Mixture Management and Cylinder Head Temperatures
written for MN Wing Newsletter Dec. 2007
When most of us were taught to fly, our instructor told us to use “full rich mixture on takeoff, level in cruise, lean until it gets a little rough, then turn it half a turn in”. This seems simple enough, especially when one is renting the plane wet. However, with the RV’s, there are reasons other than economy alone to look a little closer at the mixture. Specifically, cylinder head temperature is dramatically affected by mixture. This article will hopefully help us to understand a bit more about mixture and its affect on our engines.
Let’s look first at the basic chemistry involved, namely combining fuel with oxygen. If we have just the right mixture of gasoline and oxygen (which is 21% of our air) and we burn them, there will be no fuel left over and no oxygen left over. This mixture is called stoichiometric. With any mixture richer than this, there will be extra fuel left over, while leaner will mean extra oxygen left after combustion. In piston engines, at the stoichiometric ratio, exhaust gas temperature will be near a maximum (peak egt).
So, where do want the mixture to be? Rich of peak (ROP)? Lean of peak (LOP)? It depends on a number of variables, including manifold pressure, fuel octane, RPM, compression ratio, ignition timing and intensity, and other variables. For full rated power, we need to run our engines well rich of peak, indeed, a lot rich of peak. This excess fuel does a couple things: first, it modifies the combustion process to reduce or eliminate detonation. Additionally, it reduces the amount of heat transferred to the cylinders, keeping the cylinders cooler. Running too lean at full or high powers can damage our engines, typically through detonation and overheating. A 180hp Lycoming type engine will burn about 15 to 17 gph at sea level, full throttle and mixture full rich. Anything less than that can be problematic. Lycoming recommends no leaning until at or below 75% power.
Before we continue on with less than full rich operations, we should have a look at how the fuel gets to the cylinders. The throttle’s job is simple – to control the amount of air entering the engine. The carburetor or FI servo’s job is to add fuel in some ratio to the amount of air. The mixture control simply tells the carb or FI how to ratio the fuel/air.
In carbureted engines, things are a little complicated downstream of the carb. As the air travels through the carb, fuel is metered into the air stream. Some of this fuel evaporates, but not all. So, downstream of the carb there is a mixture of air, vaporized fuel and liquid fuel droplets. As this mixture travels through the induction system, it has to make turns on the way to the cylinders. Unfortunately, due to the geometry of the induction system, the droplet portion of the fuel/air mixture does not remain evenly distributed on the way to the cylinders. The end result of this is that the fuel/air ratio is not the same cylinder to cylinder. So, what does this mean? What it primarily means is that power and waste heat will not be equal between cylinders. This can cause roughness, since cylinder 1 might be putting out 15% more power than the other cylinders, for example. Additionally, and very importantly for RV’s, it can cause large ranges in cylinder head temperatures from cylinder to cylinder. Hottest to coolest cylinders in RV’s can commonly range from 400F to the low 300’sF, in the same engine! For engines with FI, the distribution of mixtures between cylinders can be almost equal, since fuel is sprayed directly upstream of the intake valve, and does not have to travel the same path as the induction air.
When we burn fuel in our engines, almost all of the energy created goes to one of three places. One of these is creating work at the propeller, which is of course the point of the whole thing. Secondly, heat energy is transferred to the cylinders, and thirdly, heat energy is discharged out of the tail pipes. Unfortunately, the waste heat of #’s 2 and 3 comprises the majority of the energy expended. Only about 20% of the combustion energy actually gets to the propeller! To get a sense for the amount of heat that must be removed, think about pouring a pint of gasoline per minute on a bon fire and you get an idea of just how much burning is going on up front!
So, how does mixture affect cht’s? Here is Lycoming’s chart showing, in a general sense, how mixture, egt, cht and power are related:
Referring to Figure 1, if we look at the curve labeled “cylinder head temperature”, we can see that it peaks around the same point that exhaust temperature peaks. Notable is how rapidly the cht curve drops off to the left (lean) of peak, compared to it dropping relatively slowly on the right (rich) side of peak. It indicates that the cht can be lowered by about 35 Celsius degrees, or about 63 Fahrenheit degrees, just by going to about 80 degrees F lean of peak with the egt! However, there are several classic problems with running lean of peak (LOP). Recalling what we discussed earlier about uneven mixture distribution, if we have cylinder #1 running nicely at this 80F LOP point, cylinder #3, for example, might be right at its peak egt, and hence, close to peak cht temperature. So, cylinder #1 might be at 320F, while #3 is at 380F. With a carbureted engine, it is not uncommon for one cylinder to be running at peak power while another is in the best economy region! Think of your four or six cylinder engine as actually four or six separate engines, each running with the mixture knob at a different setting. Another article is planned in the near future on the art of persuading a carbureted engine to run smoothly LOP, so stay tuned...
