252 High power Lop

[carusoam]

If you take more time during the run-up... there may be some interesting information about the mags that can be discerned here...

It looks like the run-up happened faster than the rate of data collection is occurring...

 

[jlunseth]

As I have mentioned, I use a high MP setting with my turbo although not as high as Mikosch.  I use 34” and a FF of 11.3 (or thereabouts, it doesn’t hold precisely still).  I have wondered whether that is hard on the turbo.  The plane is currently in for the annual and I asked Willmar (Oasis) to take a very good look at the turbo thinking it is about time to rebuild.  It has around 1200 hours on it since it was last OHd, and a few hundred of those are at that higher cruise MP setting.  They, in turn, talked to their turbo rebuilder to find out what to look for.  The news is that the turbo is in great shape and it would be a waste of money to rebuild it at this point, so I think I am going to stick with my setting. It is about 71% HP in my aircraft.

[carusoam]

High MP will surely be working the TC harder...  rpm of the turbine is related to MP and FF... more exhaust volume driving the turbine...

Temp control will be a tremendous part of the wear equation...

 

It would be really good if Miko’s data collection could include MP, rpm, and TIT... to get a clearer picture of what is going on inside the engine...

For a few euro-amu more...   

PP thoughts only, not a mechanic or CFI...

Thanks to @Mikosch for being kind enough to share his data!

Best regards,

-a-

[kortopates]

43 minutes ago, jlunseth said:

As I have mentioned, I use a high MP setting with my turbo although not as high as Mikosch.  I use 34” and a FF of 11.3 (or thereabouts, it doesn’t hold precisely still).  I have wondered whether that is hard on the turbo.  The plane is currently in for the annual and I asked Willmar (Oasis) to take a very good look at the turbo thinking it is about time to rebuild.  It has around 1200 hours on it since it was last OHd, and a few hundred of those are at that higher cruise MP setting.  They, in turn, talked to their turbo rebuilder to find out what to look for.  The news is that the turbo is in great shape and it would be a waste of money to rebuild it at this point, so I think I am going to stick with my setting. It is about 71% HP in my aircraft.

MAP alone isn't important, its high MAP and high altitude that makes the turbo really spin or work. Add in high TIT to increase the wear rate.  

I should add, 1200 hrs on your turbo is doing great. You're obviously taking care of it.

Edited April 4, 2020 by kortopates

[kortopates]

11 hours ago, Mikosch said:

 

Here is the link to the test flight I did. 

 

At 12:48 I did the big mixture Pull back to about 13GPH and at 25:52 I tried the same powerseeting again in cruise the the cowlflaps fully closed to get a feeling how hot the CHT would get at that power setting. 

What was the objective of your test flight? 

[MikeOH]

28 minutes ago, kortopates said:

MAP alone isn't important, its high MAP and high altitude that makes the turbo really spin or work.  

Paul,

Can you please elaborate a bit more?

Questions/thoughts that come to my never had/flown a turbo mind:

1) What are the real wear factors for a turbo?  I would think temperature (TIT) would be the crucial one. Not so convinced that higher rpm in and of itself is a huge factor as long as sufficient cool oil is being supplied to the shaft bearing.

2) Why would the turbo be working 'harder' at altitude if the compressor is working with a lower air density?  Sure, it will spin faster to get the required boost but, again, is the higher rpm an important factor in wear/reliability?

Thanks in advance!

[kortopates]

2 hours ago, MikeOH said:

Paul,

Can you please elaborate a bit more?

Questions/thoughts that come to my never had/flown a turbo mind:

1) What are the real wear factors for a turbo?  I would think temperature (TIT) would be the crucial one. Not so convinced that higher rpm in and of itself is a huge factor as long as sufficient cool oil is being supplied to the shaft bearing.

2) Why would the turbo be working 'harder' at altitude if the compressor is working with a lower air density?  Sure, it will spin faster to get the required boost but, again, is the higher rpm an important factor in wear/reliability?

