RAFSA 2000 accident 8-8-15


Now, all of the above assumes that PN39's curious blade R (radius) of 73" is correct, vs. something closer to the 156" (13 feet) length of the RAF rotors. In fact, for Eben's "radius of the circular path" shouldn't R be measured from the blade tip to either the hub bar center, or at least to the ¾" bolt's underhead (since we're discussing its load)? How did RAF come up with a mere 73" for a 13'-15' rotor system? Perhaps Vance can explain that.

Kolibri

The reason the total length of the rotor blade is not used to calculate the centrifugal load is the entire mass of the rotor blade is not at the tip.
 
Understood, thanks, then I'll let stand my figures with their equation.

Have you looked at PN39's formula for hub bar stress?
Something seem off to you there?
 
Understood, thanks, then I'll let stand my figures with their equation.

Have you looked at PN39's formula for hub bar stress?
Something seem off to you there?

The formula looks good enough for a guess.

I feel a strain measurement would be the best way to determine the actual loads.

In my opinion your focus on centrifugal force and tensile strength of the bolt is misguided.

I feel tensile load is not what broke the bolt in the Oytmayer accident.

I don’t know what broke the bolt in the South African accident and do not have enough information to speculate.

I suspect a blade strike is what began the process of the bolt failure in the Oytmayer accident.


On a separate subject I feel you may be misunderstanding the effect of G loading in a turn.

Speaking in general terms a gyroplane rotor speeds up with additional load.

For example The Predator has 30 foot 8.5 inch chord blades similar in size and chord to an RAF and weighed in the neighborhood of 1,100 practicing air show maneuvers. She typically flies straight and level around 315 rotor rpm at that weight. In a steep turn it is not unusual to see around 450 rotor rpm. The highest loading I have seen on a G meter is a little over two Gs for a total effective weight on the rotor of around 2,200 pounds. These are rough numbers.

The RAF 2000 has a gross weight of 1,536 pounds and is approved for a sixty degree banked turn (2 Gs) for a load on the rotor or around 3,072 pounds. I feel it would be reasonable to expect the rotor speed in a 2 G bank to be higher than the limit quoted of 395 rotor rpm. In my opinion the pilot would need to reduce his weight of angle of bank to stay below 395 rotor rpm.
 

Attachments

  • Gs.jpg
    Gs.jpg
    51.8 KB · Views: 0
Last edited:
Inspecting and Replacing bolts in the hub bar at 500 hrs is reasonable but RAF doesn't share the torque spec. for the bolts . You also can't buy these bolts at Aircraft Spruce
or anywhere else that I am aware of. They may list them but when you call looking for a couple bolts they will be less than helpful . Before removing the bolts be sure you have new ones in hand but you still won't have the torque spec for the hub bar !
I'm certain these bolts aren't torqued to the max spec. off a reference chart . They have to be tight but still adjustable . There are ways to get it done but if you don't fully understand
What you are doing you will have problems.
Just thought I would put that out there for anyone that may be thinking of changing there bolts themselves . (Just food for thought.)
 
My bolts have washers underneath them,its just that my winglets where the rotor blades

attach are the older style,no cracks showing .I feel that the safety record of the RAF hub-

-bars are actually quite good,certainly not bad enough to worry about.


best regards,
 
The formula looks good enough for a guess.

I feel a strain measurement would be the best way to determine the actual loads.
Vance, it seems that PN39's "good enough for a guess" formula is about all the computation RAF Canada did to justify their design. I've not heard of any strain measurements from them. It calls to mind Chuck B.'s phrase . . . "by gosh and by guess".


In my opinion your focus on centrifugal force and tensile strength of the bolt is misguided.

I don’t know what broke the bolt in the South African accident and do not have enough information to speculate.
Centrifugal force and bolt tensile strength are all we can work with for now (hence my unavoidable focus), and I agree that bending loads are probably a large factor in ZU-RHO.

I am reassured by my source that the rotor blades had not experienced any impact or flap after the rebuild/prior the August failure. Furthermore, the pilot owner did not loosen the hub bar nuts for any reason, such as to readjust pitch or tracking (which was satisfactory from RAFSA).

However, my source did wonder aloud to me if the original bolts from the previous rollover were reinstalled in Upington by RAFSA. Such a shocking possibility had never occurred to me, and he is not actually making the accusation (nor am I). I'd have a hard time believing such could be true, even though it would certainly explain a lot for such a <50 hour bolt failure.



On a separate subject I feel you may be misunderstanding the effect of G loading in a turn.

Speaking in general terms a gyroplane rotor speeds up with additional load.