Many of us chase this elusive cylinder head temperature problem by mucking around with baffling and other hardware. An additional problem with this situation is the uneven power output cylinder to cylinder mentioned earlier, which we will now discuss.
Referring again to Figure 1, look at the curve labeled percent power. Just to the left of those words are numbers, starting at 100 and counting down the left side to 80. Let’s not think of this as the classic percent power (100% = 180hp, for example). Rather, think of that as percent of maximum power available at a particular manifold pressure (MAP) and rpm. Say that we are running full throttle at about 8500’. This roughly corresponds to about 75% power (in the classic sense). In our 180 hp engine example, this is 135 hp. Looking at that curve, and assuming that one cylinder is in the peak power portion while another is in the best economy area, we can see that the lean cylinder might be only putting out 80% as much power as the one operating in the peak power range. This uneven power distribution cylinder to cylinder will cause roughness. For sure, this drop in power as we go LOP will show up in reduced indicated airspeed, which we may choose to do, provided the engine is running smoothly, for economy purposes.
The final curve we will discuss on Figure 1 is the one at the bottom, labeled “specific fuel consumption”. This curve tells us that when we operate in the shaded area labeled “best economy range”, we get the most power per gallon of gasoline consumed. It says that if we operate in that range, the engine is burning the gas in the most efficient way possible with respect to power output. Do not confuse this with the most efficient airspeed to fly, as that is a different topic.
Additionally, engine roughness can result when one cylinder is so lean of peak that it simply misfires occasionally. Many of us will attempt to run LOP only to find that misfire occurs in one cylinder while another of the cylinders is running too hot (sitting right at peak). Indeed, the very description “running my engine LOP” is misleading if the mixtures aren’t balanced. We should say “running engine #1 LOP”, when one thinks of each cylinder as a separate engine as described above.
Another challenge with running LOP is to get the fire started. At low MAP’s and lean mixtures, a hotter spark is needed. It is also helpful to advance the spark more than the typical fixed 25 degrees (that mags provide) in this situation. Both of these problems are solved with most of the electronic ignitions available for use on experimental aircraft.
So, how well does the ancient Lycoming chart in Figure 1 hold up in the real world? Amazingly well, it turns out. This data is from an RV6A, taken about a year ago. Note that in the following charts, the engine is fuel injected, and reasonably well balanced with respect to mixture distribution among cylinders (i.e., running “one” engine instead of four separate engines).
Figure 2 shows the relationship of mixture (fuel flow) to cylinder head temperature. Note that the cylinder to cylinder variation in temperature is fairly constant over the entire mixture range. This probably means that the differences are indeed due, in this case, to baffling inconsistencies or cowl airflow differences. Note also that by going from peak egt to about 60 to 80 LOP, the cht’s dropped about 60 to 70 F degrees! This is the main point of this entire article!
It is an interesting side note that cylinder #3’s temperature was brought into line with the others (after this data was taken) by putting a spacer behind #3. The range of hottest to coolest cylinders on this engine is now typically about 25 F degrees. It should also be again recalled that this engine is fuel injected, so egt’s tend to peak at about the same total fuel flow, as can be seen in Figures 2a and 3.
Figure 3 shows how power drops as mixture goes LOP. This graph shows the power drop as mixture goes leaner by graphing IAS (power) vs mixture (fuel flow).
Figure 4 shows how the maximum fuel economy, shown here as indicated n.m. per gallon, is at a maximum when running LOP, for a given MAP and rpm. While we know that we can increase fuel economy by pulling the throttle and flying much slower, in this case we are attempting to keep as much speed and power as possible by getting the engine to convert avgas to mechanical energy in the most efficient manner. As a bonus, the engine will also run cooler! In other words, the miles per gallon could be much higher just by flying slower, but that is not what is being illustrated here. This correlates very nicely with the Lycoming chart in Figure 1, specifically the specific fuel consumption curve.
The RV6A from which the above data was collected regularly flies along side other RV’s that are burning 1 to 1.5 gallons per hour more fuel! This is $4 to $6 per hour at today’s fuel costs. The total savings over a couple thousand hours can pay for an engine overhaul.
Update June 8, 2008:
This chart is from an RV7A, O360 with a carburetor. It can be seen that the #1 cylinder is quite a bit leaner than the other three, and causes the leaning process to be halted (due to its, cylinder 1's, lean roughness) while the other three are right in the "hot zone". If the leaning process were to be continued (to the left side, less gph), the other three cylinders would cool down into the 300F range also. Indeed, #3 is on the way already, having dropped about 20 degrees already.
Here is the same run, only this chart shows the egt data. #1 can be seen to have peaked around 9.6 or so gph, while #4 hasn't even peaked at 8.3 gph. There are indeed four different "engines" being run here.