Thanks in advance!

Mike, you are so right about TIT. I was editing my my response to correct that oversight while you were writing your reply. But absolutely TIT plays a very important role by increasing the wear rate as the TIT climbs to higher temps. The turbo charger is already glowing with cherry red color at 1550F yet TIT limitations are another 100-200F higher than that. Most of the earlier turbos have a TIT limitation of 1650F while some of more modern designs, theoretically with better metallurgy,  have pushed that up to 1750F. Yet even with those, we recommend not exceeding 1600F to improve turbo and other exhaust component longevity. 

I should back up a bit though begin with the different controllers used to regulate the upper deck pressure (UDP) since different controllers influence how hard the turbo is working and then go into more detail about what wears out in the turbo and what causes it.

Some of the controllers strive to maintain a constant UDP a few inches above redline MAP regardless of how much MAP you're maintaining. This is one fundamental reason why MAP alone doesn't influence wear. An example of this is the Absolute  Pressure Controller (APC). Imagine in this case, flying with lower MAP isn't helping  reduce the turbo output since it trying to maintain a constant set point UDP. In fact as the pilot reduces power, the controller will increase oil pressure to close the wastegate to force more exhaust through the turbo to maintain UDP at the desired constant.

A better controller used by most pressurized twins, and my Mooney, is the the Variable Absolute Pressure Controller (VAPC). The variable part of the controller is that it no longer strives to maintains as constant UDP but variable one limited too a couple inches over the required MAP setting. At full WOT it operates just like a APC, but below that its variable to be just above the required MAP. As such its not working nearly as hard as the APC type. 

Moving onto wear points. We find the turbo fails or exceeds limits in two principal areas, i) the garlock oil shaft seals on the shaft connecting the turbine to the compressor and the i) the compressor blades. The spinning turbo shaft is constantly cooled with oil pressure fed from an engine galley and then scavenged back into the engine case with a separate oil scavenger pump. The turbo shaft doesn't have anything like real bearings but only a metal shaft rubbing against a metal housing except the shaft and housing is pressurized with oil held in with a garlock seal which is also only metal. The shaft is spinning at very high rpms  in excess of 50,000 rpm with only a film of oil to protect it. When the garlock seals begin to wear excessively they leak oil that is either pumped overboard out the exhaust or even worse into the induction stream out the compressor side; depending on what part of the shaft the seal is failing. Also as the seal and/or the shaft wears, the shaft develops too much free play which leads to blades touching the outer housing leading to accelerated blade wear and sometimes blade failure. The blade wear is the other common cause of turbo replacement. Blade wear due to excessive shaft free play is only one mode. Blade wear also stems from  blade stretch which is directly related to blade temperature and the rotational speed of the blades which can well above 50K rpm. When rotating at very high rpm's at high altitudes and operating at very high temperatures blade stretch wears the thin curved compressor blades at a higher rate. The garlock seal is also under greater stress as shaft RPM increases, and temperature increases. In fact as the turbo ages, it suffers from some degree of internal coking from burnt oil that reduces the effectiveness of lubrication to prevent wear and hot spots. This is why its recommended to flush the turbo every few hundred hours to reduce coking before it gets bad enough to cause premature failure. 

Other indirect signs of the turbo working much harder at altitude is the significantly increased CDT or Compressor Discharge Temperature. CDT redline is a limitation on installations without  a factory intercooler since too high of an induction air temperature significantly reduces detonation margin of the mixture. Down at low altitudes the CDT is never a concern but as we climb higher and higher the CDT goes up as the compressor  has to breathe more air or spin faster to get to the desired MAP. Interestingly, another indirect sign is oil consumption. When the garlock seal is first beginning to leak it will first appear to the pilot as very sporadic changes in oil consumption. One flight it'll appear to burn a quart or more of oil over a relatively short flight, such as 1-2 hrs. Then no oil consumption at all for several flights. And then excessive oil consumption returns on another flight, frustrating the pilot. What's happening is that the garlock seal is wearing out but at lower altitudes, say 6-10K, the turbo is hardly spinning. But take the plane up to the upper teens or flight levels and now the turbo is spinning much faster since its compressing air nearly only half as dense and with its great increased rpm its pumping a lot more oil out the shaft.  