I feel it would be reasonable to expect the rotor speed in a 2 G bank to be higher than the limit quoted of 395 rotor rpm. In my opinion the pilot would need to reduce his weight of angle of bank to stay below 395 rotor rpm.
Sure, I understand and don't dispute that, but my purpose was to recalculate within RAFSA's operating limits so that I wouldn't be accused of employing an unfair example of overspun rrpm. A conscientious RAFnaut would keep within those limitations, but one could get caught in an updraft at the worst timing during a steep turn.


______________
I'm certain these bolts aren't torqued to the max spec. off a reference chart . They have to be tight but still adjustable . There are ways to get it done but if you don't fully understand
What you are doing you will have problems.
Boots, great point, thanks for saying so. The clamping force is probably within a very narrow "Goldilocks" range, else the hub bar/winglet join would be a mere hinge, or the pitch could not be adjusted at all. You're right: we don't know RAFSA's torque figure.


_______________
My bolts have washers underneath them,its just that my winglets where the rotor blades attach are the older style,no cracks showing .I feel that the safety record of the RAF hub--bars are actually quite good,certainly not bad enough to worry about.
eddie, my friend, if "Santa" gave you for Christmas an installed set of SC rotors/bar/head/cheek plates, would you accept it? Or, would you direct Santa to install these parts on another's RAF?

I understand one's emotional inertia to stick with something that you already own, and hasn't not (yet) hurt you. I also understand out-of-the-blue warnings in life. Even according to RAFSA you should have replaced your bar/rotors 100 hours ago. So, please, don't miss Kolibri's cue, and let's not whistle past this graveyard. You've fully enjoyed your money's worth from the existing rotors. Step up and install something proven for 1000's of hours vs. (sort of) for only a few hundred.


_________
What I've so far been left with is that the RAF hub bar assembly is adequate only for moderate flying with (vis-a-vis better gyros) frequent parts replacement. Other RAF subsystems, such as their choice of small ¼" spherical rod ends on the pitch control tube, tend to bolster my view.

YMMV, but you'll be sorry if it does . . .

Regards, Kolibri
 
I believe that the RAF rotor system has a good safety record,good maintinence and

inspections are required on all aircraft,if we start worring about every thing that we don't

like then about the only thing left to do is quite flying ! I have been flying for 47 years and

have heard the doomsday naysayers that are at every airport in the country,these guys

seem to enjoy instilling fear into others,over the years they have badmouthed just about

everything that flys or has flown.I truly believe that some of these guys are really afraid

of just about everything,how they manage to get away from the recliner and out into the

real scary dangerous world is beyond me.Flying at best is a calculated risk,you just have

to step up, evaluate your flying machine yourself, and then decide.

Your are so worried about a 3/4" certified bolt that you have failed to take into account

all of the other things that could break,for example what about the 1/2" bolt in tension

that holds the rotor head on to the mast.The point is that you and only you can decide if

your gyro is safe to fly,Iam able to decide for myself about my Gyro,I can't speak for you

and have no intention to,so please leave me to make my own decisions regarding my gyro

I have a pretty good record so far. Think about it,if flying was 100% safe everbody would

be flying,the skys would be so full that you would have to file a flight plan months in

advance to fly for 15 minutes in the pattern.






Best regards,
 
as you can't inspect and test everything ...eventually it comes down to
do you trust the concept , the design, the meterials , the manufacturing process and testing of the meterials, the build, the complience to the designers spec, the environment in which you use it and the way you use it
The on going maintenace you do, on going support from the manufacturers and component suppliers.
at the end of the day it's all about trust
do you trust all the above, do you trust yourself the maintain and rebuild and modifi and select the correct, tested components.
99% of all suppliers want their customers to be happy and safe so that word gets around that theor products are good ... so they sell more and make more profit !
It's all about trust.... if you have lost trust in this safety critical sport, then simply look elsewhere or try fitting a BRS
 
Thats right on peter,trust becomes stronger when the product has a proven record of time

and hours in service.


Fergus, thanks for the kind words of support.






Best regards,
 
I'm going to keep flying mine too, they track well, no stick shake and compared to aluminum blades, are real easy to clean, Try eco friendly Awesome cleaner on your composite blades. It can be bought at the dollar store. I haven't found a safe cleaner that works as well. Eddie, when was the last time you pulled 3.5 Gs at gross weight of 1540 lbs? (can't be done). IMO, this 3/4" bolt is plenty over built. If I needed a new set of blades, I would seriously consider another set from RAF. My experience buying parts from RAFSA, has been wonderful. Eben and Yolande have always been great to work with, which for me is worth its weight in gold. I know that SportsCopter makes a great product as well, and I'm sure would make an excellent replacement. Nice to have choices.
 
eddie, no need to wrongly generalize me into your described camp of Nervous Nellie Naysayers. I'm also a pilot, who owns an E-AB gyro, and an RAF2000 to boot. However, I will not fly on a rotor system with a mfg. recommended service life of only 500 hours.

