Home | Rick Koch: My engine temperature (oil temperature), when as warm as it ever seems to get, BARELY touches the green. It’s always right at the yellow/green division. The previous owner (for many years) said it always stays at that point, so at leas

Rick Koch: My engine temperature (oil temperature), when as warm as it ever seems to get, BARELY touches the green. It’s always right at the yellow/green division. The previous owner (for many years) said it always stays at that point, so at leas

Rick Koch:

My engine temperature (oil temperature), when as warm as it ever seems to get, BARELY touches the green. It’s always right at the yellow/green division. The previous owner (for many years) said it always stays at that point, so at least I’m seeing a “normal” indication for this engine (it’s a O-360, 670 SFOH). But shouldn’t it get “deeper” into the green? What could be reasons for never getting or registering “hotter”? I’m also not too sure that the fuel gauges are reading correctly.


In most cases, low engine temp indication is the result of an out-of- tolerance panel gauge. The sending unit is a Mil-spec part, at least for the Rochester gauge systems, and rarely goes bad. Poor connections have occasionally been found to be the problem, but much more often it is the gauge. A “rough” but very easy check is to simply go to the plane on a warm day (well above 60 degrees F). See what the gauge reads, on the cold engine, with the Master switch on (and with a fully charged battery). If the gauge reads well below the ambient temperature, or the needle remains pegged on “cold”, you know that it really is an instrumentation problem and not an over-cooling problem.

In cases where you really do have low oil temp (anything sustained much under 175 degrees), as confirmed (for example) by a probe connected to a JPI EDM-700/800, you nearly always will find a bad Vernatherm (oil thermostatic control) valve. They fail to fully extend when hot, and give high oil temp; or they bind and fail to fully retract, giving low oil temp. In extreme cases the wax pellet can migrate around the plunger inside, and it won’t retract at all.
There is a Lycoming test spec in one of their SB’s for testing the valve in hot oil. I don’t have access to it where I am at the moment. If it does not move at the specified temps, or does not have the minimum specified travel, or does not move smoothly straight in and out as it moves, it is no good. BTW, I would never buy a used one; it is one of those parts, like an auto thermostat, that weakens as it ages.

Assuming your gauge has the little “normal point dot” in the green range, that’s where you should see the temp needle at 65%-75% cruise, almost regardless of ambient temp (within 1/16″ of either side of the dot).

Editor’s addendum from another post response:

Before doing much of anything else, you have to know who manufactured your gauges/cluster unit. Sometimes if you look very closely at the face of the cluster, you can see a brand name in very small print, somewhere on the front of one of the gauges. Failing that, there may be a label on the back. Failing that, if you can find the part number printed on the cluster housing somewhere, I may be able to identify the manufacturer based on the PN.

Most (but not all) of the C-model Sierras used Rochester gauges, in
conjunction with a MS28034-1 sensing bulb (sender). The
specifications for the bulb are available at:

Part description: http://www.norwichaero.com/rtd3.html

Specs table: http://www.norwichaero.com/rtd8.html

As you can see from the table, 100 ohms is a valid testing value, and
would correspond to an ambient temperature of almost 90 degrees F.
I’m not sure how you are connecting the wiring for testing; the needle should not have pegged at 100 ohms, or at any value below about 150 ohms. However, it probably doesn’t matter much. I’ve never even heard of a bad MS28034-1 sensing bulb, probably because it is made to a Mil-Spec, but nearly every old gauge cluster unit I have seen has contained one or more gauges that were far out of spec, or outright bad. One even had a piece of debris in it from the original manufacturing (30 years ago), that would periodically cause the needle to stop at mid-range. Can’t help but wonder how many people lived with that over the years.

The 12-14V gauge PN is 169-380036-29, and the 24-28V gauge is 169- 380036-45.

RAPID shows the dash 29 for $319, but it doesn’t matter. That was what they charged for the last one they sold, and they have not had any for years. If they ordered more they’d probably want to charge $1,000 each for them; they know that no one will pay that, so they won’t get any more. You can get the gauge units overhauled by Air Parts Of Lockhaven, 800-443-3117. They were based in Lock Haven, PA, with a branch at Vero Beach, FL, last time I dealt with them. They were charging $113 per instrument for overhauls. If you have all three of the electrical sender-based gauges in your cluster done (two fuel gauges and oil temp), it would be three times $113. The ammeter and mechanical oil pressure gauges seldom go bad, as they have basically mechanical rather than electrical-bridge internal calibration. If your A&P wants to fiddle with the two calibration screws (as he will be tempted to do if he pulls it out), he might get it to read about right with the test resistance, but the range will not be correct; the gauge will not be reliable above and below that point. There is also most likely an out-of-tolerance component on the circuit board, which he will be “covering up” by changing the adjustment screws. Not to mention that he isn’t supposed to recalibrate gauges unless he has the correct equipment and is approved to do it.

