What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Viking 110 notice

KeithO

Well Known Member
There have been some RV-12s built with Viking 110 engines, so I just wanted to post some news on this engine here.

In the last 6 months there have been 2 in flight fatigue failures of the gearbox input drive flange ears, one each on 2 different aircraft. One failed at 130 hours total, the second failed at approximately 800 hours. I am attaching a picture showing the fatigue crack on the 800 hour engine where the drive ear departed the airframe after tearing apart the radiator and cooling tube on the side of the engine.

Both failures occurred during take off. one caused the total loss of the airplane, the second the pilot was able to do a 180 and downwind landing on the same runway used for take off. Both had the potential to be fatal accidents, it was just a question of how high above ground the airplane was able to get prior to failure.

If you have one of these engines, PLEASE remove the gearbox and inspect both drive flanges for fatigue cracks. Dye penetrant will find even small cracks. Al that is needed is a new centering bushing from Viking so that the gearbox is properly located when it is put back on.

If you inspect your drive flanges and do find cracks, inform the factory, you will need to get replacement parts. Replacing the flange on the flywheel side is pretty simple but for the pressed on gearbox input flange you may want to ship it back to Viking to get replaced. I'm assuming none of the replacement parts will be free, but if quite a few are needed Viking will be able to make a bigger batch which may save some money. Viking is claiming that only engines that did not comply with the flywheel and drive flange AD are at risk, but based on my investigation the failure at 130 hours was in full compliance with all the correct parts. I would suggest better safe than sorry. Im pretty sure the centering bushing does not cost much if indeed your parts turn out to be fine.

Be careful out there...

Keith Olivier
 

Attachments

  • Fatigue fracture.JPG
    Fatigue fracture.JPG
    123.1 KB · Views: 669
I was active on the Sonex site around 2009 and, possessing expertise in liquid cooling systems, I started investigating the background of Viking’s Honda conversion. A series of claims advanced by Jan proved to be not just negligent but patently false. In response to one of my questions he claimed to have consulted a radiator manufacturer, but his extended response clearly revealed that he had no idea (at the time, I don't know now) on how to size a radiator. I contacted the radiator manufacturer’s technical department and actually spoke with the design engineer. He was the person that would respond to any such inquiry, had never heard of Jan Eggenfellner and never spoke with anyone regarding an aircraft application for that radiator.

My continued investigation turned another patently false premise. I made a point to visit the Viking tent at OSH that year and not seeing Jan, I spoke with a Viking Rep expressing my desire to speak with him. I was asked to wait while he went in search of Jan. He came back to say he was unavailable and to come back later. As the Rep left, another bystander told me that Jan was there, saw me coming, acted disturbed and left the site.

Just went to the Viking webpage and read the following:

"Here at Viking, we spend every waking moment thinking about torsion dampers, radiator efficiency, and all sorts of things related to engines and flying machines."

Anybody interested would do well by reading the Landoll thread on this site or visiting the Homebuiltairplanes forum and read the thread on PSRU's

Reading the above post I’m reminded of the wise saying:

“ plus ça change plus c'est la meme chose?”
(the more things change, the more they stay the same).

As I said so many years before, Caveat Emptor!
 
Last edited:
Not sure if it applies to this but I just finished reading a post on Facebook that a SB for the 110 is going to be coming out for adding a 5th gearbox attach point to increase gearbox rigidity. Maybe there will be a SB for this issue too. Not sure, don’t own/know much about a Viking, am interested in one for my rv12 though.

Jacob
 
Jacob, trust me the 2 things are related. If Jan comes up with a better 110 gearbox mount I expect these issues will go away because he is not having the same problem on the 90, 130, 150 or the turbo engines. Its just a matter of convincing him to do the right thing so the 110 fleet can keep chugging along without falling out the sky...

Not sure if it applies to this but I just finished reading a post on Facebook that a SB for the 110 is going to be coming out for adding a 5th gearbox attach point to increase gearbox rigidity. Maybe there will be a SB for this issue too. Not sure, don’t own/know much about a Viking, am interested in one for my rv12 though.

Jacob
 
Just went to the Viking webpage and read the following:

"Here at Viking, we spend every waking moment thinking about torsion dampers, radiator efficiency, and all sorts of things related to engines and flying machines."

“ plus ça change plus c'est la meme chose?”
(the more things change, the more they stay the same).

Jan might spend a lot of time "thinking" about these things but he actually still doesn't understand much about either topic as evidenced by poor design after poor design he puts out. Cutting his rubber/ corded damper with a saw ranks right up there with his other guesses.

He should have a qualified engineer on staff experienced in these fields. Could have saved a lot of money, time and grief for himself and customers.

Yeah, with enough trial and error you may eventually get it right- or not. In the meantime customers are at risk due to your lack of knowledge.

Why not do a proper TV study and the design the PSRU properly from the get go? The cost could be amortized over many hundreds of engines.

Thanks Keith for your research and bringing this to the attention of Viking owners.
 
Last edited:
I made a posting on ZENITH ZODIACS 601/650 LOW WINGERS and first Alissa then Jan attacked anyone who had anything to say and eventually the entire thread was deleted and Jan and Alissa got blocked. I was disapointed that the original post same as this one was not left up. There are a lot of people that need to know.

The attached did get posted on Zenith Aircraft Engines FB page yesterday after some public shaming.
 