There are other turbo failures not really related to how hard the the turbo is working or how hard the pilot may be pushing it which I didn't go into, such as blade erosion. The quickest way I know of on how to ruin a perfectly good turbo is to run it up on the ground with alternate air door open (which can pop open on its own; especially if de-cowled for maintenance) or even worse de-cowled with no air filter on it and it then sucks in some light FOD ruining the compressor blades - ouch!

sorry, didn't mean to write such a lengthy reply but its a great question!

Edited April 5, 2020 by kortopates

[carusoam]

Mike,

That was a great question...

Because, it earned a fantastic answer about the inner workings of TCs... to a level that I’m not familiar with...

 

Thanks for the extra added details Paul @kortopates!

Best regards,

-a-

[gsxrpilot]

That level of detail usually comes with university tuition attached. It's experts like @kortopates that really set's MooneySpace apart from other pilot communities.

Thanks Paul!

[MikeOH]

Thanks, Paul!

Bunch of stuff you discussed that NEVER would have occurred to me...like blade stretch at high rpm.

Awesome answer

[aviatoreb]

2 hours ago, kortopates said:

...This is why its recommended to flush the turbo every few hundred hours to reduce coking before it gets bad enough to cause premature failure. 

 

What is involved in turbo flushing - I know nothing about this so please explain from scratch.

I loved your long posting.

[Mikosch]

Thank you all for your response. Very interesting! 
I have a very badly worn out turbo charger from my old BMW X5 3.0 diesel on my desk at home.  The amount of play from the shaft is incredible. 
 

As soon as I have added all the features to the JPI, I’ll be back with some more engine data. 

[tmo]

7 hours ago, kortopates said:

Interestingly, another indirect sign is oil consumption. When the garlock seal is first beginning to leak it will first appear to the pilot as very sporadic changes in oil consumption. One flight it'll appear to burn a quart or more of oil over a relatively short flight, such as 1-2 hrs. Then no oil consumption at all for several flights. And then excessive oil consumption returns on another flight, frustrating the pilot. What's happening is that the garlock seal is wearing out but at lower altitudes, say 6-10K, the turbo is hardly spinning. But take the plane up to the upper teens or flight levels and now the turbo is spinning much faster

Is this true for both types of pressure controllers, or just the smarter VAPC?

The stock TSIO-360GB/LB engines have a static APC, the variable one comes with the MB/SB versions, correct?

But even with the static APC the turbo works less down low, because there is less of a pressure delta to overcome and the air is denser, correct?

Thank you for the very detailed post.  I would much rather read a longer post like yours, than a one-liner reference to a PM.  So looking forward to your presentation at the Summit.

[jlunseth]

5 hours ago, tmo said:

The stock TSIO-360GB/LB engines have a static APC, the variable one comes with the MB/SB versions, correct?

The stock TSIO-360GB/LB doesn’t have an APC at all.  It has a fixed wastegate, adjustable only on the ground by a mechanic, it does not have the ability to maintain a set pressure.  That is up to the pilot.

Many have an aftermarket Merlyn wastegate controller, which is not an APC, it is a differential pressure controller, it attempts to maintain a constant differential between the pressure at the output side of the turbo and the pressure in the induction system.  The idea is to prevent the turbo from running away, suddenly raising or lowering the output pressure.  So think of a takeoff.  As you roll down the runway you put in an MP setting with the throttle.  As the engine gains speed the exhaust output increases, which causes the turbo to spool up sharply, which causes the MP to increase sharply even though you have not changed the throttle.  You want that to happen (the turbo to spool up), but you don’t want the turbo and the engine constantly chasing each other, the differential pressure controller damps that tendency.  It also, by the way, enables the wastegate to fully close, which the fixed wastegate can’t do, giving the aircraft a significantly higher Critical Altitude (from 15,500 to about 22k). APC is one type of wastegate controller, there are several.

