broken drive shaft

AKSNOWRIDER said:

mod, what I am saying is if the glued joint holds as far as the glue goes, the joint will hold and the tube will fail somewhere else(most likely twist up in the center as the truck driveline did)..ideally if you have a big powered sled(turbo bigbore, i think a 11-12 steel shaft would be the best bet) on a stocker..well I would get some milage on it(with anything I think I would need for a tow just in case in the truck) and inspect it closely on every ride..once it got miles on it..I would gain trust in it..but thats me...

Click to expand...

Gotcha, thanks for clarification.

I think you're comparing apples and oranges, using the example of a long automotive driveshaft where the input and output are on opposite ends of a long tubular...compared to the hexagonal shaped Pro driveshaft where the output is spread across the drivers.

I'm not an engineer, but I would bet that if enough rotational forces were applied, the input spline would fail long before the hex shaped extruded shaft with the internal "X" gusseting (this is a VERY strong design).

Again, just my opinion.

mod..technically speaking round is stronger due to no edges for a crease to form on..but the extruded tube does have internal walls as well so theoretically it is probably stronger in this instance..now as far as length goes, the longer a tube/shaft is, the more it can flex, yes, but a shorter tube/shaft is more brittle...as an example, a 4x4 truck, spinning all 4 wheels and catch traction on the frt tires..it will snap the short axle every time unless the long axle has an obvious flaw, but do the same on a rear axle(both shafts are almost same length) and either axle can be broken...just what we have seen over the years of failing stuff(read heavy rt foot/thumb)

Why would they make something like that. ? How many have broke already.

i think we should chill and sniff the glue...

Plug welding

byeatts said:

Welded. isnt the shaft aluminum and splines steel?

Click to expand...

That was my thoughts... good job!!!

Like I tell my buddy's just because it's light. Iam not pulling you off the mountain with my yamaha

Wow! Lords 406 failed at 1980 psi shear at 180 deg F. Considering full stress reversals every revolution and a factor of safety of 2.5, means that the glue is good for about 100 psi. The shear from the red hot belt pulley, will cause about 1120 psi pullout shear on the lower glued joint. The glue also creeps under load, and I will bet that it especially creeps when it is red hot from the belt pulley heat. Therefore, the glued joint is overstressed by about 1000%. In otherwords, if it is still holding together for you, you are riding on a prayer. Perhaps that mechanical engineer for Polaris, would have been a good investment after all. Which brings to mind one possible reply that came from R&D. "We don't need no stinkin Engineer!" Well, as it turns out, Polaris did need an Engineer. It will be very tough to hang this on the dealer and expect them to foot the bill for Polaris's apparent gross negligence on an untested driveshaft.

http://www.ambercomposites.com/downloads/datasheet/lord-403---406--410-with-19-19gb-march-06.pdf

whoisthatguy said:

Wow! Lords 406 failed at 1980 psi shear at 180 deg F. Considering full stress reversals every revolution and a factor of safety of 2.5, means that the glue is good for about 100 psi. The shear from the red hot belt pulley, will cause about 1120 psi pullout shear on the lower glued joint. The glue also creeps under load, and I will bet that it especially creeps when it is red hot from the belt pulley heat. Therefore, the glued joint is overstressed by about 1000%. In otherwords, if it is still holding together for you, you are riding on a prayer. Perhaps that mechanical engineer for Polaris, would have been a good investment after all. Which brings to mind one possible reply that came from R&D. "We don't need no stinkin Engineer!" Well, as it turns out, Polaris did need an Engineer. It will be very tough to hang this on the dealer and expect them to foot the bill for Polaris's apparent gross negligence on an untested driveshaft.

http://www.ambercomposites.com/downloads/datasheet/lord-403---406--410-with-19-19gb-march-06.pdf

Click to expand...

Who,
You are more than welcome to post whatever you want, but all those numbers you keep throwing around like they mean something are just numbers you pulled out of your azz and don't mean squat to anybody. I don't know your personal background but you have posted the same stuff on every thread about this issue multiple times and IMO it is all useless, baseless data.

Is there an Issue with this assembly? Yes to an undetermined quantity of them, there seems to be an issue. Either with how is was assembled or some other manufacturing variable, most likely. I want it to work and think it will, and a few of you have likely glimpsed at least a portion of the underlying factor. I will refrain from my speculation but as soon as this driveshaft or it's replacement proves viable I will be ordering them in quantity for my 3 current builds and projected future builds. Even if I need to disassemble, re-work, and re-glue brand new ones with my own personal preparation of the gluing surfaces. (I reviewed the posted pics again and there is a glue channel as my original post proposed, I didn't catch that on my phone earlier, so I edited my response) You can see from the pics posted that the contact area of the mating surface is stepped, with the largest next to the shoulder (to center the extrusion) and reduced where the injection ports are (I didn't catch the step back up in size forming the channel for the glue to flow around the hex and then squeeze both ways to the inboard and outboard). The injection dispenser is likely timed or pressure triggered so they get a consistent delivery of glue. As any shaft I purchased would not be covered under any warranty, if I had any doubts about it I would disassemble it and add my personal touches to the mating surfaces. A good cleaning and sand blasting with a mild abrasive or acid etch and re-glue the stub in and give it a good stress test. "Ride it like I stole it", which I do anyway. I'm not scared, I've broken my fair share of stuff and I'm still here to tell the tales they have inspired.