Attachments

  • Viking 110 change.jpg
    Viking 110 change.jpg
    116.3 KB · Views: 624
Last edited:
I made a posting on ZENITH ZODIACS 601/650 LOW WINGERS and first Alissa then Jan attacked anyone who had anything to say and eventually the entire thread was deleted and Jan and Alissa got blocked.

Nothing much has changed in the last 15 years since the Subaru days. Bring problems out into the open which could injure people and you're vilified.

TLAR/TLAW engineering isn't a good approach for a vendor selling critical airplane parts. You'll be forever redesigning stuff that fails and apologizing to your customers while reducing confidence in your products and capabilities.

Almost no design or product is perfect first time around, no matter how much engineering and testing went into it and all designs can benefit from field experience to be made better. But when you are guessing and don't do any long term testing, you're asking for a lot more problems down the road.

I don't have a problem with amateurs making their own PSRUs for themselves but that's in a different realm than selling it to others thinking they are getting a well tested, reliable and safe unit. Your customers should not be your beta testers unless they are fully aware of that fact.
 
Penny Wise and Pound Foolish - Rotax is certainly more expensive but it does have well tested (50 000+ engines ) package - and I do mean the entire package, including gearbox.
 
Folks, the problem is totally fixable. Its just not something that the average owner can do himself. When Jan developed the Viking 110 engine he had not yet discovered the "proper" way to utilize a modern car engine, but he has learned. The 130 and 150 and 90 engines are far more mature and robust.

You will see that Jan leaves these engines alone and just adds the parts he needs to have an airplane engine. The gearbox mounts are 1000% better than the Viking 110. The days of sawing bits off the block and adding a ton of cnc billet parts are over. The engines are also left vertical as the car makers intended and the gearbox gets an extra gear to put the prop hub in the right place. Steel flywheels are used, not pocketed out billet aluminum that cracked at the crank interface and has ring gears that flew off.

So, Jan knows how to fix the Viking 110 but has been downplaying and suppressing the failures that have been out there and blaming them on the customers. Given that according to their records 450 of the engines were sold, that is a lot of engines that could fail in the manner described in this thread, in takeoff mode and sooner or later people are going to be killed. It seems, after some ribbing that he has now actually considered this liability a bit more seriously. He can now develop a new gearbox back plate and add a bracket to the block to attach it to on the opposite side of the block to where all the current mount pads are at and then it becomes a mandatory update and people will pay for the parts and get their engines updated. I certainly hope he goes through with it because it will save all the owners who have the 450 Viking 110s from being forced to re-power after we get some deaths from the gearbox failing. Because that is the way this is going to go if nothing happens.
 
Indawarrior said: "This has turned into a Jan bashing. Unfairly, I think."

So, virtue, character and personal responsibility are rationalized away because the perpetrator's feelings may be hurt?

Jan put himself in this position. Back in '09, IIRC, he was representing his product was manufactured according to "accepted industry standards." * He stopped using such terminology but may have not stopped the behavior.

The consequences to victims and their families are greater by several orders of magnitude than the burden of corrective action to Jan.

Have we abandoned the expectation of holding one responsible for making a wrong thing right?

My opinion from back then has not waivered

* placed in quotes for emphasis - from his website but may have been - acceptable or similar
 
Companies with good morals issue a SB and offer free replacement parts to remedy the problem to their affected customers.

This demonstrates they actually care about their customers more than money.

Denying there is a problem and blaming the customer doesn't help your reputation.

Fess up, fix it, move on.
 
...and

"...Almost no design or product is perfect first time around..."

There is no such thing as a perfect design; there has never been and will never be a perfect design.
 
Rocketman,

The fact that there is no such thing as a perfect design while an accurate observation, does not "enable" substandard effort or performance.

Jan embarked on a project to provide a man - usable product.

As a reasonable man, he had the responsibility and the moral obligation to expend best efforts to produce a "man-rated" product. He did not and, in the past, marketed a substandard design claiming it to be made "according to industry standards." I suspect the marketing has changed but the behavior has remained the same.

I have personal experience observing the impact on an ethical man the consequences of his missing a design item that resulted in a fatality. He explained that nothing in his experience alerted him to the possibility of a problem. He spent a large portion of his assets in investigating and correcting the problem.

Fixing the problem didn't ease the conscience of this ethical, moral man.

In this case, a reasonable testing period should have given evidence of the weakness of the PSRU design. IMHO, if a "proper" designed part would have cost $100 more at the time, and $150 now, AT A MINIMUM, Jan should absorb the difference and offer a quality replacement at $50 off.
 
I am working with the Viking 110 owners who utilize the 708HU engine controller, which Jan claims is responsible for all the failures. The amazing thing is that since 2014 there has not been a single engine problem in that entire fleet. There were probably 30-40 owners who went with this engine controller when Jan refused to make changes to the controller provided which the owners felt rendered their engines non airworthy at the time.

I am working on a design proposal for a new gearbox back plate and a new bracket for the underside of the block (in the installed position) which provides a 5th attachment point for the gearbox. The centering bushing and the flexible damper will also be of a new design, although it will be compatible with the existing drive flange geometry.

The existing 4 post mounting of the gearbox will be replaced with a 1 piece casting that uses these 4 attachment points but would obviously be stiffer in the radial direction given it is a single piece. This piece could be retrofitted to any 110 gearbox original or updated, but this change alone will not solve the root cause. The extra gearbox mount point will be required.