[bcbender10]

11 hours ago, gsxrpilot said:

That level of detail usually comes with university tuition attached. It's experts like @kortopates that really set's MooneySpace apart from other pilot communities.

Thanks Paul!

I couldn't agree more! Thanks, Paul. @kortopates. I upgraded from a Piper Arrow II (great planes) last August. This forum and Paul's flight instruction in my new-to-me Encore are incredible sources of wisdom. This string is particularly helpful for me. Yesterday I applied some techniques discussed and voila... 65%+ LOP ops and all the numbers (including TIT) right where they should be. 

[kortopates]

13 hours ago, aviatoreb said:

What is involved in turbo flushing - I know nothing about this so please explain from scratch.

I loved your long posting.

Thanks!

Here is a step by step procedure from RAM: 

http://www.ramaircraft.com/Maintenance-Tips/Save-A-Turbo.htm

[jlunseth]

4 hours ago, kortopates said:

Thanks!

Here is a step by step procedure from RAM: 

http://www.ramaircraft.com/Maintenance-Tips/Save-A-Turbo.htm

I noticed in that procedure there was the longstanding and disputed recommendation to taxi or otherwise allow the engine a cool down period before shutting down.  For what it is worth, I have not ever done that during the 1200 or so hours I have on my turbo.  Not to say the procedure is not valid, just that recommendation seems to be OWT, either that, or I always take three minutes to get to the ramp.

[kortopates]

9 minutes ago, jlunseth said:

I noticed in that procedure there was the longstanding and disputed recommendation to taxi or otherwise allow the engine a cool down period before shutting down.  For what it is worth, I have not ever done that during the 1200 or so hours I have on my turbo.  Not to say the procedure is not valid, just that recommendation seems to be OWT, either that, or I always take three minutes to get to the ramp.

I agree with you there as well as APS advice on the topic regarding the need for the 3 min cool down before shutdown. I was quoting the article for the flushing procedure.

Are you also asking or wondering if coking is a myth as well? I have no reason to doubt the Turbo re-builders, such as Main Turbo,  that continue to see it as an issue. I just don't trust it comes from the shutdown procedure. I have wondered if its happening in flight slowly with excessive TIT's but I really don't know.

[jlunseth]

No, I was just addressing the spool down issue.  I don’t know where coking comes from either, if the rebuilders see it, it is no doubt real.  

[N231BN]

Thanks!

Here is a step by step procedure from RAM: 

http://www.ramaircraft.com/Maintenance-Tips/Save-A-Turbo.htm

B-12 is great stuff! We had a Cessna 340 on lease to a 135 operator and decided to park it when the engines hit TBO. When I went to sell it a couple years later both turbos were stuck, after discussing with the new owner we gave them a B-12 soak for the weekend. They both easily popped loose and ran trouble free for a couple hundred more hours.

 

 

 

 

[carusoam]

Other places coking occurs...

same reasons, same oil, same concerns... And same wacky hard to follow advice...

My M20C liked to bake its oil in the exhaust valve guides...  similar exhaust temps that a turbo sees... close to EGT...

Oil is temperature sensitive... and has a tendency to break down chemically when in contact with surface temperatures over 275°F...

This is time related... so as long as the oil is moving while in contact with those surfaces everything is good...sort of...

But, if something interrupts the flow....  or causes it stay in contact with the really hot surface for a long time... the break-down begins... and continues...

Everything gets refreshed with an oil change...

If things look like a carbon factory in the screens or filter... this is a hint of something possibly blocking the flow of oil in a hot place...

My M20C stuck valve occurrence was a valve that wasn’t operating properly. the guide had coked oil carbon in it...

