No Back Spring Update

[1980Mooney]

1 hour ago, PT20J said:

Do you know if that incident involved a Plessey or an Eaton actuator?  I know that Mooney was using both well into the 1990s and the accident reported by @1980Mooney(NTSB Accident Number ERA22LA319) involved a Plessey actuator with only 427 hours since spring replacement. Plessey and Eaton are similar but different designs. Plessey calls the spring a torsion spring and Eaton calls it a no-back spring. Mooney has taken to just referring to both as no-back springs. They may or may not be identical parts -- probably not.

The original service instruction M20-52 from 1/15/92 listed 2 failures of springs in Plessey actuators at 1200 and 1500 hours respectively. It included GEC/Plessey service instruction SI-11 requiring replacing the spring every 1000 hours. This appears to be the origin of the 1000 Hour replacement interval. I found two additional SDR reports of failures of Plessey actuators (one a broken spring and the other of undetermined cause, but likely the spring from the description). With the accident cited above, that makes at least 4, maybe 5, Plessey spring failures.

Service bulletin M20-266A cites a single failure of an Eaton spring and lists a limited number of aircraft that require removal of the Eaton actuator and it's return to Eaton. This is likely the source of the idea that it was a problem related to a particular lot of springs.

Subsequently, Eaton issued service instruction SI102000-901-1 that called for inspection of ALL actuators within 100 hrs and recommended replacement of the spring at 1000 hour intervals. Mooney published this as service bulletin M20-279. Apparently, all these bulletins and instructions were causing confusion so Mooney issued a comprehensive consolidated version as M20-282A. 

So, after all the dust settles, as of 9/22/2004 (the issue date of M20-282A), Mooney has only noted the failure of ONE Eaton actuator spring, which may have been limited to one lot of actuators, whereas 4 or 5 Plessey actuators are known to have suffered spring failures.

So, I would be concerned if I had a Plessey actuator because of the several documented failures, but I'm not so sure that the Eaton's are much of a problem. It would be nice if Mooney would publish the actual number of failures of each type.

Skip

 

 

1 hour ago, 1980Mooney said:

@StevenL757  Here is the latest No-back Spring "tang" failure documented by NTSB investigation. - November 2022

Report_ERA22LA319_105499_2_2_2023 11_04_56 PM.pdf 750.77 kB · 3 downloads

Analysis

After arrival to the destination airport near the conclusion of a cross-country flight the private pilot reported the landing gear circuit breaker (CB) tripped after he attempted to lower the landing gear electrically via the normal method. He reset the CB and again attempted to lower the landing gear via the normal method, but the same CB tripped. He overflew the airport then flew towards the coast where he reviewed the airplane’s Pilot’s Operating Handbook and attempted multiple times to manually lower the landing gear but was unsuccessful. He also contacted his mechanic to obtain assistance, but he was unable to lower the landing gear. At the suggestion of his mechanic, he diverted to a larger nearby airport where he performed a go-around on the first approach because the flight was too fast, then returned and landed gear-up.

According to maintenance personnel associated with recovery of the airplane, lowering the landing gear required them to disconnect the push/pull tubes near the landing gear actuator for each landing gear. Further, the emergency cable activation lever would not move and was “jammed….” Following removal of the push/pull tubes near the actuator the emergency cable activation lever would only travel about 90%. Following removal of the actuator from the airplane, manual operation revealed the actuator turned freely in the retracted direction but would not turn in the gear extension direction. Disassembly inspection of the clutch drive assembly of the landing gear actuator revealed one “tang” of the torsion spring associated with gear extension was fractured. By system design the emergency extension system did not bypass the clutch and damaged spring; thus, a single point failure (torsion spring) precluded manual and normal extension of the landing gear. According to airframe manufacturer, the landing gear actuator and fractured torsion spring are no longer available or supported, but they do have a retrofit for the landing gear actuator.

An airframe manufacturer service instruction specified replacement of the torsion spring in the drive clutch assembly of the landing gear actuator was last performed nearly 5 years and 427 hours earlier. The service instruction, due every 1,000 hours, was due again in about 573 Page 2 of 5 ERA22LA319 hours. The mechanic who performed the airplane’s last annual inspection nearly 8 months and 49 hours earlier reported that as part of that inspection he performed in part, normal and emergency extension checks of the landing gear and found no issues. He did not perform any maintenance to the landing gear except for lubrication.

Probable Cause and Findings

The National Transportation Safety Board determines the probable cause(s) of this accident to be: The failure of the torsion spring inside the drive clutch assembly of the landing gear actuator which prevented normal and manual extension of the landing gear.