I honestly hadn't expected Jan to make any commitment to do anything, so we will observe what emerges in December as he says. If his proposal looks good, then we can go with that. If not we will be making parts next year and doing ground testing using Mark Hubelbanks engine that has been removed from his airframe. Since his gearbox failure his wife has basically refused to get back in the airplane until it has a different powerplant, so that ship has sailed.
 
If not we will be making parts next year and doing ground testing using Mark Hubelbanks engine that has been removed from his airframe. Since his gearbox failure his wife has basically refused to get back in the airplane until it has a different powerplant, so that ship has sailed.

I understand ground testing... but isn't testing in turbulent air required with gyroscopic precession and other dynamic factors?
 
I understand ground testing... but isn't testing in turbulent air required with gyroscopic precession and other dynamic factors?

Of course it will be flown. But what do you do before you fly it ? The first thing we will do is measure the deflection of the gearbox input shaft relative to the crank with the standard gearbox and mount for increasing amounts of bending moment applied to the prop flange in several different radial directions. This would then be repeated after making the changes to the mount and adding the new gearbox support point on the opposite side of the crank centerline. So we should have definitive answers regarding the change in stiffness before to after.

Since my plan is to replace the guibo with a custom part using PU bushings and without cordage and a new permanent guide bushing (steel with glacier bushing insert and welded to the flywheel side drive hub, with a pilot for centering on the crankshaft), I do think there should be some hours put on the assembly in ground testing to make sure the PU bushings hold up, that heat dissipation is OK (for that we could try to find a resonant point near idle that really works the flexible coupling). We can also check if we get any fretting problems between the gearbox input shaft and the new guide bushing. I was going to use high pressure grease in that joint like copperslip to help prevent that problem but this is all safe work that can be done on the ground. I think everyone expects that TV issues are usually not encountered at higher power/RPM.

We dont expect problems with the gearbox internals since they seem to have held up well with no issues we are aware of. Again, having said that, since I am qualified in all 4 NDT disciplines to level 2, at the start of this exercise I will be dye penetrant inspecting all the parts, including the housing, to see if we discover any additional fatigue cracks that we are simply not aware of. If we have any new discoveries we will have to take that into account and may yet be needing even more new parts. We shall see.

Finally we will be looking for a test pilot after we have got that far, someone who flys more hours than average. We would potentially want to inspect the drive flanges which so far have been the point of failure at 50 hour intervals to confirm absence of fatigue cracking. If the metallurgical analysis of the existing drive flanges indicates embrittlement due to the welding process used for the pins, we may need new parts and have them heat treated to normalize them to prevent this problem before we start. Pending what the metallurgist says when he looks at the parts that cracked all the way through. Any part like this should be ductile and not show brittle fractures.
 
Last edited:
Rocketman,

The fact that there is no such thing as a perfect design while an accurate observation, does not "enable" substandard effort or performance.

Jan embarked on a project to provide a man - usable product.

As a reasonable man, he had the responsibility and the moral obligation to expend best efforts to produce a "man-rated" product. He did not and, in the past, marketed a substandard design claiming it to be made "according to industry standards." I suspect the marketing has changed but the behavior has remained the same.

I have personal experience observing the impact on an ethical man the consequences of his missing a design item that resulted in a fatality. He explained that nothing in his experience alerted him to the possibility of a problem. He spent a large portion of his assets in investigating and correcting the problem.

Fixing the problem didn't ease the conscience of this ethical, moral man.

In this case, a reasonable testing period should have given evidence of the weakness of the PSRU design. IMHO, if a "proper" designed part would have cost $100 more at the time, and $150 now, AT A MINIMUM, Jan should absorb the difference and offer a quality replacement at $50 off.


I cannot comment on Jan nor the engine in question as I have no background or experience with either.

That said, my comment stems from an engineering perspective; a perfect system cannot be designed or built. Anyone who claims they have done it is not being realistic.

Yes, I agree that in business there is a responsibility to the customer…
 
Bob, all of these engines are long out of warranty and I dont think that at this point anyone would realistically expect any sort of "free" solution to the problems. But if a solution is made available that one could consider modest in cost relative to the cost of re-powering, there will be many people who would choose to use it.

If I take my Jabiru 3300 gen 2 engine. It has a theoretical top overhaul limitation at 1000 hours that will cost $7500 (I have a quote from the folks at Arion). Now I have only 700 hours on my engine and it has had strong blowby and oil consumption since a little over 500 hours. If I go ahead and have the work done, the top overhaul is not going to mitigate in any way the series of issues that this engine suffers from. The head bolts will still need to be torqued at every oil change interval of 25 hours. The case studs will still be of the "known to be too weak" original small diameter and could yield and could result in fretting of the case halves and out of round wear of the main bearings.

To get the known issues addressed I have to spend more than the $7500 quoted to get the cases machined for the bigger studs and still the yielding of the head bolts will continue because there is no solution for it. The wear on the cam and flat cam followers will also continue, inspect and repair as necessary as it comes up. And best of all, the high temperature at which this engine operates will continue to cause rapid sludge buildup on the pistons and I can expect that in another 600-700 hours I will have the same blowby and oil consumption issues as I have now.

I would have to be stupid to accept this maintenance path and I wont. My Lightning will be getting a Honda engine and I can forget about all these problems and someone else can carry on with the Jab 3300. So I agree that there is no perfect engine, yet somehow I feel that all of these manufacturers are being held to a very low standard. We are well aware of the draconian restrictions the Jabiru folks are under in Australia and possibly New Zealand, yet despite the fact that our engines come out the very same factory, we in the US pretend that none of it applies to us because it has been sold in a different jurisdiction.
 