You're 100 hours past PN33, and you've the old winglets (which RAF Canada voluntarily redesigned because of inherent stress risers).

Thus, I hope that you never experience this:



attachment.php




It is amazing to me that there hasn't been more problems. Not creating stress risers is something most people that make parts learn early on. A smooth transition is very important. One of the down sides to aluminum alloys is it is how it fatigues. There is generaly very little warning.

I would like to point that these kinds of cracks are very hard to see in the early stages. From the time you can see the crack to a seperation of the part is generaly quite short.


There are actually three stress raisers on this part, conveniently placed close together to help the part break more easily. They are:

(1) A large change of section. The yoke that contains the pitch-change bolt is very rigid in up-down bending; the milled-out area that receives the straps is much thinner and more limber, forming a fulcrum that concentrates the stretching and compressing during bending.

(2) A sharp corner. People have already commented on this one.

(3) A hole. Putting one of the blade strap holes near the #1-#2 stress concentrators makes the problem worse. It both reduces the section area and creates its own stress concentration.


If they do not have the ability to design a component as critical as a hub bar that can last longer than five hundred hours how in hell can you have any faith in their five hundred hour life time?

How did they arrive at that number?


i m just amazed RAF didnt put some hub bars rotating during a month (500 hrs) with fake blades, por balanced, shaking the mast, just to see for REAL if there is a source of failure, it s not a matter of cost for them but of will...

i think also an approximative design can be "forgiven" on an amateur construction but, if you sell as an official "aircraft manufacturer" you must have a serious quality/design control, or you are not honest.


As a fellow gyronaut, with some small experience, and with access to engineering support I can make suggestions and comments.

Comments:

1. There are many hundreds of RAF aircraft flying.

2. There have been few reported cases of RAF rotor systems failing.

3. Sometimes a design is intolerant of dimensioning tolerance stack up.

4. After Paul's post I asked my staff to look at the RAF hub bar design.

5. We found tolerancing issues that may be resolved by operators.

6. We found that if the tolerance stack up between the blade straps and the hub bar interface is slightly off, that the blade strap can cut into the hub bar exactly at the point where the hub bar is milled to a thinner profile.

7. This interference can result in the blade straps cutting a small "V" notch in the hub bar at the worst possible spot.

8. We will be issuing a notice to all our mod kit customers still flying the RAF system making the following suggestions.

Suggestions:

1. Remove your rotor blade system from your aircraft.

2. Separate the hub bar from the blade straps.

3. Carefully examine the imprint where the inboard end of the blade strap touches the hub bar.

4. Determine if there are any nicks, cuts, or gouges at this point.

5. In any case, you can, looking at both the top and bottom blade straps, round off the blade strap corners with an approximate 1/8 inch radius.

6. Additionally, on both the top and bottom blade strap inboard ends, round off, with about a .06 inch radius, the portion of the blade strap in contact with the hub bar.

7. You may also wish to carefully blend the abrupt radius on the hub bar from where the hub bar transitions from the thin section and begins to slope up to the thicker section.

8. I strongly urge you to buy a cheap dye penetrant set and dye pen the hub bar at the blade strap hub bar interface.

Now in english: If the blade straps are just a smidgin too long and the sharp edge of the blade strap digs into the hub bar you may generate a stress riser that will initiate a propagating crack.

This message in no way constitutes an engineering suggestion from GBA. This message is not approved by GBA. This message does not indicate that following the steps suggested by me, as an individual, will ensure additional safety.
 
Last edited:
Well it sounds like what you are saying is that the hub will break off before the bolt fails,

which is it.

How many hubs have cracked?





regards,
 
Well it sounds like what you are saying is that the hub will break off before the bolt fails,

which is it.

How many hubs have cracked?

eddie, the old winglet is historically more prone to failure than the NAS washered bolt. If it fails on you, and you live to tell the tale, RAFSA will scold you for flying it past PN33's 500 hours.

Paul's failure was not the only one.
For example, Chuck Ellsworth posted in that thread:


In the early ninties I was booked to fly with Duane Hunn at Brookville, I did a walk around and found a cracked hub bar.