Findings

Aircraft - Landing gear actuator - Failure

Personnel issues - Use of equip/system - Pilot

Just to confirm what @PT20J stated, this latest gear actuator failure in 2022 was a Plessey actuator - they call it a torsion spring (not a "no back spring although it performs/grips the same way) and that the GEC (Plessey) actuator is no longer available.

 

[Pinecone]

8 hours ago, toto said:

Blech.  It's kind of hard to see how this "retrofit" is a solution to the NBS problem.  You're trading one actuator with a single point of failure for another actuator with a single point of failure.

Because you are changing out a design that has had at least two failures for one that doesn't seem to have any documented failures.

Yes, a new design that did not have a single point failure would be even better

[toto]

12 minutes ago, Pinecone said:

Because you are changing out a design that has had at least two failures for one that doesn't seem to have any documented failures.

Yes, a new design that did not have a single point failure would be even better

I was under the impression that there was a handful of documented failures for both the Plessy and the Eaton actuators?

[PT20J]

2 hours ago, toto said:

I was under the impression that there was a handful of documented failures for both the Plessy and the Eaton actuators?

Well, I have listed all the one’s I’ve been able to find. If someone has documentation of others, please share.

[carusoam]

We have seen the age affects on the spring… they grow cracks over time in service…

The cracks are near the the heat treated(?) tang at the end…

 

The usual, or more common, gear failure… is when the E-system is partially engaged while the regular gear system is operated…. And a brass gear gets mangled… 

This minor mangled issue prevents the E-system from engaging properly…

For pics… Find @Hyett6420 Andrew documented a bunch of details with nice pics a few years back…

Best regards,

-a-

[toto]

4 hours ago, carusoam said:

We have seen the age affects on the spring… they grow cracks over time in service…

The cracks are near the the heat treated(?) tang at the end…

 

The usual, or more common, gear failure… is when the E-system is partially engaged while the regular gear system is operated…. And a brass gear gets mangled… 

This minor mangled issue prevents the E-system from engaging properly…

For pics… Find @Hyett6420 Andrew documented a bunch of details with nice pics a few years back…

Best regards,

-a-

I tried to link @Hyett6420’s photo post recently, but it seemed to have been deleted. 

Don Kaye has a photo page online here:

https://donkaye.com/infamous-1500-back-spring

[PT20J]

I don't think the analogy to a clock spring is accurate because this is a wrap spring and they are different.

It would be interesting to take one of these apart which I've never done. But looking at Eaton's exploded diagrams and description it seems possible to make some educated guesses about how they work. First, the problem they are trying to solve is that the actuator needs a brake. Lead screw actuators that use an acme thread such as the trim jack screw have enough friction to be self locking for the same reason you cannot push on a nut threaded onto a bolt and cause the bolt to turn. But the friction also makes it harder for the motor to turn the screw to move the load. So, for highly loaded actuators, a ball screw is used where the screw threads ride on ball bearings. This type of screw has much lower friction, but now when the load applies tension or compression to the screw it can back drive the actuator.

The Mooney landing gear will stay down due to over-center locks. But there are no up locks and the weight of the gear would back drive the actuator if there were no brake. Within the Eaton actuator gear train there is a shaft with an input gear on the motor side and an output gear on the screw side. The two gears are not rigidly connected but each is connected to a hub and the spring wraps around (hence the name wrap spring) the two hubs with the tang at each end attaching it to each hub. This assembly is inserted into a tubular metal shell that does not rotate. When the gear is retracted and the weight of the gear tries to turn the lead screw and thus the output gear/hub, the spring unwinds and rubs against this shell and the friction creates the braking effect. This is generically called a wrap spring brake. 

When you put the gear down, the input hub starts to turn and this tightens the spring around both hubs causing them to turn together thereby driving the screw. The functioning of the spring is dependent on the tangs fixing each end of the spring to it's associated hub. If the tang on the input side breaks, the input hub might turn, but the spring will not tighten around the hubs to lower the gear. Or, the broken piece might jam the mechanism preventing the gear from coming down.

If you search the web for wrap spring clutches and wrap spring brakes you will find a lot of animations for how these work in general. However, I have not found any actuators that have such a simple mechanism with such a high dependence on the spring, and most have a lot more parts. It seems that this design was intended to reduce the parts count to a minimum which has the undesirable effect of making the entire mechanism dependent on the stressed spring tangs. The other problem with the design is that the emergency gear extension system only protects against electrical or motor failures and is rendered inoperative by any mechanical failure that prevents the motor shaft from turning. 

