On thermal efficiency; it is nothing more than energy sought divided by energy that costs. Unfortunately, the devil is in the details. What is the energy sought? In our case is the shaft work of our Lister(oid). How do we measure it? (a) with the aid of a dynamometer or (b) the output of our gen head.
Not so fast there, the dynamometer has some frictional losses that may be (probably are) invisable to the instrumentation, better account for it and correct the readings. If we use the electric power from the generator head as an indicator of engine work output the measured power is the product of the engine efficiency, belt drive efficiency and generator efficiency. Hard to determine the engine work output alone without knowledge of the drive and generator efficiencies.
If we choose to just lump them together and call it the overall genset efficiency that is fine if you are choosing to try to improve efficiency of your own genset. But bear in mind if you compare measurements of your system to that of some one else there is a measurable uncertaintly involved with the comparison. How do you know the both sets have the same drive and generator efficiencies. Casual comparisons are fine as long as everyone is aware of and takes into consideration thease uncertainties.
Assume you have a reliable method to measure the engine shaft work, lets take a look at the measurement of the heat of combustion. Simple right? just go to a handbook and look up the heating value of petro diesel. Wait a minute...light diesel, heavy diesel and whats this higher heating value/ lower heating value stuff? Well, being pump diesel its probably light diesel. Heavy diesel is probably that thick stuff that must be heated up before being pumped into a large diesel.Thats my assumption. Lower heating value assumes complete combustion ( no HC, CO or NOx in exhaust, just H2O and CO2) and the water in the exhaust stream is in the vapor phase at temperature of 25 deg. C or 77deg F and a pressure of one atmosphere. Higher heating value employs the same assumptions, but the water is in liquid phase. Ok, so we are using the lower heating value of light diesel fuel, 43,200kJ/Kg or 18,600BTU/lbm by my reference. Done, right? Nope. Real engines have incomplete combustion, for best accuracy we need an analysis of the exhaust products and the temperature of the exhaust stream. When we know this, the heats of formation for each chemical specie of the exhaust products can be found and the heat releassed during combustion determined.
My point in relaying this information is this: when we use terms like "thermal efficiency" in casual conversation ( watts produced divided by lower heating values of diesel) we must bear in mind that there is a lot of overhead involved with obtaining an accurate measure of this term. The value that we might calculate for thermal efficiency using the simplifying assumptions stated above may well still be 85-90% accurate, I don't know as I have never looked into it.
GF speaks of not all BTUs being equal. I can't say that I understand fully the point that he is trying to make, sometimes I am a bit dense. What I do know and perhaps this is part of what he is saying is this; sources of energy posess a quality. We all know that energy can be converted from one form to another. Mechanical work and electrical work are two high quality forms of energy. What makes them high quality is that they are readily converted to other forms of energy with minimal losses. For example electrical energy can be converted into mechaniclal work with a better than 90% conversion effieiency with an electric motor (for large motors, smaller one are less). Thermal energy is a lower quality form of energy. For example, converting thermal energy into mechanical work using a steam power plant (steam turbine) has about a 40% conversion efficiency. And that is for a relatively high efficiency plant. Temperature of the thermal energy source plays a role too. It is easier to convert thermal energy to another form if the thermal energy exists at a high temperature rather than a low temperature.