Hi John:
I've done some combustion testing not only on the Garn but also numerous other brands of OWB's and sealed system gasification units and learned a few things along the way. Probably the first thing I should point out is that CO emmisions will vary widely over the course of a burn on any wood boiler. There is also a huge variation determined by the moisture content of the wood.
CO is a normal combustion by product and the amount, usually measured in PPM (parts per million), is a function of how complete the combustion process is. The more completely the fuel is burned the less CO present in the flue gas. Any imbalance in the fuel/air ratio will result in excessive or high CO. Normally on a gas or oil fired boiler or furnace, the burner can and should be tuned to show a flue gas content of less than 50ppm. I have oil furnaces with Riello burners that will routinely hit CO of less than 10ppm once the combustion chamber is up to temp. Some of the Gas fired Viessmann boilers we install are so low you could literally breath the flue gas all day and not suffer ill effect.....from the CO that is. Oxygen deprivation would be a different matter.
Now where am I going with this?.......You'll notice the line I have underlined in the paragraph above. It is the key difference between a Garn and anything else that operates with a firing sequence that is cyclical. Most wood boilers maintain temperature by cycling the air flow available to the fuel. They have to because of a lack of btu storage available in proportion to the amount of fuel load. Obviously storage is not a problem with a Garn unless one was careless about firing it and loaded it full with the storage already at 190*.....but that's a different subject and I'm digressing from the topic.
The point is that anytime you operate a boiler on/off you allow the combustion chamber to cool in between burns. This in turn leads to very, very high CO (I've seen in excess of 10,000ppm many times) when the air hits the fuel once more when another call for heat is initiated. The combustion chamber has to heat up again before it will acheive the cleanest burn. This is why you see a lot of downdraft gasifiers with serious amounts of clay refractory in them trying to retain some of the heat of combustion.
With a Garn the combustion process is a steady burn which is hot, fast and continuous through the entire load of wood. This cut's overall CO emission by a large fraction compared to any other wood burner I have had the oppotunity to "sniff" with my analyzer. That being said, I have measured CO in excess of 2000ppm from a Garn depending on where the wood is in the combustion process and how wet it is to start with. Higher moisture content equals lower combustion temps which in turn equal higher CO readings. From what I have seen, a Garn will over the course of a burn emit much less CO than anything else. Once everything is up to temp with a load of correctly seasoned wood the CO drops far below that 2000ppm number. A boiler that cycles the fire is going to hit those extremely high peaks everytime is goes on/off.
AFA your question about wood smoke smell is concerned, you will get the normal wood smoke odor for a short period of time after initial light up. In actual operation, once the flue gas thermometer on the Garn hits about 325-350, you can literally stick you hand into the exhaust stream for a few seconds and then not be able to smell any wood "flavor" on it at all. There is very little odor present through the whole burn once that flue temp is established.
Most of the installations we have done are horizontally vented. I think there are 3 out of the whole bunch that are vertical for a variety of reasons. None of those 3 have a stack height of over 12'. Usually just enough to clear an eave. You do not need a vertical stack for draft with a Garn.
If it were me I would not be concerned with venting the Garn horizontally unless there are things particular to your installation that would dictate otherwise. In most cases it is not needed because the flue gas dissipates rapidly. At a distance of 50' from the house I wouldn't think you'd have any problem under normal circumstances. No wood burner, be it a horizontal or vertical flue, should be installed "upwind" of a house. That's just not smart for a lot of reasons in addition to CO and smell.
Hope that helps. :)
Edit: After re-reading my blurb here, I just want to clarify what I mean by very low CO once things are cooking. The initial stage of the burn is when you will see the high numbers. This happens even in gas and oil fired appliances. As I stated, I have seen numbers in the 2000ppm range but that is very dependent on the moisture content of the wood. Once the combustion process gets rolling, about 5-10 minutes the CO will drop to well below 50ppm in nearly all cases unless something is amiss or the MC is above 25%.