Skip

 

[toto]

So what is the current MS hive mind thinking about NBS replacement? I’m on my original NBS, and to be totally honest at this point I don’t feel much difference in risk between keeping the original or rolling the dice on a replacement. 

[A64Pilot]

11 hours ago, carusoam said:

We have seen the age affects on the spring… they grow cracks over time in service…

The cracks are near the the heat treated(?) tang at the end…

 

The usual, or more common, gear failure… is when the E-system is partially engaged while the regular gear system is operated…. And a brass gear gets mangled… 

This minor mangled issue prevents the E-system from engaging properly…

For pics… Find @Hyett6420 Andrew documented a bunch of details with nice pics a few years back…

Best regards,

-a-

Almost certain the entire spring is heat treated. It’s the heat treat that makes it “springy”

If there are of have been cracks at the bend of the tang and or broken tangs, it to me could possibly indicate that the tangs were formed after the springs were heat treated, which is normal, as I don’t think it’s normal to heat treat small springs after being formed, but bending heat treated steel in a small radius can fatigue it

For normal small common springs they are wrapped or formed from already heat treated bulk stock.

We had problems with a batch of elevator down force springs where the ends broke off. They looked exactly like screen door springs and the ends were formed by merely bending the last loop out 90 degrees to leave an open circle to attach a cable too. They broke because bending the wire that much after it was formed overstressed it, fatigued it I believe.

But I don’t remember what the fix was, but it was a manufacturing change of no great consequence, maybe they heated the end prior to bending or opened the radius of the bend, I don’t know.

 

[A64Pilot]

3 hours ago, toto said:

So what is the current MS hive mind thinking about NBS replacement? I’m on my original NBS, and to be totally honest at this point I don’t feel much difference in risk between keeping the original or rolling the dice on a replacement. 

That’s the $50,000 question, I’ve gone back and forth on that myself. I had supposed though that this little spring had gained so much exposure, that and as it’s cost to me is $1,000 I had assumed it would have been made to NASA standards and inspected sixteen ways from Sunday. 

Maybe not?

So now I’m wondering if I should ask for my $1,000 back or wait until it comes in and just keep it.

‘I do feel that my 1981 actuator should be disassembled, cleaned, inspected and re-greased though, grease is surely way past needing replacing. 

[N201MKTurbo]

3 hours ago, PT20J said:

I don't think the analogy to a clock spring is accurate because this is a wrap spring and they are different.

It would be interesting to take one of these apart which I've never done. But looking at Eaton's exploded diagrams and description it seems possible to make some educated guesses about how they work. First, the problem they are trying to solve is that the actuator needs a brake. Lead screw actuators that use an acme thread such as the trim jack screw have enough friction to be self locking for the same reason you cannot push on a nut threaded onto a bolt and cause the bolt to turn. But the friction also makes it harder for the motor to turn the screw to move the load. So, for highly loaded actuators, a ball screw is used where the screw threads ride on ball bearings. This type of screw has much lower friction, but now when the load applies tension or compression to the screw it can back drive the actuator.

The Mooney landing gear will stay down due to over-center locks. But there are no up locks and the weight of the gear would back drive the actuator if there were no brake. Within the Eaton actuator gear train there is a shaft with an input gear on the motor side and an output gear on the screw side. The two gears are not rigidly connected but each is connected to a hub and the spring wraps around (hence the name wrap spring) the two hubs with the tang at each end attaching it to each hub. This assembly is inserted into a tubular metal shell that does not rotate. When the gear is retracted and the weight of the gear tries to turn the lead screw and thus the output gear/hub, the spring unwinds and rubs against this shell and the friction creates the braking effect. This is generically called a wrap spring brake. 

When you put the gear down, the input hub starts to turn and this tightens the spring around both hubs causing them to turn together thereby driving the screw. The functioning of the spring is dependent on the tangs fixing each end of the spring to it's associated hub. If the tang on the input side breaks, the input hub might turn, but the spring will not tighten around the hubs to lower the gear. Or, the broken piece might jam the mechanism preventing the gear from coming down.

If you search the web for wrap spring clutches and wrap spring brakes you will find a lot of animations for how these work in general. However, I have not found any actuators that have such a simple mechanism with such a high dependence on the spring, and most have a lot more parts. It seems that this design was intended to reduce the parts count to a minimum which has the undesirable effect of making the entire mechanism dependent on the stressed spring tangs. The other problem with the design is that the emergency gear extension system only protects against electrical or motor failures and is rendered inoperative by any mechanical failure that prevents the motor shaft from turning. 