Duane phoned RAF and they sent a new hub bar. Of course I never went flying in it after that.

There was zero communications with the RAF owners until I raised proper hell about the coverup. It was a long time ago but I'll never forget the pressure that was put on me to shut the f.ck up. It would not take much research to find out how long it was between RAF knowing they had a flawed design in the hub bar and their sending out a letter informing everyone........

eddie, think of it this way: you're flying winglets that not even 2004 RAF Canada continued to believe in (else they wouldn't have redesigned it). That old straight-cut/narrow winglet was discontinued over a decade ago.

However, if you have more faith in it than the manufacturer, then . . . good luck!

 
The history I have access to (NTSB accident reports) has one of the old hub bar bolts failing on the ground and none of the hub bars failing in flight on an RAF 2000. The South African accident report has one of the new hub bar bolts failing in flight,

As I read the accident history of RAF 2000s it appears to me that poorly trained pilots are a much greater hazard flying an RAF 2000 than anything to do with the hub bar.

I have been impressed with the safety record of the RAF 2000s flying in South Africa when compared to the USA RAF 2000s and I feel it is in a large part because of the quality of the training.

If I owned and flew an RAF 2000 and I felt that the instructor I used did not do a good job with my gyroplane instruction I would seek out better training particularly if the instructor’s students had what appeared to be a high accident rate.

Knowing what to do with the aircraft controls in an uncommanded descent seems a critical part of gyroplane instruction in an RAF 2000.

I feel that learning to do a proper preflight inspection and learning to make good aviation decisions are important parts of training.
 
Its like Vance has said over and over,its all about maintainence and good aviation

decision making.

I suspect that there a lot of RAF's and gyro's in general with several thousand hours on

the rotor blades/hubbars without incident.




Best regards,
 
Vance, in this thread we're not discussing RAF pilot training. Also, the cracked Duanne Hunn hub bar which Chuck Ellsworth described would not be in the NTSB accident reports (which, by nature, are incomplete in regards to overall parts failure which had not -- through miraculous timing -- caused an accident).

Yes, RAFSA has apparently reduced student and low-time pilot accident rates with either longer or better training. However, now that the South African fleet of RAFs has accumulated some higher airframe hours, any latent engineering issues will begin to surface, and such seem to have begun.

For example, besides this thread's ZU-RHO, there is the March crash and double fatality which has all the earmarks of a trim or control system failure:


Gyro down in South Africa - Fatal
http://www.rotaryforum.com/forum/showthread.php?t=42850
According to the witness, the aircraft took from runway 26 and climb without incident. On downwind at about 200 feet AMSL the aircraft pitched to a nose down attitude where after it crashed into a bushy terrain and burst into flames.

ZU-DVB - CAA reference #CA18/2/3/9426


________
Its like Vance has said over and over,its all about maintainence and good aviation decision making.
OK, eddie, then ask Vance if flying provenly underengineered, stress risered (and thus discontinued) winglets 100 hours past their service R&R is an example of what he meant by "maintainence and good aviation decision making".

Or, put another way: How many more hours past a meager 500 hour service life do you plan on flying a discredited and effectively recalled rotor system? (It's more of a rhetorical question, for your own reflection.)

Regards, Kolibri
 
news on the failed bolt in ZU-RHO

news on the failed bolt in ZU-RHO

There's been some light shed on this over at microlighters. Here are the highlights:

RAFSA seems to suspect an undertorqued ¾" bolt in ZU-RHO caused its premature failure <40 flight hours after factory replacement.

RAFSA suspected that up to 8 hub bars may have had undertorqued bolts from the factory (Wherever that is: Canada? Upington? Who knows?

They thought perhaps an out-of-calibration torque wrench was the cause, but claim that subsequent testing showed them within spec. (RAFSA did not, however, explicitly rule out human error during assembly.)

During RAFSA's recall of these 8 hub bars (per their Product Notice 51 dated 14 August 2015, which I cannot find on their website), they charged the owners ~$81, according to Eben:


. . . as a precautionary measure, recalled and replaced all of the bolts that were under our suspicion. We did not use any of the bolts that came back, we removed them, inspected and re-did all of the hub bars and did not charge for any of the hardware we replaced. We charged for the re-inspection and re-certification a total of R1140 just as a precaution.

You assume we are at fault, however we have no obligations to replace or change anything that is in the field unless there was an obvious reason for the change and in that case you would order the parts and replace as necessary. We were being proactive and didn't want to take any chances. So we called them back and checked all the torques and for your information there were no bolts found out of spec, as a further precaution the bolts were removed and we sent them to the manufacturer to have them X-rayed.

