Installation of a coolant thermostat in my Rotax 912S
As for so many Rotax 912 owners, I’ve been concerned by not being able to reach recommended operating temperatures before applying take-off power. Ground running at 2500rpm for long periods on cold mornings shows not much increase in temp at all. Need a much higher power setting to generate the heat necessary, and can’t do that on the ground. I really don’t like blasting off into a full power climb with the heads still at low temp, and watching those temps rise quickly with all the heat generated from that full power operation.
Duct tape over the coolant radiator does work to increase head temp sooner, but it’s a hassle to apply, and a worry when it’s still there on a warmer day and the temps are rising too high.....
I looked at the possibility of variable shutters, but not easy in the confined space. I know of someone who did build a tidy multi-blade shutter arrangement, but he found that it didn’t really do the job of regulating the temp because as soon as it was opened a little bit, the cooling effect was almost the same as full open. So it was an on-off situation.... To be effective, it would probably have to be a sliding shutter that tightly covered more or less of the radiator area, but I couldn’t see how to do that dependably in the space. And even then it would still have to be pilot monitored, which gets busy with an aircraft engine that’s run full power for an extended climb and then at idle for a long descent.
Joe Spencer in
I had a look at the following information, and was impressed with the development and testing that Bill ‘Watt-Man’ had done on the Thermo-Bob.
(If the link doesn't work for you, copy and paste it into a New Tab)
I’m really impressed that all his claims are backed by real life test results published for all to examine. He’s done a lot of work on this. I really like the idea of the bypass that keeps a significant flow through the engine at all times, rather than a conventional thermostat that completely blocks the coolant flow at times.
There are photos of the unit at
and instructions for installing that show good detail of the construction of the T’Bob in figures 11 through 18 at:
That housing is machined and hard-anodized aluminum, and the halves are held together by six stainless machine screws and sealed with an O-ring. Real ‘aircraft quality’ already. The actual thermostat inside the housing is for a 4-cylinder car, so is big enough for the Rotax.
So I pretended to own a 2006 KLR, and ordered a Thermo-Bob to give it a try.
The inlet and outlet fittings on the T’Bob for the KLR are 18mm (3/4”), while the hoses on the Rotax are 25mm. But I found that a couple of stubs of ¾” heater hose sleeved it up for a tight fit. Rubber to rubber when cleaned with thinners makes a really good bond, and the tang on the T’Bob fitting locked the hose clamps on securely. The smaller inlet/outlet size will probably restrict flow a little bit, but my engine has always run cool, even in hot conditions, so I think it will be adequate. Won’t really know that until next summer in the southern latitudes....
Installation was easy. The T’Bob goes in the line down from the expansion chamber to the inlet of the radiator. I had to cut 40mm (1.5”) off the aluminum pipe that curves down to the radiator. The ‘T’ for the bypass goes in the line from the radiator outlet to the water pump. I had to cut 20mm (3/4”) from the aluminum pipe that comes back from the radiator. All in all, it’s a very compact and tidy installation that fits in the space easily, which isn’t always easy with all the hoses around a Rotax....
This is the installation on my Rotax 912S.
My testing showed that at outside air temperature of 12ºC (54ºF), head temp was 50ºC (122ºF) at 2 minutes and 80ºC (176ºF) at 4 minutes. Before the thermostat, I hadn’t been able to get to 80ºC at all while on the ground. This feels much better to me for the full power take-off than at the much lower temps before. Once in flight the head temps locked in at 91ºC (195ºF) on the right side, and 87ºC (189ºF) on the left side, and stayed there, regardless of full power climb or idle descent.
Oil temp was not as much changed as I would have expected. It still took 6 minutes to get to 50ºC (122ºF) at warm-up. 6 minutes seems like a long time when you’re just sitting there ground running, and the prop is sucking up any debris off the ground. At cruise, oil temp stays pretty much at 90ºC (194ºF), but still varies higher or lower for climb or descent.
I had hoped to get quicker warm-up and more steady oil temps, but not so....