Last edited:
Bob, many years ago I sat in front of a hiring panel trying to get a job at Volkswagen. One of the questions they asked me is "how do you define quality?" My answer was that "a product exhibits quality when it meets the customers expectations". Apparently that was the best answer they had heard all day.... The expectations of a soccer mom, or someone who drives in a sporty way or a professional rally driver, or a business executive taking a client somewhere are obviously not the same.

I would therefore say that in the real world, perfect does not mean zero defect. It means meeting the expectations of the customers. No one has an infinite amount of money to spend or time to wait for the development to be completed. No car that is launched is perfect either, but the manufacturer has the opportunity to keep refining the product as it is in production to address unforseen issues. The Ford Focus, which today has a very good reliability record, had a very bad launch when the spark plugs specified caused a considerable amount of misfiring, so bad in fact that a large % of the catalytic converters were destroyed within the first few months of service. Because of the precious metal loading this was a very expensive warranty for Ford, several hundred $ per car. Once the spark plug problem was identified the issue went away entirely and following a general upgrade on materials used for the exhaust system became a very reliable car that they sold lots of. Especially once they introduced the variable nozzle turbo diesel that was simply a rocket (sorry it wasn't sold in the US)...

I cannot comment on Jan nor the engine in question as I have no background or experience with either.

That said, my comment stems from an engineering perspective; a perfect system cannot be designed or built. Anyone who claims they have done it is not being realistic.

Yes, I agree that in business there is a responsibility to the customer…
 
If I was given the responsibility for ground testing the new PSRU plate, I'd mount the engine/propeller assembly on a Tilt-Yaw table to check (both the old) and the new plate's response to the gyroscopic loads.

There should be digital measuring "cells"? to check relative motion created by the gyroscopic load testing
 
Last edited:
Marc, there are a few problems with what you are proposing.

The first is that this work is entirely self funded with the goal to prevent people getting killed by the failures and the reputation of experimental aviation being drug further into the gutter. Obviously I would not have access to the kinds of facilities needed to do what you are suggesting.

The second is to consider that in its original configuration the crankshaft side drive hub contains a plastic bushing which serves to center the gearbox when it is mounted to the engine. The attachment holes have quite a lot of slop in them and there are no dowels for alignment. The locations dowels would go in the original bellhousing were cut off by Jan when preparing the block for the conversion. The end of the gearbox input shaft is sized so that it slides into the plastic bushing when you assemble it and you then tighten the mounting bolts without applying lateral pressure to the gearbox, thus it should be decently centered in the unloaded state. Then in service, when the prop loads are applied, because the mounting points are all on one side and the end of the gearbox that is supposed to be aligned with the crank is hanging way off in space, the deflection in the circular section mounting posts can now allow the input end of the gearbox to move away from the crank centerline in the axial and radial directions, proportional to the magnitude of the gyroscopic forces. In other words, how abrupt was the pitch up, down or turn rate left or right. Just imagine a stall/spin...

So, when assembled and when gyroscopic loads are applied to the prop, the gearbox input shaft is going to try to move out of alignment with the crank. So long as the alignment bushing is new and undamaged, it will prevent this from happening by transferring the load into the pilot diameter of the input shaft. However many owners have reported that when they removed their gearbox, they found the guide bushing to be so destroyed that they were confused regarding what it was. If it got destroyed, then clearly it will allow relative movement of the input shaft vs the crank and it is that misalignment which will fatigue the drive flange ears because of the rotating bending load put on the pins which is then transferred into the flange itself.

Thus, when I do my stiffness measurement I will not have a centering bushing installed in the assembly because I want to see the stiffness of the system without a radial load being applied to the gearbox input shaft. If I allow the gearbox input shaft to be loaded in these tests there is a good chance that all I would accomplish is move the failure to where the input shaft is breaking off due to the bending load on it. If we see a big reduction in deflection with the new mount, it also means that the chance of having a big side load applied to the input shaft is greatly reduced, which is what we want.

I attached a picture so you can see where the mount locations are relative to the prop centerline and the input shaft centerline in case you missed the earlier one.
 

Attachments

  • Back of Viking 110 gearbox.JPG
    Back of Viking 110 gearbox.JPG
    172.9 KB · Views: 114
Last edited:
Marc, there are a few problems with what you are proposing.

The first is that this work is entirely self funded with the goal to prevent people getting killed by the failures and the reputation of experimental aviation being drug further into the gutter. Obviously I would not have access to the kinds of facilities needed to do what you are suggesting.

The second is to consider that in its original configuration the crankshaft side drive hub contains a plastic bushing which serves to center the gearbox when it is mounted to the engine. The attachment holes have quite a lot of slop in them and there are no dowels for alignment. The locations dowels would go in the original bellhousing were cut off by Jan when preparing the block for the conversion. The end of the gearbox input shaft is sized so that it slides into the plastic bushing when you assemble it and you then tighten the mounting bolts without applying lateral pressure to the gearbox, thus it should be decently centered in the unloaded state. Then in service, when the prop loads are applied, because the mounting points are all on one side and the end of the gearbox that is supposed to be aligned with the crank is hanging way off in space, the deflection in the circular section mounting posts can now allow the input end of the gearbox to move away from the crank centerline in the axial and radial directions, proportional to the magnitude of the gyroscopic forces. In other words, how abrupt was the pitch up, down or turn rate left or right. Just imagine a stall/spin...