Regarding the overly beveled washer, Eben had this to report:

The washer provides a relief for the radius under the bolt head you keep saying is not there. We install a beveled washer because the hole that is drilled is not chamfered because you can’t reach in with a chamfer tool so we cut it into the washer for the radius that is under the bolt head.

The "hole that is drilled is not chamfered because you can’t reach in with a chamfer tool" must be the inside of the hub bar for the bolthead. But, why such a huge bevel in the washer? It just isn't necessary for the imperceptible radius of the bolt underhead. A standard hardened washer would not seem to crash into it, according to two A&Ps I've queried.

____
Some folks apparently still need to grasp the detrimental nature of RAF2000 hub bar "joint", in which nearly 5 meters of rotor/winglet twists, bends, and pries on the ¾" bolthead — making the underneath hub bar face act like a bottlecap opener.

Below are photos of my own old RAF parts. The ¾" bolt is on top of its winglet, showing its position inside. Notice the significant circumferential scoring of the bolt shank at the winglet mouth (i.e., next to the hub bar). The cadmium plating is gone, and the scored rings can be felt with a thumbnail. (This vigorous dynamic also ovals the mouth of the winglet, as I've previously mentioned with measurements of my own parts.)

file.php


file.php


Where has RAFSA published data for those forces? Tensile strength (which Eben continues to harp on) isn't everything here. If it were the weak link then the bolt would be failing at the threads, not at the head.

Now view the two photos of the inside of the joint (i.e., the mating ends of the hub bar and its winglet). Notice the visible ring striations about the bolt hole of each part. This is additional evidence that the winglet ends are twisting back and forth against their hub bar.

file.php


file.php


As your RAF's 5m rotor moves like a torsional cheater bar in flight, the 6061-T6 aluminum parts are squished and deformed under the bolt head and nut washers. This reduces bolt torque as was set at the factory (Canada? Upington? Who knows?), which allows greater movement and prying, further reducing torque, etc. This is a cascade which can only eventually lead to bolt fretting and then a head separation. (Best theory on ZU-RHO so far: that particular bolt was undertorqued, and probably by human error vs. a bad wrench.)

It is good that RAFSA inspected their torque wrenches and had suspect bolts sent out for X-ray analysis. However, RAFSA's obvious blind spot is that their hub bar design itself is grossly flawed. NOBODY else in the rotorcraft world employs singleton primary hub bar bolts in tensile, much less in a hub bar already weakened from the start with huge cutouts.

Why the hell is RAFSA so stuck on their kludgy, low-time, and risky design? Why are safety- and quality-conscientious RAF owners forced to go elsewhere for a high-time and rugged hub bar/rotors/head?

An owner's pre- and post-flight check of the ¾" bolts can only confirm that they are still present.
You won't be able to confirm factory torque, or a nondeformed aluminum face underneath the washers.
You won't know if the bolt shanks are becoming scored.
And if a bolthead is just about to pop, you won't know that in advance, either.​

If you're still flying on RAF's rotor system, I'd very quickly become very informed about the issues.

Safe flying!
Kolibri


file.php
 
Last edited:
Igor Bensen designed a trouble-free version of this joint for his metal rotorblades more than half a century ago. The pivot around which you could adjust the pitch was a little rollpin. It carried no loads in use; it just held the blade and hub in alignment once you loosened the bolts and pitched the blades up or down. The tensile loads were carried, very ably, by four AN-6 bolts arrayed around the roll pin. Since these bolts were radially distant from the center of hub, any bending (coning) loads were converted into pure tensile loads on the bottom pair of bolts.

Various other metal blade-makers happily copied Bensen's idea. The nickname for the joint was the pitch-adjusting "knuckle." The only improvement in the original knuckle design that would be advisable would be to build in a coning angle to reduce the bending load carried by the hub bar.

It's pretty basic engineering to make the portion of a cantilever beam that experiences the largest bending moment (the middle, in the case of a 2-blade rotor hub bar) the thickest. Stacks of leaf springs, of course, are designed this way, yet the RAF hub reverses the practice. It's thinner in the middle.

I don't understand the need for this complex and troublesome (and expensive in machine time!) re-invention of such a simple design element.
 
Interesting backstory, thanks, Doug. Would a photo of that "knuckle" be handy for you to post?

I don't know why RAF continues with their "all-wrong" design, either. They got a "freebie wake-up call" in the non-fatalities of ZU-RHO's crash. (So did all the RAF owners!) We can all hope such a failure never happens again, but if it does then nobody can claim astonishment.

Regards, Kolibri
 
Top