Unless your fuel gauges are reading spot-on (they say “full” when you fill the tanks to the top of the slots, and they hit the E redline when they are down to a couple of gallons), I would pull the cluster and send in the entire unit. Just tell them that you only want the two fuel gauges and the oil temp gauge overhauled. They should be willing to handle the disassembly and reassembly of the cluster as part of the deal, and they might even freshen up the appearance of the other two gauges just for the heck of it. Unless your A&P is familiar with the disassembly of the cluster, you’d be better off letting the overhauling shop do it. It isn’t really difficult, but the gauges interlock, and have to be disassembled in a certain order. Someone who doesn’t understand how it’s put together could do a lot of expensive damage while figuring it out, unless they are the patient type.

And one more addendum from the Editor, to get all this in one place:

The engine’s operating oil temperature needs to be at least 165 degrees, and a normal minimum of 185 degrees is preferred. At 185 degrees, about a half-hour’s operation is supposed to be enough to drive any accumulated moisture out of the oil. If the moisture stays in the oil, the oil’s acid neutralizer and corrosion inhibitor additives are rapidly exhausted. This is why ground operation only (with no associated flight activity) is strongly discouraged.

Lycoming’s Vernatherm (thermostatic bypass valve) is designed to start opening at 185 degrees. It is typically mounted in the pressure screen housing or the oil filter adapter housing. Much like the water thermostat in a typical car engine, this valve enables colder, thicker oil to bypass the oil cooler (by leaving open a passage/bypass hole into the engine’s main oil gallery). However, because of the way it works, by plugging the bypass hole as it expands, it is always possible for some amount of oil to circulate through the cooler. The Vernatherm does not block off the cooler passage when the oil is cold; it just makes it all go through the cooler, if the oil temp makes it up to 200 degrees or so. In contrast, an auto thermostat lets almost no water make it to the radiator until the minimum temperature is reached. Because of this operational difference, it can be helpful to block airflow through some or all of the cooler’s fin area, OAT and oil temperature experience permitting, during the winter months. If you are seeing CHT’s of 360 degrees or more, I would advise you to not block any of the cylinder head cooling airflow.

The Vernatherm valves are finicky. Half of the ones in service are probably out of tolerance. They are on the Lycoming mandatory parts replacement list at overhaul. Nonetheless, I have read dozens of overhaul work orders that did not show this valve on the list of replaced parts. In fact, it is almost uncommon to see a new one. There is a Lycoming procedure for testing the valve. I have yet to test a used valve that passed the test for extension, alignment, and retraction at the specified temperatures. They fail to fully extend, they cock sideways during extension or retraction, and they fail to fully retract. New ones easily pass the same test. In most cases a faulty valve will fail to extend at the required temperature, or to the required length. Most often this just results in an abnormally high oil temperature, since too much of the oil goes through the bypass hole, rather than being routed through the cooler. However, if the valve is cocking during extension or retraction, or if it is failing to retract due to binding, it could be forcing oil through the cooler when it should be bypassing the cooler.

JPI (EDM700 manufacturer) recommends that their oil temperature probe be mounted at the front right side of the case, in the oil gallery (the plug must be removed). This is intended to give a valid reading of the oil temp as it exists inside the operating area of the engine. I have found that this location results in a virtually perfect match of the readings shown by a properly-calibrated factory gauge and the JPI instrument. If you are seeing correct temperatures from this sensor location on a JPI, the moisture should be getting driven out of the oil. A sensor mounted at the outlet of the cooler (or in that section of the oil system path) would read too low.

A sensor mounted in the oil sump might or might not read correctly.
While most of the sump oil has drained out of the engine’s operating parts, some of that oil will have picked up more or less heat than the oil coming from other areas. There are also stagnant areas of oil movement in the sump, where part of the oil may sit for a while and transfer heat out through the aluminum sump cover. Mounting the sensor in an area of the case that tends to remain at core operating temperature, and where there is fairly rapid and constant oil circulation, will give a more reliable reading.

Many of the original sensor-driven panel instruments (fuel level, oil
temp) have long since lost their calibration. I have found the original Mil-Spec oil temp sender itself to be highly reliable. If there is a difference between the factory gauge temp reading and the JPI temp reading (with a properly located JPI probe), I would believe the JPI. One crude way to evaluate the factory oil temp gauge is to know what the OAT is, while it is 70 degrees or higher (so the gauge needle should be well off the low-end peg). Before starting the engine, mentally extrapolate where the ambient temperature would fall on your gauge face (the minimum shown is probably 60 degrees). Turn on the Master, and see where the oil temp needle goes. If it fails to move up to the point where it matches the ambient temp, your panel gauge is probably not showing a correct oil temp reading.

You most likely will not be able to get a valid corroborating temperature test (using some other instrument) during a ground run- up. The plane has to be in flight at 65%-75% power in order to get proper, adequate, and even heating throughout the engine. I have never been able to get a light aircraft oil temp any higher than 165 degrees with the plane on the ground, despite ambient conditions that caused CHT to start going too high.

Thank you for adding to the resources available for your Fellow BAC Members.