So, when assembled and when gyroscopic loads are applied to the prop, the gearbox input shaft is going to try to move out of alignment with the crank. So long as the alignment bushing is new and undamaged, it will prevent this from happening by transferring the load into the pilot diameter of the input shaft. However many owners have reported that when they removed their gearbox, they found the guide bushing to be so destroyed that they were confused regarding what it was. If it got destroyed, then clearly it will allow relative movement of the input shaft vs the crank and it is that misalignment which will fatigue the drive flange ears because of the rotating bending load put on the pins which is then transferred into the flange itself.

Thus, when I do my stiffness measurement I will not have a centering bushing installed in the assembly because I want to see the stiffness of the system without a radial load being applied to the gearbox input shaft. If I allow the gearbox input shaft to be loaded in these tests there is a good chance that all I would accomplish is move the failure to where the input shaft is breaking off due to the bending load on it. If we see a big reduction in deflection with the new mount, it also means that the chance of having a big side load applied to the input shaft is greatly reduced, which is what we want.

I attached a picture so you can see where the mount locations are relative to the prop centerline and the input shaft centerline in case you missed the earlier one.

I have doubts that the plastic centering bushing was doing much of anything to help maintain alignment ( which is why they got destroyed).
For that reason, testing a modification with a bushing installed may be beneficial in that it may provide a clue whether any movement is taking place.
 
The plastic bushing is described by Viking as an assembly aid ONLY that has to be replaced every time the gearbox is separated from the engine. They KNOW it is getting severely damaged when the engine is run. Thus it is acknowledged that there is relative movement between the gearbox and crankshaft. Just not in those words.

If the current failure mode is to be addressed there has to be a reduction in relative movement. A significant reduction that is. I can verify that by measuring the deflection without a guide bushing as I described. If the design goals are met the new system should have near to no deflection. If that condition is met then it is safe to use a permanent guide bushing without the risk that we break off the input shaft.

I have doubts that the plastic centering bushing was doing much of anything to help maintain alignment ( which is why they got destroyed).
For that reason, testing a modification with a bushing installed may be beneficial in that it may provide a clue whether any movement is taking place.
 
The bushing is for alignment only when installing the gearbox. The original design with the aluminum flywheel used the stock Honda needle bearings in the crankshaft end to carry the end of the drive pinion when installing the gearbox. Those were removed and the plastic bushing was used. I have had my gearbox off twice and had no appreciable wear on the alignment bushing. I suspect it may have more to do with how good you are at installing the gearbox attachment bolts while maintaining the centering with no side loading. Once you torque the mounting bolts, the bushing should do nothing but sit there if it is still centered. There is a fairly simple way to stop the gearbox from moving while the bolts are being tightened.

John Salak
RV-12 N896HS
 
Mark told me that when he took his gearbox off, his bushing was so badly wallowed out that at first he didnt know what it was.

John, can you confirm whether there is any pilot feature between the flywheel drive flange and the crankshaft itself. In other words, does the plastic bushing center the flywheel side drive flange, as well as centering the gearbox input shaft ? I'm trying to understand whether it is only the flywheel bolts providing run out control on the flywheel side drive flange or whether some other feature is provided to center the flywheel and drive flange.

Remember that in this system the thing that counts is how well all the parts run true to the crankshaft itself. If the flywheel drive flange and possibly the flywheel itself is only located by the flywheel bolts, that would provide a significant amount of run out that could not be fixed later no matter what you did, without replacing those parts with ones that are better designed.
 
John, this fact alone could be the reason why you have not had any issues with your setup. How many hours on your engine ?

I have had my gearbox off twice and had no appreciable wear on the alignment bushing.

John Salak
RV-12 N896HS
 
I haven't seen an auto flywheel that wasn't located by a central spigot or counter bore.

The Egg Subaru drives also didn't use the factory dowel pins on the engine for location and required an alignment tool.

Some Ross drives were machined incorrectly and had misalignment causing rapid wear.

Lesson: use the factory dowels to locate the drive concentrically with the crankshaft. It's important and makes things easy for maintenance. That's why auto OEMs do it this way.

This would involve a casting or machined billet housing like almost all other commercial PSRUs use. More expensive, slightly heavier but way stiffer.

Stiffness here is critically important to gearbox longevity. Deflection and bending create loadings which can fail components prematurely.
 
John, this fact alone could be the reason why you have not had any issues with your setup. How many hours on your engine ?

I have 350 hours on the engine, average time per flight is about 2 hours.

The flywheel and drive pin plate center on the crankshaft end nipple as you can see in the photo. The steel drive pin plate is CNCed, so I presume the pins centers are concentric to the center hole around the crankshaft nipple. The plastic alignment bushing is inserted in the center pilot hole.

The other photo is the gearbox as it came off the engine with 125 hours from the last removal. The arrow is pointing to the plastic alignment bushing which shows little wear. The bushing has a small step machined on the end to match the step in the crankshaft pilot hole. The part that broke on Mark's drive was one of those three pin ears. The gearbox drive pin plate is pressed on to the splined drive pinion gear shaft. There are two bearings on the drive pinion shaft, one in the mounting case and the other in the removable cover on the other side of the mounting case. It is important to check that the drive pin plate stays parallel to the mounting case as it is pressed on.

