Leaning The Jabiru 3300

For the past year I’ve been reading and following Mike Busch’s engine articles, and generally consuming as much of his advice as I could get. Mike recently launched a new website SavvyAnalysis.com, and this site has some fantastic features. The site allows you to upload engine monitor data straight from your engine monitor (I use the Dynon Skyview system and can export my data logs including engine data to a USB drive). The website programming takes care of the magic and allows you to look at each flight’s engine parameters (CHT, EGT, RPM, Fuel Flow, etc) graphed along a timeline. What a fantastic tool! Go explore the site and you’ll see what I mean…

One area that benefits greatly from this style of graphical data display is the ability to easily spot trends and relationships between the various sensors. For the past 3 years I have been experimenting with leaning in cruise flight, trying to determine how lean to run my engine. It’s a classic dilemma: how to find the sweet-spot where the engine runs smooth and clean internally, but where temps are under control and stress on the heads and valves is acceptable? I had settled in to a typical cruise of 2900 rpm at 4000-8000 ft DA leaned to approx. 6.5 gph. This would yield approx. 65% power, and EGT’s in the 1350-1380 range on my hottest cylinders. I figured this was conservative, and wanted to see if there was a better way.

I recently collected some in-flight test data while leaning slowly at cruise. The task was to fly as consistently as possible, and lean slowly and smoothly to collect temperature data for later analysis. This resulted in a 2900 rpms cruise at 8000 ft DA, initially burning 7.2 gph, leaning slowly until the onset of engine roughness, and then richening slowly back close to the starting fuel flow. The trick in my plane is to make small adjustments of the mixture knob – this is difficult because the button-lock mixture cable I installed lacks the fine control needed. Each adjustment was generally too course, but careful practice made the best of the situation. I’ve selected a few of the best results for this analysis, and a few other test runs (test 1, test 2) were also reviewed for consistency.

Mixture Sweep #1.

Looking at the graphic of Mixture Sweep 1, below, you’ll notice that I’ve zoomed in on just the few minutes of flight of the actual leaning process, and highlighted EGTs (various colors for each cylinder) and Fuel Flow (thick blue line).

The data shows that as the mixture is leaned, fuel flow is reduced (somewhat erratically, again due to my button-lock control), and EGT responds by increasing. The interesting part starts occurring at 00:32, as EGTs 2, 1, 4, 3 all peak at about the same time. Although you can’t see from just this snapshot, EGT #2 peaks first at 00:32:14 at 1475 deg F at 5.1 gph, and #3 (the last of the 1st 4 cyl’s) peaks 15 seconds later at 1170 F at 4.9 gph. I’m not pleased about the spread in EGT’s, mainly because I don’t know if there’s a problem with this spread, or it’s simply the way it is….

So the 1st take away is that the front 4 cylinders all peak very close to each other, with only about 15 seconds and 0.2 gph between them. This seems pretty good. However, take away 2 is that the rear cylinders (#5 and #6, green and blue lines, respectively) will not have peaked yet at that mixture setting. They won’t peak until a fuel flow of 4.7 gph after an additional 3 mixture adjustments over the next 60 seconds. By the time 5 and 6 peak, 1-4 will be running significantly lean of peak and have dropped by 50 deg F, and the engine experienced a 100 rpm reduction in power. By this point the engine will likely be running rough, with fluctuating rpm (see chart below, rpm now shown as the thick blue line). Further leaning at this point will feel like a bad idea!

Take away 3 is that if you run 50 deg F lean of peak on the front 4 cyl’s, then the rear 2 cly’s will be operating right at peak, at the point of highest stress, and the engine will be grumbling with rpm surges and roughness. This doesn’t seem like a place to operate for long periods in cruise. Looking back at the data, we can identify a place where the engine stress is lessened (50 deg F rich of peak). This occurs at 5.7 gph. If we back off the EGT even more to be extra conservative, we aim for 100 deg F rich of peak and that works out to 6.5 gph – right where I normally try to fly!

So it looks like the options are to 1) fly conservative at 6.5 gph and 100 deg F rich of peak, 2) lean a bit more to 50 deg F rich of peak at 5.7 gph and accept the higher stress on the engine, or 3) continue to lean to just before the onset of roughness at 4.8 gph, and get lean of peak on 4 cyls and near-peak EGT’s on the rear cly’s. The big unknown is how much stress option 2 or 3 places on the engine?

Somewhere there is an economic case to be made for each option. For discussion sake, let’s assume that option 1 results in baseline performance for 1000 hrs (the recommended top overhaul on the Jabiru). Burning an additional 0.8 or 1.7 gph over options 2 or 3 (respectively) results in an additional 800 or 1700 gal of fuel over the life of the overhaul period. Assuming $2.50 per gal of auto fuel, this works out to spending an extra $2000 to $4250 in “wasted” fuel. Saving $2000-$4000 in fuel by leaning more is attractive, but not if doing so places additional stress on the engine and requires additional parts or maintenance. One or two bad cylinder heads will cost this much to repair, not to mention the cost if it happens in flight and trashes more than just that cylinder. Additionally, there is a benefit to flying longer legs made possible by a lower fuel burn. Option 2 gives you 20 minutes greater duration [in case you want to run the math: time in hrs = 16 gal * [ (1/5.7) – (1/6.5) ] at the same speed (probably), and option 3 gives you 45 minutes at a slightly slower speed (again, probably slightly slower due to the lower rpm when lean of peak). Each person will have to decide how much this extra 20-45 minutes is worth.


After considering all the data and the 3 options, I guess the decision comes down to the effect of the extra leaning on engine stress. If the engine can handle the extra stress of hotter EGT’s, when running close to or lean of peak, then saving fuel becomes an attractive reason to lean. However, if the increased stress takes its toll on the engine and increases the cost of maintenance or the risk of premature failure, then that’s a hard sell to lean the engine aggressively.


NOTE: If you do not see a menu frame on the left, click here to reload the full page.
Updated: 30 Jan 17