CHT probe styles and installation will vary depending on your engine type and installation. On my Lycoming 320, for example, I bought these “Bayonet-type” probes that come with a screw-in adapter. You take a big flathead screwdriver, install the adapter, and then the probe is spring-loaded and installs like a BNC coax connector. Pretty straightforward, so what’s the issue?
When this Type J thermocouple probe produces a differential voltage and your engine monitor converts this voltage to a temperature readout, what does it really mean? Does it mean the hottest piece of aluminum on the cylinder is this temp? Does it mean the combustion chamber or exhaust valve guide material is this temp? None of the above: It only means that the location of that probe is that temperature. Let me repeat that … it only means that the location of that probe is that temperature.
This seems important. Lycoming gives us recommended temperature ranges for that exact measurement location, and if you have a Lycoming, you are likely targeting somewhere between 380° F or 400° F. These temperature guidelines correspond to many years of testing and experience. It is kind of like if you wanted to measure your car’s engine temp by sticking a probe on the muffler. You can do it, but you’ll have to collect a bunch of data to determine the correlation between 150° F at the muffler to your boiling coolant in the engine.
You’re thinking, “Well, of course. What’s the big deal?” The big deal happens when you use a non-traditional engine with very little data or history behind it. I have a Revmaster VW conversion in a plane, and even with many years of existing, the CHT guidelines are loosely defined at best. With a digital engine monitor, I’m able to collect some good data on the subject.
The largest discrepancy on a VW comes from where you install the CHT probes. There are three different methods I am aware of on a VW: 1) ring probe under the spark plug, 2) ring probe under the cylinder head mounting nut, and 3) welded junction with set screw attachment.
To be clear, Revmaster does give a recommended temperature range for Method 3 installation. Unfortunately, this “acceptable” temperature range has shown to send many cylinder heads to the scrap pile due to excess valve recession and wear. As a result, many in the community have ditched the set screw probe installation and opt for the spark plug ring probe instead. When you query the community for data on the subject, it is common to get a mixed bag, and you’re left more confused than when you started.
My solution was to install four set screw probes per Revmaster recommendation and an additional spark plug ring probe for comparison. The results were staggering.
The difference between the probe data varies by up to 50° F on the same cylinder for the two probe types! To put this in perspective, if you only had the ring probe on the rear cylinder spark plug, you would think you’re cruising at 370° F CHT. Meanwhile, with the set screw location on the same flight, you’re showing 420° F. So which one is right? Both are! But you don’t want to fool yourself by changing your instrumentation. I’ve found that the set screw probes should stay below 380° F for best longevity on a VW, but this would equate to a 330° F limit if using the spark plug probe!
It is always important to qualify your measurement data and not assume it is absolute in any way. When reporting data, you should always highlight the details of the specific installation. The actual values of your CHT readings are not as important as where they fall in the acceptable range for that scenario.
This is especially important information for non-primary engine monitors where the aircraft came with one bayonet probe from the factory and must be retained. The engine monitor installation typically uses a ring/gasket style probe under a spark plug or under the bayonet probe and those often read 20-40° cooler than the bayonet type. When installed under a spark plug, the variation depends on installation on the top or bottom plug with top installation reading cooler than bottom.
I have always wondered what to do about the spark plug gasket with the under the plug temp sensors.
That’s a good point Don. Assuming you use the gasket on top of the probe ring, you now must rely on the CHT probe ring to seal to the cylinder head. The ones i have seen are copper and can make this seal, but will get worn out over time removing and replacing spark plugs.
I think you need to mention a very important aspect of using thermocouples in temperature measurements. For accurate data, you need to make sure the display instrument is cold-junction compensated. Otherwise, readings are going to significantly affected by ambient temperature. You’ll find that some popular but not very expensive EGT/CHT are uncompensated. You also need to make sure that any extension wire is also thermocouple wire. Cheaper instrumentation uses copper extension wire.The hotter “readings” under the plug is quite well known and really makes no difference as long as you compensate for it in the published operating limitations.
Hello and thank you for the comment. While this article really just focused on the installation location, you are absolutely right about the other details of thermocouple instrumentation. I am always surprised to see things like copper extension wire and “connectors” with set screw attachments. I do want to point out for consistency that on my engine, the spark plug ring probes read colder, not hotter, so it very much depends on the specific application.
That’s very interesting. I’ve never seen “colder” readings under the spark before as a matter of normal operation. I mean the “sensor” is on a thin part of the cylinder head and closest to the combustion chamber. Did you verify accuracy by, say, immersing both probes in boiling water and verifying their output? I’ve only seen spark plug probes read low because of poor contact (the copper gasket was not annealed and made poor thermal contact) or the crimp location of the thermocouple junction was poor. Even an Amazon IR temperature probe could help here. I guess it’s possible, but…I’d be suspicious without more data.
Both sensors in this case were verified to be accurate outside of the plane. The reason for the difference in readings is mainly the proximity to the exhaust port on the cylinder head. The spark plugs on this engine are 3/4″ reach which is a good estimate of the casting thickness.