John Salak
RV-12 N896HS
 

Attachments

  • Honda 110 Crankshaft End.jpg
    Honda 110 Crankshaft End.jpg
    452.5 KB · Views: 105
  • Honda 110 Gearbox drive with centering bushing installed.jpg
    Honda 110 Gearbox drive with centering bushing installed.jpg
    410.3 KB · Views: 144
I hate to step into something I know nothing about. However, when I look at that frame structure with the leverage afforded by those long rods, all located on just one side, and compare it to all the bell housings I have seen over the years (always SOLID, half round structure), I am not surprised that it exhibits a lot of flexing. Not an ME, but as I look at it, it seems that meaningfull flexing is all but guaranteed, affecting the area near the crank, that is well offset from the supports.

Larry
 
Last edited:
I am an ME and I’m looking at how the Viking 110 gearbox is cantilever-mounted on four thin standoff’s – yikes…

I’m very happy to be flying behind Rotax 912… The excellent gearbox includes hunting gear ratio, overload clutch to protect engine in event of prop strike, and substantial gear case for rigidity to maintain alignment with dowel pins for precise location on front flange of engine crankcase….
 
Not sure if it applies to this but I just finished reading a post on Facebook that a SB for the 110 is going to be coming out for adding a 5th gearbox attach point to increase gearbox rigidity. Maybe there will be a SB for this issue too. Not sure, don’t own/know much about a Viking, am interested in one for my rv12 though.

Jacob

The SB for the 110 came out - years ago!
 
Folks, the problem is totally fixable. Its just not something that the average owner can do himself. When Jan developed the Viking 110 engine he had not yet discovered the "proper" way to utilize a modern car engine, but he has learned. The 130 and 150 and 90 engines are far more mature and robust.

You will see that Jan leaves these engines alone and just adds the parts he needs to have an airplane engine. The gearbox mounts are 1000% better than the Viking 110. The days of sawing bits off the block and adding a ton of cnc billet parts are over. The engines are also left vertical as the car makers intended and the gearbox gets an extra gear to put the prop hub in the right place. Steel flywheels are used, not pocketed out billet aluminum that cracked at the crank interface and has ring gears that flew off.

So, Jan knows how to fix the Viking 110 but has been downplaying and suppressing the failures that have been out there and blaming them on the customers. Given that according to their records 450 of the engines were sold, that is a lot of engines that could fail in the manner described in this thread, in takeoff mode and sooner or later people are going to be killed. It seems, after some ribbing that he has now actually considered this liability a bit more seriously. He can now develop a new gearbox back plate and add a bracket to the block to attach it to on the opposite side of the block to where all the current mount pads are at and then it becomes a mandatory update and people will pay for the parts and get their engines updated. I certainly hope he goes through with it because it will save all the owners who have the 450 Viking 110s from being forced to re-power after we get some deaths from the gearbox failing. Because that is the way this is going to go if nothing happens.

Viking stopped selling the 110 10 YEARS AGO!

Viking issued the SB correcting this issue YEARS AGO.

A few people thought they new better than Viking and chose not to comply with the service bulletin.

Now, years later, they are complaining about a problem then chose not to fis.
 
There have been some RV-12s built with Viking 110 engines, so I just wanted to post some news on this engine here.

In the last 6 months there have been 2 in flight fatigue failures of the gearbox input drive flange ears, one each on 2 different aircraft. One failed at 130 hours total, the second failed at approximately 800 hours. I am attaching a picture showing the fatigue crack on the 800 hour engine where the drive ear departed the airframe after tearing apart the radiator and cooling tube on the side of the engine.

Both failures occurred during take off. one caused the total loss of the airplane, the second the pilot was able to do a 180 and downwind landing on the same runway used for take off. Both had the potential to be fatal accidents, it was just a question of how high above ground the airplane was able to get prior to failure.

If you have one of these engines, PLEASE remove the gearbox and inspect both drive flanges for fatigue cracks. Dye penetrant will find even small cracks. Al that is needed is a new centering bushing from Viking so that the gearbox is properly located when it is put back on.

If you inspect your drive flanges and do find cracks, inform the factory, you will need to get replacement parts. Replacing the flange on the flywheel side is pretty simple but for the pressed on gearbox input flange you may want to ship it back to Viking to get replaced. I'm assuming none of the replacement parts will be free, but if quite a few are needed Viking will be able to make a bigger batch which may save some money. Viking is claiming that only engines that did not comply with the flywheel and drive flange AD are at risk, but based on my investigation the failure at 130 hours was in full compliance with all the correct parts. I would suggest better safe than sorry. Im pretty sure the centering bushing does not cost much if indeed your parts turn out to be fine.

Be careful out there...

Keith Olivier
V

Viking stopped selling the 110 engine 10 YEARS AGO. And issued an SB to correct the gearbox issue, also YEARS AGO.

Yet, some people thought they new better than Viking and chose not to follow the FREE SB at the time. And now, years later they are complaining about the problem.

How a Viking 100 engine can have 130 hours on it when it was sent from Viking 10 years ago., also raises so many questions????

This thread is based on half-truths and complete falsehoods.
 
There are still new unused 110 engines sitting on shelves after the owners lost faith in the reliability of the product before they even finished installing it. Eric bought one of those, did the SB and put it in his plane and it failed 134 hours later. In fact, Eric knows of a brand new zero time 110 engine right now that the owner has not found a buyer for in several years. And there are 450 more of them that could fail at any time.

The primary purpose of this thread is to let owners know that there have been 2 failures in the last 6 months and what to look for on their own installations to identify if they are at risk. Why would you oppose that ? Are you an engineer ? Do you know what you are talking about or just a fanboy ?
 