So coking is probably not a normal occurrence... but it helps to understand the causes and effects....

PP thoughts only, not a mechanic...

Best regards,

-a-

[gsxrpilot]

I'm at almost 1600 hours on my turbo and I've never once sat with it idling to "cool down". I do believe that's an OWT and actually does more harm than good.

 

[Bolter]

On 4/4/2020 at 5:45 PM, kortopates said:

<snip>

Moving onto wear points. We find the turbo fails or exceeds limits in two principal areas, i) the garlock oil shaft seals on the shaft connecting the turbine to the compressor and the i) the compressor blades. The spinning turbo shaft is constantly cooled with oil pressure fed from an engine galley and then scavenged back into the engine case with a separate oil scavenger pump. The turbo shaft doesn't have anything like real bearings but only a metal shaft rubbing against a metal housing except the shaft and housing is pressurized with oil held in with a garlock seal which is also only metal. The shaft is spinning at very high rpms  in excess of 50,000 rpm with only a film of oil to protect it. When the garlock seals begin to wear excessively they leak oil that is either pumped overboard out the exhaust or even worse into the induction stream out the compressor side; depending on what part of the shaft the seal is failing. Also as the seal and/or the shaft wears, the shaft develops too much free play which leads to blades touching the outer housing leading to accelerated blade wear and sometimes blade failure. 

<snip)

@kortopates This is an area I work with (bearings and turbines), so I wanted elaborate on what is happening within the turbo bearings.  There is a critical phenomena occurring within the bearing that permits the very high rotor speeds without constant wear. 

Before ball bearings, plain bushing and sleeve bearings were the "normal" bearings.  The clearance between rotor and bearing is precisely manufactured (typical clearance tolerance is ±.0003") and filled with oil fed from the sump.  At start up, the rotor rubs the bearing, and the materials of the bearing are intended to wear gently, wet or dry.  When the rotor reaches a certain speed, the oil film will develop enough internal pressure within the tight clearance to float the rotor off the bearing so there is no longer rubbing contact.  This is not from the pressure of the oil feed, but pressure developed within the bearing due to the surface speed difference between bearing and rotor across a small clearance.  In large equipment with plain sleeve bearings, the self generated oil film pressure can exceed the stiffness of steel, yet still permit low friction without wear.  The bearings will wear only during the start and stop events when the speed is less than lift-off speed, and this is what leads to large "play" in the rotor.  Too much clearance and the bearing is not the same as it was designed, permitting too much rotor motion, for example.   

Regarding cooldowns, each machine is different.  In my experience that is NOT small GA turbo's, cooldown adds stress to the wheels and shortens their life (fatigue crack failure).  However, fast shutdowns from max power will leave heat conducting into the bearings, and will coke the oil that is remaining in the bearings and not being flushed with fresh oil.  You get to choose your demons.  If your turbo's are failing from coking, then a cooldown may help.  Otherwise, I would skip it.  I suspect that the long low power descent and low power taxi has removed enough heat that by the time you are parked, you can just shutdown.  Plus it is an accumulated failure mode.  No single event does it.  

Stepping further outside of my direct knowledge, but for the sake of a though experiment...If oil gets unhappy at over 400F range.  And there is some loss of heat from conduction from the turbine.  So maybe a metal temperature of 500F is OK.  And gas temperature is higher than metal temperature.  So if your TIT is under 700F for a prolonged period, you may be below coking temperatures.   How does that jive with turbo experience of @gsxrpilot and some others who have long turbo life and do NOT practice an intentional cooldown?  What is your TIT on descent and taxi?

On that note, if you like cooldowns, go for it, but skipping it once will not cause a failure.  If you have people around you, and there is a flow of planes behind you, like when you are parking with the Caravan at the North 40 at OSH, just shut down your engine for the safety of others, just like was briefed. 