Skip

 

The best illustration of how a clutch spring works is the spring on your rivet gun. You have to use the top tang to screw it on and the bottom tang to screw it off.

[Fly Boomer]

13 minutes ago, A64Pilot said:

‘I do feel that my 1981 actuator should be disassembled, cleaned, inspected and re-greased though, grease is surely way past needing replacing. 

I think I read that the lubricant requires mixing powdered molybdenum into grease.

[A64Pilot]

18 minutes ago, Fly Boomer said:

I think I read that the lubricant requires mixing powdered molybdenum into grease.

Maybe, but even though I’m rated to do the work, I think sending it to someone that’s done several times is probably a better idea.

Especially since these things must be made from Platinum

[PT20J]

25 minutes ago, Fly Boomer said:

I think I read that the lubricant requires mixing powdered molybdenum into grease.

That's for the Dukes actuators which are a completely different design. They used a worm and wheel gear drive (like the flap actuator) that inherently resists back driving. But the gears are very highly loaded and need the moly to prevent wear. 20:1 gears are the worst; 40:1 gears are better from a wear standpoint. You end up having to mix that up (or LASAR used to sell it premixed) because the spec is for 10% moly and the highest commercial moly grease is generally 5%. The fact that they specified so much molybdenum tells you something about the extent of the problem.

[PT20J]

4 hours ago, A64Pilot said:

‘I do feel that my 1981 actuator should be disassembled, cleaned, inspected and re-greased though, grease is surely way past needing replacing. 

Maybe it makes sense to remove it and take it to Maxwell or LASAR or someone else who has done a lot of these and have it disassembled, cleaned, lubricated and inspected. If an inspection showed no incipient cracks in the spring, it's probably good for another 1000 hours.

[toto]

5 hours ago, A64Pilot said:

I had assumed it would have been made to NASA standards and inspected sixteen ways from Sunday.

I know nothing at all about this, but it’s still kind of surprising to me that there isn't some sort of keep-the-thing-from-rotating-only-well-enough-to-survive-one-emergency-extension thing that stays in place when the spring breaks.

[1980Mooney]

41 minutes ago, toto said:

I know nothing at all about this, but it’s still kind of surprising to me that there isn't some sort of keep-the-thing-from-rotating-only-well-enough-to-survive-one-emergency-extension thing that stays in place when the spring breaks.

Reality check - how big is this market? TINY and shrinking.  Incredibly unlikely that any new units will ever be built again.  No one is going to spend 2 cents re-engineering the Mooney gear extension system - Not Eaton and certainly not Mooney.

If this becomes too big of a safety issue the more likely outcome will be an AD to lock the gear down permanantly......  

[toto]

5 minutes ago, 1980Mooney said:

Reality check - how big is this market? TINY and shrinking.  Incredibly unlikely that any new units will ever be built again.  No one is going to spend 2 cents re-engineering the Mooney gear extension system - Not Eaton and certainly not Mooney.

If this becomes too big of a safety issue the more likely outcome will be an AD to lock the gear down permanantly......  

Not sure what this is about. I’m not suggesting that someone re-engineer anything - I’m just expressing surprise that it passed engineering reviews in the first place. 

[1980Mooney]

8 minutes ago, toto said:

Not sure what this is about. I’m not suggesting that someone re-engineer anything - I’m just expressing surprise that it passed engineering reviews in the first place. 

Of sorry - I misunderstood. I thought you were wondering if someone could come up with a fix for this actuator which was originally designed in ~1974-75 when it went into production for the first M20J built in September 1976.

So you are wondering why this wasn't better designed back in ~1974-75?

[toto]

54 minutes ago, 1980Mooney said:

Of sorry - I misunderstood. I thought you were wondering if someone could come up with a fix for this actuator which was originally designed in ~1974-75 when it went into production for the first M20J built in September 1976.

So you are wondering why this wasn't better designed back in ~1974-75?

No worries. Yeah - it seems crazy that there wasn’t some kind of mechanism in the original design that would give you one manual extension in the event of a spring failure. I don’t care if it’s safety wire or a zip tie - just something that doesn’t require calling the insurance company if the tiny spring fails. 

[EricJ]

1 hour ago, 1980Mooney said:

Of sorry - I misunderstood. I thought you were wondering if someone could come up with a fix for this actuator which was originally designed in ~1974-75 when it went into production for the first M20J built in September 1976.