Last edited:
Message from a Moderator - information is good, better if backed up by facts!

Name calling and personal attacks are not. Keep the thread informative, keep away from personal attacks, and you’ll be within the rules….

Carry On
 
I would like to say something about personal attacks too. The group of airplane owners that elected to develop their own engine controller, known as the 708HU engine controller have been the subject of vicious attack and vilification by Viking because they dared to address engine controller problems that Jan would not resolve.

The project was lead by Mark Hubelbank who owns a company called NorthEast Monitoring, who make medical devices. Mark is an electrical engineer and embedded controller specialist who happened to buy one of the early Viking 110's.

This list below, were SOME of the issues they wanted to address:

1. The original Viking ECU and wiring harness has a number of clear and some possible single point failures. As designed, if any ignition coil or injector shorts to ground or draws significantly excessive current, the whole system will fail. No provision was made to shut down a single item and allow the engine to continue on three cylinders in this case.

2. The Viking supplied wiring harness has many splices and used parts. This is not considered acceptable construction for any critical design.

3. The two ECUs in the Viking design share a common power point of entry and may share some common power conditioning components. No design review was done to verify this. When initially asked, Jan did say there was a common voltage regulator. This may still be the case in the "red" ECU.

4. Although the Honda engine only allows for a single spark plug for each cylinder, it does have separate ignition coils and injectors for each cylinder. As such redundant electronics are justified so there will be no single point failure. The engine will operate on 3 out of 4 cylinders. While the Viking ECU does have two sets of electronics, there is no evidence that they are separated in such a way that the failure on one can not effect the other. All switching is done electronically. Thus the failure of a switch or of components which are "wire-ored" will not allow the second set of electronics to recover if the failure was on the on state. As such, the ECU was considered a non-redundant design.

5. There no diagnostic outputs for the Viking ECU, something as simple as the ignition timing was said to be a "company secret". There is no display to help the pilot understand any problems which may be observed.

6. Wasted spark ignition is used. This increases the temperature of the ignition coils more than Honda provided for thus putting the ignition coil operation into a unknown realm.

7. If the alternate ECU was selected and the backup power wire was used, the entire engine supply current would be going through a single pin on the connector. This could cause a secondary failure.

The controller and its information display unit have been in operation since 2013. The 2 drive flange failures that occurred within this group in the last 6 months involved a newer engine with only 134 hours on it, which was fully SB compliant and Marks own system which had all but the gearbox drive flange updated. His engine had 800 hours when the failure occurred, identical in nature to Erics failure. By now there are likely many engines in the fleet approaching the 800 hour mark. The shortest lived system so far was an SB compliant system.

Any competent engineer who looks at how the gearbox is mounted will agree that it should never have been released in that form, and will agree that it is never going to perform reliably long term. People who dont know the actual facts need to stop commenting on how this failure has only occurred on non updated (SB compliant) engines because that is a patent lie. Also, if you look at the nature of the fracture of the flange itself, an experienced person would also know that one should never use a material with a heat treatment status that will produce the kind of brittle fracture that we see in this example. Whether the heat treatment was deliberate or accidental we currently don't know but we will find out soon enough and will be reporting back on that.
 
Paul, I am obtaining the 800 hour failure engine, it is being shipped to me from MA so it may still be a week or 2 before I can get my hands on it and start the material analysis, dye penetrant inspection of all the parts, and the needed reverse engineering to allow the needed engineering analysis.
 
Is the number of pages in this thread starting to exceed the number of people with Viking 110 engines on RV-12's?

- mark
 
I hate to step into something I know nothing about. However, when I look at that frame structure with the leverage afforded by those long rods, all located on just one side, and compare it to all the bell housings I have seen over the years (always SOLID, half round structure), I am not surprised that it exhibits a lot of flexing. Not an ME, but as I look at it, it seems that meaningfull flexing is all but guaranteed, affecting the area near the crank, that is well offset from the supports.

Larry

As the lawyers say, don't assume facts not in evidence. No one has shown any evidence of flexing, just speculation based on photos. The mounting plate is almost 1" thick 6061-T6 aluminum with a bolted-on cover casing that contains another set of bearings with a similar thickness for the bearing pockets. The mounting plate is about 11" long; the distance from the top standoff bolt to the centerline of the pinion drive gear (which passes through the center of the prop shaft bull gear) is 8". I have a hard time believing that a bolted together boxed structure with 1" thick aluminum on both sides of the two gear shafts is bending in any meaningful way. The suggestion that the 1.75" diameter standoffs are flexing would require two AN6 attachment bolts to stretch under whatever load you may be imagining. The one item I do not like is the potential for the standoffs to move around the bolts at the engine block interface. Eliminating that movement relies entirely on the clamping pressure of the bolts. There is an easy way to fix that with a bracket that is pinned and bolted to the engine block that will provide precise gearbox alignment.

From the documentation I have seen, there is a very detailed 3D CAD model of the engine and the gearbox somewhere that could be useful in proving or disproving the various speculative engineering theories. Hopefully we will get some real data to work with from the fracture and metallurgical analysis Keith O is working on for the failed part. BTW, this is a nice engine package to fly behind and the experimental ECU makes it even better.

