Bad fluctuating Oil pressure on a Rotax 912ULS

[Kyle Communications]

This could be a huge post but I will try to keep it shortish.

 

Recently (last 6 to 12 months looking back now) I have been experiencing bad oil pressure fluctuations. Its hard to actually go back and think when this first started or if it was slow in getting worse to a point that it was stupidly obvious. I started having issues with my Honeywell 4-20ma sensor..thought it was the sensor so ended up getting rid of it and putting in a remote mounted VDO old style analogue sensor. Thought it had solved the issue but the oil pressure still moved around a fair bit on my MGL Xtreme EFIS. This being a digital style input works a certain way and the ADC inside can make things look worse than they are. But over a reasonably short period of time it just got worse. Then alarms started coming up on the limits I had set for the oil pressure but the engine seemed to be working fine. I did some research on it and found a thread on Rotax owners with a fair amount on it and no one seemed to find a permanent solution. I got a analogue gauge and hooked it up to my analogue sensor and was shocked what I saw. You can see it in the video. Further research found a guy in the USA ...Bill hertzel who came up on the Rotax owners thread and said he was working on a solution. Mind you I bought the Rotax upgraded lasted oil pressure relief valve (mushroom style they call it) and installed it thinking this would fix it. It only half fixed the issue my oil pressure was still bouncing around heaps at the bottom end when the oil came up to temp.

 

I got in contact with Bill and he did initial testing with a guy down the road from me here funnily enough...the world is becoming a small place. John Raison worked with Bill and they made some prototypes and got it to a final design that worked a treat. John has more then 100 hrs on this design now and no dramas at all. His fluctuations were far worse than mine too...I mean off the chart stuff. I saw the video of it. The issue is the ball type and the mushroom head style can vibrate around inside the pressure relief valve tube and this causes the problem. Bill came up with a new style for the 912ULS something similar to the one used on the new 912IS. There are some differences though the ULS has a 4 lobe oil pump where the 912IS has a 6 lobe oil pump so the pulses are no where near as hard from the pump. Bills valve will also allow you be be able to adjust your oil pressure as well. Although he recommends keeping the oil pressure around the 30 psi when hot at idle and 45 to 50 psi at cruise rpm.

 

I fitted my new valve on friday and test flew it this morning. Certainly very different now. I am getting some movement but that is because I am running a analogue sensor on a digital instrument. I understand this being in the electronics game. The 4-20ma sensor would be far better to use and I now may go back to one of those as I believe what I thought was a faulty sensor most likely was not. I dont think I threw that $350 sensor out...I hope not

 

The last 2 videos show running with the new valve..one on the analogue and one on the Xtreme once I hooked it back up. Probably wont get a chance to do any more flying as I am off overseas in 9 days for 3 weeks

 

Picture of the mushroom valve from Rotax that doesnt work and Bills new valve and the reason why the mushroom valve doesnt work you can see the layover which bounces it around.

 

If anyone wants Bills email if you having similar issues just contact me and I will forward on his details.

 

 

This is the original issue

 

 

 

[Downunder]

I had the honeywell fail at about 350 hrs. They are known to be crap.

 

Now running the rotax kellor sensor on a skyview for a couple of hundred hours.

 

Some "bouncing" of pressure at low rpm's (nothing extreme) warming up, but smoothing out at higher temps/rpm.

 

Just running the ball in the valve in the ULS.

 

Funny how some engines seem to be ok, even with the ball and others playing up....

 

 

[facthunter]

Fluctuations should be rectified but a relief valve is just that. It tries to limit it when a certain pressure is reached., depending on the spring tension. Thick oil can make it show a higher reading than the lift of setting., when cold if the engine is revved. You can't set for a cruise setting and the idle as well, as the idle will have the valve sealing (closed) so the oil viscosity and the bearings and pump condition will affect that. The flat faced end is unusual. Why not just have it located positively by the fit in the body and still have a slightly coned end ? A sliding plunger always runs the risk of a metal chip or something jamming it. Keep everything scrupulously clean. Nev

 

 

[Kyle Communications]

The ball type and the new replacement by Rotax with the mushroom head which is the one on the left in the picture wobble around inside the oil passage. This causes the fluctuations....this new one pictured on the left the valve sleeve is held in place by the plug skirt and this solves the issue

 

 

[facthunter]

I can see that but a flat face is "unusual" unless the manufacturer can't guarantee the seat being on centre. Since everything is machined at the one setting, it should be easy to achieve... A plunger with ports works fine. These sort of problems are not rocket science. Nev

 

 

[Kyle Communications]

Rotax must have had these issue for a while but of course wont admit it. There are a lot around when you search. The 912iS Oil Pump has 6 lobes producing the 180 pulses per second

 

The ULS Oil Pump has 4 lobes producing only 120 but larger pulses per second at 3600 rpm so the pulse in the 912IS is smaller but their relief valve is very similar although not a trapped sleeve like this one on mine and mine is adjustable internally you can see it in those pics

 

This is the original thread I found about it

 

Severe Jittering Oil Pressure

 

The 912IS valve and inside the oil passage looks like this..very different to the ULS

 

 

This is what mine looks like now

 

 

 

 

 

 

[Yenn]

That must work more like an old style diesel injector pump, should be able to smooth out any jitters.