-dan 

[kortopates]

51 minutes ago, Bolter said:

Before ball bearings, plain bushing and sleeve bearings were the "normal" bearings.  The clearance between rotor and bearing is precisely manufactured (typical clearance tolerance is ±.0003") and filled with oil fed from the sump.  At start up, the rotor rubs the bearing, and the materials of the bearing are intended to wear gently, wet or dry.  When the rotor reaches a certain speed, the oil film will develop enough internal pressure within the tight clearance to float the rotor off the bearing so there is no longer rubbing contact.  This is not from the pressure of the oil feed, but pressure developed within the bearing due to the surface speed difference between bearing and rotor across a small clearance.  In large equipment with plain sleeve bearings, the self generated oil film pressure can exceed the stiffness of steel, yet still permit low friction without wear.  The bearings will wear only during the start and stop events when the speed is less than lift-off speed, and this is what leads to large "play" in the rotor.  Too much clearance and the bearing is not the same as it was designed, permitting too much rotor motion, for example.   

Great explanation on the mechanism behind lubrication and how important turbo speed is to effective lubrication.

51 minutes ago, Bolter said:

Regarding cooldowns, each machine is different.  In my experience that is NOT small GA turbo's, cooldown adds stress to the wheels and shortens their life (fatigue crack failure).  However, fast shutdowns from max power will leave heat conducting into the bearings, and will coke the oil that is remaining in the bearings and not being flushed with fresh oil.  You get to choose your demons.  If your turbo's are failing from coking, then a cooldown may help.  Otherwise, I would skip it.  I suspect that the long low power descent and low power taxi has removed enough heat that by the time you are parked, you can just shutdown.  Plus it is an accumulated failure mode.  No single event does it.  

Stepping further outside of my direct knowledge, but for the sake of a though experiment...If oil gets unhappy at over 400F range.  And there is some loss of heat from conduction from the turbine.  So maybe a metal temperature of 500F is OK.  And gas temperature is higher than metal temperature.  So if your TIT is under 700F for a prolonged period, you may be below coking temperatures.   How does that jive with turbo experience of @gsxrpilot and some others who have long turbo life and do NOT practice an intentional cooldown?  What is your TIT on descent and taxi?

WRT to TIT, they cool down as soon as we reduce power on approach, either from pattern altitude or from the FAF. A leaner mixture than rich helps to keep them a bit higher as well as the CHTs for a possible go around, but the reduced power cools them down on the final approach rapidly to below 1200F slowly decreasing to near 1000F. By the time we're landing they are at their minimum which is below 800F, and as we taxi to the hangar they'll actually go back up closer to 900F. Idling at this point isn't going to get them much cooler and not as cool as at landing. This is the basis of the rational for why a further cool down isn't needed despite the POH warnings calling out for a 3 min cool down. 

In your explanation of bearing wearing primarily "start and stop events when the speed is less than lift-off speed," is the "lift-off" speed a turbo RPM that bearing and rotor shaft separate? and if so any idea if we might approach that at a typical idle RPM? Or is repeated idling in the run-up area waiting several minutes for a IFR release well below this and similarly what Paul's @gsxrpilotcomment above about the cool down actually does more harm than good by contributing to significant bearing wear?

 

Edited April 6, 2020 by kortopates

[gsxrpilot]

Thanks for all the good info. It's always good to be learning, and I'm certainly learning from this thread.

On the topic of letting a turbo cool down. I'm not trying to say that cooling a turbo down isn't beneficial, just that sitting and idling for 3 or 5 minutes is the wrong way to do it. What I learned at the APS class in Ada, OK where a lot of testing with well instrumented turbos and engines has been done, is what @kortopates says above. And that is that the turbo is coolest as I come over the numbers on landing. And that between touch down and reaching my hangar or parking space, the turbo is actually warming up a little bit. And sitting at idle power for an additional 3 to 5 minutes just serves to warm the turbo even more.So that would indicate to me that if a cool turbo is a happy turbo, then I should land, get to parking, and shut it down.

So that's what I've always done. So far, it seems to be working.