John Salak
RV-12 N896HS
 
John, the plate might start out 1" thick but they did machine away a considerable amount of that material, so its not suitable for a hand calculation, the only way to determine deflection is a) to measure it like I have proposed or b) do an FEA calculation. Obviously I can pretty readily do the first, while I would have a lot of work to do the second. If Jan would make a step file available, that would make it easy, but I think the likelihood of that is very low.

I did discover today while doing a google search that the alternate style flex disc that I was proposing is already made by several suppliers. Picture attached. This style of flex plate has a carrier plate that transmits the engine power instead of relying on cordage embedded in the flex coupler. It also has very specific amount of elastomeric material (much more than in the current part) which can be compressed in either direction. The PU bushings can be changed so one can try different shore hardness for the specific application.
 

Attachments

  • V110 Mounting plate.JPG
    V110 Mounting plate.JPG
    54.8 KB · Views: 138
  • Flex disk alternate.JPG
    Flex disk alternate.JPG
    82.9 KB · Views: 146
I did discover today while doing a google search that the alternate style flex disc that I was proposing is already made by several suppliers. Picture attached.

Keith, suggest you consider natural rubber torsional couplers made specifically for this sort of application. Lovejoy Centaflex is a personal favorite, but there are other couplers in various styles. PU bushings are poor for alternating angular displacement...too much hysteresis.

BTW, don't be pushed toward a BMW driveshaft coupler because Viking made that choice. Looking for something to fit the existing driving/driven flanges is a constraint. Climb out of the box.
 
Ross, and Dan, your feedback is appreciated. I was looking for a coupling that would be softer than the stock BMW unit which is mostly cordage and extremely stiff when in tension (from torque application). Since Viking cuts the stock one into 3 pieces, I'm not sure what it is doing in an over-run situation.

All I know is that if you misalign the shafts, one is very quickly stretching the segments of the current coupler, which is very strong in that mode and one can quickly apply the sorts of loads which will result in fatigue failure. If we find only 1 ear damaged on the failed gearboxes, it will suggest that there was asymmetry in the pin positions of either flange to the point that a single pin was carrying the entire load, together with any load created by misalignment and one could see that failing quite rapidly. It is interesting to note on the originally pictured fatigue crack how it runs almost exactly square to the centerline of that "leg" of the spider. If one thinks about it, the engine load is applied at almost a 90 degree angle to the centerline of the drive ear. So one would not expect a fatigue crack from engine load to be where the crack has appeared. On the other hand, a load from misalignment could be expected to be in the radial direction and that lines up perfectly with the orientation of the crack.

Given that the gearbox input flange broke in both cases we could model that up since its not very complex and make a determination how high the forces would need to be to get stress levels high enough that one could fail in the kinds of service durations that we had seen.
 
This video illustrates some of the kick back issues on starting that was one of the complaints against the stock ecu.
https://youtu.be/lXb_N60vAHs

The 800 hour engine has shipped in MA, now just a question of how long it takes to thread through the snow disaster that is the I80 corridor.
 
Pictures of the failed 134 hour SB compliant drive flange

See attached the first sample failed gearbox input flange. This is the 134 hour unit that had all the SB compliant version parts in it and lead to the total loss of the customers airplane in a forced landing.
I think anyone can understand why it is important to get both flywheel side and gearbox side drive flanges inspected for cracks as soon as possible if you own one of these.
 

Attachments

  • IMG_0373.jpg
    IMG_0373.jpg
    490.9 KB · Views: 200
  • IMG_0374.jpg
    IMG_0374.jpg
    412.5 KB · Views: 170
Last edited:
See attached the first sample failed gearbox input flange. This is the 134 hour unit that had all the SB compliant version parts in it and lead to the total loss of the customers airplane in a forced landing.
I think anyone can understand why it is important to get both flywheel side and gearbox side drive flanges inspected for cracks as soon as possible if you own one of these.

Scary stuff if one is not over the flat lands of Kansas.
 
Adding a few more pictures. The first showing the gap between the input gear and the bearing on the inside of the gearbox, where I plan to try to fit a bearing splitter to support the bearing inner race while pressing the input shaft, with gear out the drive flange.

Also a wider angle shot showing how the engine slung the broken off ear around and "machined" out the back plate flange until the guibo finally let go. It seems fairly obvious that the crack initiated at the outer diameter of the boss in the center of the drive flange where the bending load was concentrated. Likely the outer crack originated where the boss is located on the ear to hold the drive pin, there is another sharp change in section there.

Fundamentally the material looks brittle. Based on the coloration this may have been caused by the welding process that Jan used to "further improve" the drive flange.
 

Attachments

  • IMG_0375.jpg
    IMG_0375.jpg
    161.5 KB · Views: 176
  • IMG_0376.jpg
    IMG_0376.jpg
    278.3 KB · Views: 227
Gyroscopic Loads

If I was given the responsibility for ground testing the new PSRU plate, I'd mount the engine/propeller assembly on a Tilt-Yaw table to check (both the old) and the new plate's response to the gyroscopic loads.

There should be digital measuring "cells"? to check relative motion created by the gyroscopic load testing

Gyroscopic forces are easily calculated. The PSRU plate can simply be statically load tested with the calculated gyroscopic forces applied the the propellor drive flange. No need to design and build a complex and expensive gimbaled test stand for this.

Skylor
 
Last edited:
Granted, I didn't consider that. On further "associative memory" reflection, recalling the Lockheed Electra in-flight failures there could be a "whirl mode" issue?

This is a "just thinking-what if" observation. Not saying it's necessary.

Hoping you have an additional cogent insight.

:D
 
Back
Top