To paraphrase Edision "I did not fail, I have just proven X number of ways that will not work" We are close, the Devil is in the details.

Whoisthatguy....

I can see that you have some serious technical background and I can respect that. I'm glad you are part of this dialog.

There are some assumptions that you are making that may not build an accurate picture for the other readers...The figures you are quoting are for lap shear testing, which has a contact area of about 0.5 sq-in. And is only in linear shear. As you can see in the vid posted above... the aluminum failed before the glue at 16,000 lbs. pull.

Also, with all the snow and air flying around in the tunnel, and the surface area of the shaft...there is no way that the sprocket is approaching the 350+ degrees required to weaken the adhesive bond. You would be seeing the grease cooked out of the shaft bearing before that happened.

I think that is is Hyperbole and irresponsible to say that the shafts are "holding together on a prayer"... even more to say "gross negligence on an untested driveshaft"

I personally, as well as 6 others, pounded on these sleds for a day with no issues... and a log was jammed in the tunnel for 30 min.

Those same sleds were rotated into the demo tour.

The demo fleet was pounded by hundreds of others at the end of last season... and no shaft failures. (we would have heard about it on these forums...it's a small world out there in "demo land")

The design, IMO, is NOT faulty... IMO, the adhesive application in the production of the shafts may have been "off" on some units... And that would be a quality control issue, not a design issue.

If there were 100 examples of failed driveshafts... which there aren't... then I might be more worried... too early in the season to to speculate.

I like Modsledrs simple description.

If we assume that a certain amount of glue is required to meet the structural engineering requirements of the application, and there are a certain % of units that are "under glued"...then the glue fails first, then the insert spins and blows out the sides.

One of the first pics posted of a failure in snobyrd's post showed a shaft where the insert had spun, but had not yet broken out.

Click to expand...

Snobyrd's shaft... spun... ballooned out... but not torn.
I'm thinking adhesive starvation or surface prep.. but not defective design.

Mountainhorse, What is the difference between directionally loaded lap shear and linear shear? None I used manufacturer's test data which indicated the Failure load at 180 degrees F, not 350 degrees F. With 1000% overstress due to the belt force alone, it would be impossible for this joint to have been properly designed. Here is some ballpark numbers since I do not have a sled in front of me to take measurements.

(120 ft lbs torque x 12 in/ft) divided by 2 inch lower pulley radius = 720 lbs belt tension.

720 lbs tension x 2 inch moment arm to center of driveaxle bearing = 1440 lb-in moment

1440 lb-in moment divided by 2.5 inch fence post end diameter = 576 lbs tension on lower glue face of 6 sided polygon

576 lbs tension divide by (0.5 inch x 1 inch glued surface) = 1152 psi pullout shear on glued surface

1980 psi failure load divided by 2.5 factor of safety for static loads = 792 psi working stress

792 working stress divided by (10 factor of safety for more than 500,000 full stress reversals) = 79 psi allowable glue shear

(1152 psi divided by 79 psi) minus 1.0 = 13.6

13.6 x 100 = 1360 % overstressed

8 in drive sprocket diameter x pi = 25.1 inches

(5000 miles x 5280 feet per mile x 12 in per foot) divided by 25.1 inches = 12.6 million full stress reversals

The math cannot be denied. The numbers can be fined tuned but are not even close to the 100% overstressed number. Consequently, it is fairly clear that Polaris made a huge blunder.

As people in the know have pointed out, these particular driveshafts were not on any demo sleds. Therefore, there was no testing by people who did not work for Polaris. The people that may have tested it, would not be able to talk about driveshaft failures.

:noidea: :shocked: :doh:

mountainhorse said:

Whoisthatguy....

I can see that you have some serious technical background and I can respect that. I'm glad you are part of this dialog.

There are some assumptions that you are making that may not build an accurate picture for the other readers...The figures you are quoting are for lap shear testing, which has a contact area of about 0.5 sq-in. And is only in linear shear. As you can see in the vid posted above... the aluminum failed before the glue at 16,000 lbs. pull.

Also, with all the snow and air flying around in the tunnel, and the surface area of the shaft...there is no way that the sprocket is approaching the 350+ degrees required to weaken the adhesive bond. You would be seeing the grease cooked out of the shaft bearing before that happened.

I think that is is Hyperbole and irresponsible to say that the shafts are "holding together on a prayer"... even more to say "gross negligence on an untested driveshaft"

I personally, as well as 6 others, pounded on these sleds for a day with no issues... and a log was jammed in the tunnel for 30 min.