 

 

[bexrbetter]

Get a dial gage on the end of the crank and measure end float and side play.

 

 

[facthunter]

What's to stop the spacer unscrewing? and why is it needed? the length of the plunger could do what the stop is doing.. Every trochoidal pump has pulses. while it may make some noise why would it make the problem? The original set up probably gets the ball pushed to the side the outlet is on by the flow direction of the oil.. Nev

 

 

[Kyle Communications]

The spacer in the top is a tight thread adjustment on the little allen key grub screw and also loctited with a special loctite. This adjuster allows you to adjust your oil pressure higher or lower if you want. The unit came adjusted to a level that was known to work fine but if you wanted to make it more or less you can. The previous proto units had specific thickness shim spacers in the bottom of the plug to get the right pressure range this all depends of course on the tension of the spring. It took a few goes at getting the right spring tension during development. John the local guy here who did the testing of the proto units and also helped getting the size of things right told me yesterday that 0.5mm spacer made 5 psi difference in the pressure

 

 

[pluessy]

There is a general misconception that the oil pressure relief valve REGULATES the oil pressure. All it does is LIMIT the maximum pressure at the location of the relief valve. Even the Rotax manual is wrong by saying "The oil pressure from 1.5 to 5 bar (22 to 72 p.s.i.) is controlled by the pressure relief valve (8). The surplus oil returns to the oil pump rotor via the channel (9)."

 

The relief valve can't regulate the pressure below it's opening pressure. According to the Rotax manual that would be 5bar (the operations manual says max 7bar). Below the relief valve setting, the oil pressure is a function of volume (rpm), viscosity (oil temperature), pressure loss (filter, passages) and leakage (lube oil demand from all lube points).

 

I had originally a thermostat fitted with very poor plumbing (90deg brass elbow, 180deg bend in the hose) and lower-than-ideal pressure and pressure fluctuations. Removed the thermostat and re-hosed the suction lines and now have rock-steady 4bar at 5,000-5,200rpm and 80-90deg oil temp. Oil temp control is not with an adjustable blank in front of the oil cooler (25-40% of the area summer-winter).

 

The Rotax engine is the only engine I know of that has the oil cooler in the suction line. Any restriction will cause pump cavitation and erratic oil pressure, the relief valve is more likely a symptom rather than the cause.

 

 

[IBob]

That's a very interesting set of observations, Pluessy.

 

I had not considered cavitation in an oil pump...though I am currently dealing with cavitation on a solar hot water pump, due to unusually long pipe run (and it's looking as though the correct response is to slow the pump down, rather than speed it up, as the manufacturer's instructions suggest).

 

I can confidently state that, once you have cavitation, you can say goodbye to any sort of control.

 

A question, Pluessy: with cavitation in oil...what are the bubbles composed of? Is the composition of the oil damaged???

 

 

[pluessy]

To my knowledge, the oil is not affected in this application. The life of the oil is too short, 50-100h max compared to a hydraulic system (several thousands of hours). The most likely component(s) to cause cavitation (vapour) are also not oil components but fuel! When my oil temps wouldn't come up to 80-90deg C, I had over 2% of fuel dilution in the oil (most of the time it is under 1%).

 

Cavitation is caused by a small part of the fluid experiencing a local pressure that is below the vapour point eg forming a tiny vapour bubble. Speeding up a pump that already struggles to get enough fluid supplied is not the fix. The usual remedy is the reduce the suction loss (reduce pipe length, increase dia and replace/remove fittings with low-loss versions). If that is not possible, increase the source pressure (elevated reservoir or pressurising it).

 

I have a Tecnam P92 and Tecnam, in their wisdom, is installing the oil cooler with the inlet/outlet fittings down (contrary to Rotax installation requirement). After every oil change, I'm cranking the engine with spark plugs removed until I have steady oil pressure. When cleaning out the tank, I also drain the lines and then use a vacuum pump (oil sample pump) to bleed the suction lines and oil cooler at the oil pump inlet.

 

 

[IBob]

I have been reading up on cavitation, Pluessy, and apparently there are a number of circumstances that will cause it.

 

In my case (with the solar hot water) it is a combination of an unusually long pipe run, resulting in large pipe losses, so the pump is working against a high head pressure. In this situation, speeding the pump up does not deliver more liquid, it causes some of the liquid to recirculate in the pump. This, combined with rising liquid temperature as the water heats, was causing cavitation once the inlet temp went over approx 40'C.