So you are wondering why this wasn't better designed back in ~1974-75?

The first year of J models had the old-style Dukes actuator with the crank handle emergency extension.    I think this is one of the attributes that makes the 77 J model the most desirable year. 

[toto]

3 minutes ago, EricJ said:

The first year of J models had the old-style Dukes actuator with the crank handle emergency extension.    I think this is one of the attributes that makes the 77 J model the most desirable year. 

Out of curiosity, what were the superior features of the Eaton?

[201Steve]

3 minutes ago, EricJ said:

The first year of J models had the old-style Dukes actuator with the crank handle emergency extension.    I think this is one of the attributes that makes the 77 J model the most desirable year. 

Yes sirrrr. Only thing about the 77 I really dislike is the location of flap and trim indicators. Other than that, I’m ok with the other qualities. 

[PT20J]

10 minutes ago, toto said:

Out of curiosity, what were the superior features of the Eaton?

It was beefier than the Dukes it replaced.

https://www.donmaxwell.com/ad-75-23-04-sb-m20-190

 

[1980Mooney]

34 minutes ago, PT20J said:

It was beefier than the Dukes it replaced.

https://www.donmaxwell.com/ad-75-23-04-sb-m20-190

And based upon a KNR Shop Talk article it sounds that in some ways the GEC-Plessey was a better design than the Eaton (Vickers, Avionics Products) design. The Emergency Gear Extension coupler system consists of a steel key block, a bronze (Eaton) or steel (Plessey) slot for the block to engage and a spring to hold the halves together. The Plessey had steel to steel drive attaching the pull cable to the transmission where Eaton had bronze to steel.  The bronze was subject to wear (probably a good reason to not test the Emergency Extension System very often!) The Plessey is also easier to rig.

However it seems that Eaton had better design and quality control of the spring resulting in fewer failures than the Plessey (see Top Gun - Tom Rouch article on page 17 of Mooneyflyer August 2017. (see after KNR)

200107 Emergency Gear Extension Systems (knr-inc.com)

"There are two types of landing gear actuators used in both twelve and twenty-four volt aircraft. The early actuators were manufactured by Avionics Products (Eaton Corp., El Segundo, CA). This system has an adjustable rod end requiring some rigging adjustment. This is the system with the bronze coupler slot, mentioned above. The later type of system is made by GEC (Plessey), UK Aerospace, Whippany, NJ, and is almost identical with the following exceptions: the jackscrew is a complete unit without the adjustable rod end. The rod end is built into the jackscrew itself and is much easier to rig the landing gear system. The other key difference is that the emergency drive system in the Plesley system uses a steel-to-steel drive to attach the pull cable to the transmission. The bronze-to-steel system is subject to excessive wear; the steel-to-steel system is much more durable. Each of these actuators has one service instruction against it. See Mooney Service Instructions, number M20-52B (Avionics) or M20-92 (Plessey).

In conclusion, the Plessey system is more robust than the Avionics system because of the steel-to-steel coupler. "

"Ask the Top Gun" Tom Rouch Page 17

Present Position (themooneyflyer.com)

The No-Back Clutch Spring

These newer models have a common problem with the no-back clutch spring, which is recommended to be changed every 1,000 hours of aircraft time. A no-back clutch spring is what keeps the gear up when retracted. It keeps the gear from unwinding after the gear up switch shuts off the power. I have dealt with planes that have a broken no-back clutch spring. Most of these had Plessey actuators, which were used in the late 90s. These use a slightly different spring from all other actuators. The Plesseys are now obsolete and their no-back clutch springs are not available. A later problem we have found on the Eaton actuators is chipping of the jack screw. While difficult, I have been able to buy a new jack screw. I don't know current prices, but I do know that a new actuator is well over $10,000.

If the no-back clutch spring breaks while retracting the gear, the gear will not go down for landing. If it breaks during extension, the gear will go down, and probably hold the gear down. You probably won't know that the no-back clutch spring is broken until the next flight when the gear won't retract. The main defect in this gear system is that the emergency extension depends on an intact no-back clutch spring.

Another problem with the newer actuators is the wear on the emergency extension cable. It is Teflon coated and the Teflon can tear and peel loose and that loose Teflon can actually jam the actuator. It actually happened during a practice emergency landing gear extension. The Emergency Clutch The other wear item is the emergency clutch, which is made out of soft brass. If the emergency cable is not rigged correctly, this may allow partial connection of the emergency clutch to the actuator drive gear, which causes the brass on the emergency clutch to wear off, and the clutch will not engage.