A bit more study shows why this is so.
I tried taping over the oil cooler radiator completely, and it made no difference to warm-up time.
It still took 6 minutes to warm to 50ºC (122ºF) even though the head temps were up at 90ºC (194ºF)....
This seemed puzzling until I realized that because these heads are water-cooled, the oil flow through the heads is only enough for lubrication and is not the high oil flow rate used for cooling inside the heads as would be in an air-cooled engine. So this low flow rate doesn’t heat the oil very quickly. Then consider that the oil goes to the large 3 litre metal holding tank, which also acts like a very effective oil cooler. After running the engine for a few minutes, touching the oil tank showed it to be warm at the top, but cool at the bottom, from where the oil is drawn to feed the engine.
I had previously tried two brands of oil thermostat, Permacool and Thermostasis. Only minor improvement in flight regulation, and no improvement in warm-up time. Not worth all those extra hoses and hose clamps full of hot oil....
I can’t see any way to hurry the oil warm-up with the size and exposure of that oil tank/cooler. Can’t put a thermostat in the line from the sump to the tank in order to bypass the tank at low temps, because the flow at that point is irregular as the scavenging system ‘burps’ both air and oil. A more effective oil thermostat in the line from the tank to the oil pump could stabilize in-flight oil temps, but won’t be able to significantly shorten warm-up time, due to the cooling effect of that storage tank.
I’m starting to realize that we may be getting a bit too fussy about oil temp just because we can monitor it. If we had oil temp gauges in cars, I bet they would never be over 50ºC (122ºF) before we drive away. The Rotax manual does call for 50ºC (122ºF), but it’s very impractical to get to it, and is it really necessary?? Certainly the oil does have to be warm and circulating well, and 3-4 minutes ground running now does that, and is enough to allow heat soak to stabilize. The AeroShell plus 4 oil that I use is rated as 10W40, so that viscosity should allow good circulation and lubrication even at just warm temps. So I’m now satisfied to go at 4 minutes, when the head temp is well up and the oil temp is just starting to climb on the gauge.
The T’Bob makes that possible, and so I consider it to be really worthwhile. It fits in the system really well, and feels like it belongs there. I’ll certainly be keeping mine.
The only possible downside I can see would be if the thermostat wouldn’t open. The bypass would still allow some flow through the engine, but none through the radiator, so the temps would rise. I don’t know how high they would go, but have heard that the Rotax can be run at low power with no coolant at all. I can’t verify that, but it’s very likely, since only the heads are water-cooled, and the cylinders are already air-cooled, and the circulating oil does still give some cooling inside the heads.
Thermo-Bob suitable for the Rotax
I’ve informed Watt-Man about these results, and suggested that he might consider manufacturing a T’Bob with 25mm fittings that could be used for the Rotax. He’s interested, and re-tooling for that right now, but it would be sold with a ‘not for aircraft’ sticker on the box. (That label will be for liability reasons, as we are all aware.) It will have 25mm fittings that will suit the Rotax hoses, and the greater internal flow to suit. It will be labelled as a ‘Thermo-Bob 1’ model, and will cost a bit more due to the re-tooling and limited production, US$159 I’m told. Still good value for a really effective top quality product, I reckon.
Update May 2013 - I'm now told that they have been in successful production since then, and the price has come down to US$139.
Having seen my results, Joe Spencer (701) has ordered the very first one, and friend Hans (also 701) has ordered the second one. I realize that if you’re in the northern latitudes, you’re more interested in flying rather than tinkering right now. But if you’re going to buy one sometime anyhow, to install next winter, how about ordering it right now to encourage this small business and help cover his re-tooling costs. I’ve told him there are a lot of Rotax owners out there with this same cooling problem, and I think it’s true, judging by all the chatter on the forums. I’m wishing him well in the venture.
You can contact Watt-man directly at firstname.lastname@example.org .
I do have to stress that I have no financial interest in this business at all.
I just like to give assistance and encouragement, and help spread the benefits to all.