is no fuel consumption from a fan. It's also kind of an experiment for a cogenerator I want to build later on, I want to try some different things with the engine (like changing the compression ratio) and see how it effects fuel consumption. I'm also thinking about trying some things like ceramic and anti friction coatings in various places.

I'm also curious if I can't use the hot water from the engine and cold water in the creek to generate some power with a thermoelectric generator. I'm also going to install a small turbine in the creek eventually.

I am attempting to find a good liquid cooled engine in the 30hp range to run from natural gas. I am trying to find something that will have the lowest BSFC. I expect to either increase the engine compression or add a turbocharger to increase efficiency and take advantage of the high octane of NG.

In this power range, I have found 3 major engine types to choose from:

1. Briggs and Stratton or Kawasaki aluminum block V-twin.
2. Geo/Suzuki 3 cylinder 1.0L aluminum block inline 3.
3. Briggs and Stratton .7L cast iron block inline 3.

I'm leaning towards #3, mainly because it has a cast iron block. Is an engine of the same design with an aluminum block going to have a higher BSFC due to the aluminum losing more heat? Or can that be controlled as long as the engine has a proper thermostat? Also, the Briggs engine is a converted diesel, so it should be able to handle turbocharging quite well. (turbo diesel is a factory option)
http://www.ebay.com/itm/131797201535

My other concern is number of cylinders. I think two is better than three. Is a two cylinder engine likely to have a better BSFC than a three cylinder of the same displacement?

Are there any other engines I should consider? I would also think about the Kubota 3 cylinder gas engines, much like the Briggs, but I can't find anywhere to buy one that is new.