 

Prior to slowing the pump, I had increased the size of the suction pipe, ensured that it had swept bends, and modified the return to the tank to ensure I was not recirculating the same hot water, and fitted an additional air/steam trap to ensure I was not air locked in the return to the tank..

 

Fortunately, this has nothing to do with aircraft, since once cavitation sets in, system temperatures rapidly rise out of control, and once you are making steam, the only way to recover control is to turn the sun off!

 

 

[M61A1]

The Rotax engine is the only engine I know of that has the oil cooler in the suction line. Any restriction will cause pump cavitation and erratic oil pressure, the relief valve is more likely a symptom rather than the cause.

And that I suspect is likely the real issue.

Most aero engines use a multi stage pump with one or more scavenge stages and the oil is returned through the cooler, Rotax uses crankcase pressure to return the oil, so any restriction might lead to a crankcase full of oil.

 

 

[Kyle Communications]

people with the issue have changed their oil coolers and pipes and it made no difference. The relief valve change has been to only real change that has made any difference. I have not replaced my oil cooler or pipes but I am pretty sure they are not blocked but replacing the valve with the rotax updated mushroom version half fixed my issue but after changing to Bill Hertzels version it seems to have fixed it completely

 

 

[Kyle Communications]

this is Bills version

 

 

 

 

[M61A1]

It's an interesting issue. I like the way the Bill's version is adjustable. But, I was was amazed when First looked at the rotax oil circuit schematic, to see that they sucked the oil through the cooler. Blockage is not so much an issue, but I have seen a few oil coolers on bikes that get pinhole leaks. On a pressurized system any leaks in coolers, hoses or coupling will show up, but a system under vacuum will just introduce air, creating cavitation and fluctuations.

 

 

[pluessy]

The relief valve is supposed to stay shut to 5bar. If there are pressure fluctuations that happen below that pressure, then an oscillating relief valve is a symptom and not the cause. Changing the relief valve in this case most likely "fixed" the issue as it would have a very different harmonic frequency (different spring and different plunger mass).

 

Jabiru has that problem with oscillating pressure relief valves when the PRV is activated (high oil pressures). The pressure spikes are splitting oil coolers! It doesn't happen at oil pressures below the relief valve setting.

 

 

[Kyle Communications]

This first video is of the problem...although mine was not as bad. This is in a RV12 locally here and it is with the manual style oil pressure gauge using the oil tube

 

 

No comparison. This owner also changed everything inc the pump housing all the pipes and cooler etc as well as total cleaning of the whole system and oil tank. Fitting the new valve fixed the issue. He has almosy 100 hrs now on the new valve and perfect operation

 

 

[pluessy]

OK, defies logic. 45psi is 3 bar and that is well below the opening pressure of the PRV. Mine is on 4bar (5,000-5,200rpm) and changes ever so slightly with oil temp. I don't know what my relief pressure is, I have never revved the engine high enough with cold oil to see that pressure. One interesting fact with Rotax engines is that the running oil pressure actually increases with increasing oil temp. When I take off with 50deg C oil temp, the pressure is initially 3.6-3.7bar and then increases during climb to 4 bar when the oil temp reaches 80degC. This is a result of the diminishing suction line losses with decreasing viscosity, more than off-setting the increasing pressure loss in the post pump circuit.

 

 

[Kyle Communications]

Actually normal ops from what I have seen and on my engine is as the oil heats up the oil pressure goes down. Cold it was about 65psi and once the engine gets to around 84deg it used to be about 54psi.

 

The new Hertzel valve is adjustable. I can actually adjust what oil pressure I want to have. The rotax needs to be no less than around 20psi at idle when hot and really doesnt need to be any more than 40 to 50psi when hot at cruise rpm. Any more pressure than that is a waste of time anyway...all your doing is throwing more oil at the engine that it actually needs.

 

SpringPressureProfile.pdf

 

SpringPressureProfile.pdf

 

SpringPressureProfile.pdf

[pluessy]

Sorry, need to correct you here. The oil pump is fixed displacement, the only control over the volume you have is the engine rpm. You actually throw oil away when you lower the PRV pressure as a portion of the oil leaving the pump is now being returned to the pump inlet and does nothing to lubricate or cool the engine.

 

 

[cscotthendry]

My engine also shows a higher pressure when cold than when warm. As I understand it, this is because of a temperature sensitive bypass valve in the oil filter.

 

 

[Kyle Communications]

I have not seen a engine that gives higher oil pressure when it get hotter.... it always goes down. It makes sense as the oil gets thinner it flows better and not as dense as the pump is still spinning at the same engine rpm regardless of the oil temp. The more viscious it is the easier it will flow. I can understand fluctuations being there if you have lots of bends etc in the oil system due to the added restriction but mine doesnt at all. They are as short as possible and no joiners or extra bends