Those same sleds were rotated into the demo tour.

The demo fleet was pounded by hundreds of others at the end of last season... and no shaft failures. (we would have heard about it on these forums...it's a small world out there in "demo land")

The design, IMO, is NOT faulty... IMO, the adhesive application in the production of the shafts may have been "off" on some units... And that would be a quality control issue, not a design issue.

If there were 100 examples of failed driveshafts... which there aren't... then I might be more worried... too early in the season to to speculate.

I like Modsledrs simple description.


Snobyrd's shaft... spun... ballooned out... but not torn.
I'm thinking adhesive starvation or surface prep.. but not defective design.

Click to expand...

--The end caps should be extended at least another 1 inch into the shaft in order to spread out the load and 3x thicker metal driveshaft also. Engineers need to recalibrate their computer models. These things are failing with engine power output at less than 120 HP- and a lot less torque at altitude. Pump gas turbo or BB, forget it.

--The purpose of the glue is just to hold the driveshaft assembly together as one piece, it is not structural or designed to carry any load.

--If designed properly, RKT clamps will have the same effect as having a thicker wall driveshaft at the location it is clamped to. The clamp should help prevent endcap spinout and also help transfer load from endcap to shaft. The clamp will need to be tight at each point area of the hex shaft as these are the areas that have actual metal to metal contact area. The clamps should be a descent way to have some piece of mind until a properly designed factory unit is available.

But a factory unit is availabe now. So why waste your time conconcting an untested bandaid fix, that can likely only compress the corners of the 6 sided polygon driveshaft, without applying any confining forces to the flat surfaces where the failure also occurs. Furthermore, the end cap is pulling out of the aluminum center section. Your speculation that the glue is not acting in a structural capactiy, has no factual basis. Your alleged fix will not likely prevent the end cap from pulling out of the aluminum center section.

Dose any know the manufacture date's of the the units that have failed ?
or the serial #'s, Remember the recall on the earlier A arms, my dealer told me that it was due to the bonding agent not sticking to the painted shaft. Well anyone should know you have to prep an area being glued, are we looking at the same thing with the drive shafts? If so hopefully it will turn out to be a limited amount that have the problem.

Does anyone know how many cases we are looking at. I myself have a 13 pro and would like to know

whoisthatguy said:

But a factory unit is availabe now. So why waste your time conconcting an untested bandaid fix, that can likely only compress the corners of the 6 sided polygon driveshaft, without applying any confining forces to the flat surfaces where the failure also occurs. Furthermore, the end cap is pulling out of the aluminum center section. Your speculation that the glue is not acting in a structural capactiy, has no factual basis. Your alleged fix will not likely prevent the end cap from pulling out of the aluminum center section.

Click to expand...

--A clamp should support the driveshaft exterior and help prevent spinout, keeping the endcap and driveshaft aligned thereby preventing misalignment and separation.

--The endcap does not seem to be "pulling out" laterally, but rather spinning inside of the driveshaft, misalinging, and separating.

--If the endcaps were another 1 inch longer inside the driveshaft and the exterior metal of the driveshaft was thicker, an interference fit would probably work fine with no glue needed. Probably add $3 to the cost of manufacturing the sled and then subtract $3 dollars since no glue needed so OEM fix would probably be a wash. I just do not like the overall design, period. External location of driveshaft where endcap ends and internal driveshaft support begins is also a weak spot.

--A 12 shaft is a great option for hundreds of dollars, but a OEM redesigned or 12 shaft replacement at no extra charge would be better

whoisthatguy said:

But a factory unit is availabe now. So why waste your time conconcting an untested bandaid fix, that can likely only compress the corners of the 6 sided polygon driveshaft, without applying any confining forces to the flat surfaces where the failure also occurs. Furthermore, the end cap is pulling out of the aluminum center section. Your speculation that the glue is not acting in a structural capactiy, has no factual basis. Your alleged fix will not likely prevent the end cap from pulling out of the aluminum center section.

Click to expand...

Our kit does apply forces to the flat areas as well as the corners..

But I will agree, installing the steel 11-12 shaft is the better route BUT it is also the most expensive route and shafts are back-ordered

so, WHY?? because it is very inexpensive, it may save you a cracked head and a broken sled, and because getting the old shaft installed is very expensive and not currently available...

The end cap can not pull out unless the outer extrusion wall fails..

These are two different metals , how can they actually bond ??

Kelsey, just so you know..I just tried ordering 100 2012 driveshafts..they show available and in stock..so they are not backordered..........

AKSNOWRIDER said:

Kelsey, just so you know..I just tried ordering 100 2012 driveshafts..they show available and in stock..so they are not backordered..........

Click to expand...

This is not true unless you are some wizard or something.

Literally just got off the phone with my dealer and the 2012 driveshafts show a backordered date of DECEMBER 21st.

I don't know where you can get 100 of them, but maybe you should and sell one to me???