Drive Shaft Question

Hello

If you are able to advise the detailed usage of the driveshafts it may result in a better reply.

1. Engine size - HP or kW
2. Engine maximum torque
   
3. gearing ratio
4. wheel (tyre) diameter
5. Other - Advise please....

That way persons experienced in vehicles could give you more complete advice.

Kind Regards....

Hi and sorry for the lack of info , the car is a Toyota Aygo but I fitted a Starlet GT 4 ef-te 1.3 ltr turbo engine which is running at about 160bhp dont know the torque fugures but. The car weighs about 810kgs gear ratio is standard toyota and the wheels are 15" with very worn out toyo T1r's . So nothing that big to break a shaft like that!

Can you post a well focused picture of the end of the shaft showing the cracks, the grain in the metal, every picture tells a story to the Non-Destructive-Test guys. A picture of the entire shaft would also help. Was the spline cut into the shaft or welded onto the shaft ?

Most stub-shafts on front wheel drive cars are about the diameter you mention for 4-6-8 cylinder engines.

i will take pics later and post them up

Do you or wnyone there drive the vechile in reverse at high speed a lot ?

I use to back up my long drive way at high speed because of the hill, to lazy to turn around or to dumb. I had a lot of CV problems from that.

No we dont reverse at high speed here we only speed farward

I've seen a lot of broken shafts to your description, normally they are fatigue failure, the source is normally the bottom of the spline cut. They are usually square cut with negligible fillet. The cracking starts at the corner as you would expect. The fix is to Either reduce the spline number, normally you want big fat splines for high power transmission rather than lots of little ones Or get the machinist to round out the spline valley so there is no sharp corner. The shaft end of the spline cut is also crucial. Ideally it should curve up gently to the top - about 5-6 spline widths in length. The turn up point is quite often where the spline fails so this should be kept as smooth as possible, no sharp lines.

Finally you want to match the female (inside) spline to the shaft and ensure they are not binding and not really loose. There is no real figure for it but 10 thou" backlash cold is about correct if a little on the wide side.

If all that fails you can try either reducing the shaft diameter prior to the splines so the splines sit proud Or use a different alloy but as you found the correct metal can be hard to come by and a bugger to machine!

Yes it looks like fatigue failure but what after a month? and the car isnt even used every day! I dont know about the splines but when he give me the new one i will post pics of it and you can see what type of splines he is doing as im not an engineer and im not that technical.

Havn't seen it mentioned yet, but why not use a new factory shaft?

-John

as i said i fitted the 4 ef-te engine from a starlet Gt 1.3ltr turbo into my aygo which was 1.0ltr 3 cyl N.A and the cv at the gbox side was a match but the outside CV on the Aygo is pretty small so i opted for a bit bigger one from the yaris which come out with a 1.3ltr na engine so the cv's are a bit stronger.But u guess i have to go to a smaller spline count.At the moment i have 24 outside 19 inside , so i might try 24 / 23 combination as the shaft would be thicker and also the splines.

Hello nexia201,

Since you are opting to have the parts machined on site rather than a custom order from facilities that do this regularly. You could instead order the splined ends having rolled splines then stubbed to the shafts. This would simplify future repairs as you experiment with your design too.

I could not find a quick reference on heat treatment of EN24 but did find its a .40 carbon steel formulated for toughness.

Using some vaguely remembered rules of thumb the improvement in strength from heat treat and proper anneal should boost the shaft strength 50 to 100% giving you some safety margin.

Shock loads are the usual cause of shaft failure. If your drive train is reving up while the wheel is air born and then the shaft has to carry the load of slowing down the engine, transmission, and engine power and the inertia of the engine guts in a very brief period while the suspension is bending that is the type of high impulse load and or strain that fractures a shaft.

Could it be the way you drive?

Hope she holds together for you this time. But, even a 100% safety factor is a little slim for typical automotive applications. No guarantee from me on this one.

Mr. Gee

hello nexia201,

During manufacture of the failed shaft were the splined ends stubbed to the shaft?

If you have invested for a spunkier mill then spring for a better than average hot rolled shaft material such as a 4140 or 4340 annealed and if the splined ends are stubbed which is fine; have the shaft end to be welded heated to cherry red 2" each side off welded area before and after the welding procedure then air cool. This preserves the grain structure of the chromoly steel and prevents the situation you currently endure.

You will find the proper fit allowance table for the spline in "The Machinery Handbook" google that or Barnes and Noble or Amazon, just the paper back handbook is all you'll require.

En24 is a good steel often used for shafts but normally used in the heat treated condition to improve the toughness, the ISO designation for En24 is 817M40. I would tend to agree with previous comments that you should be looking at the machining, in particular at the root radius of the splines in the area of the failure.

Hello nexia201

Rolled splines are far stronger than cut splines.

But I guess you are limited by the existing flanges or gears into which the axle parts fit.

A good photo of the broken axle end would be useful, as mentioned above.

Kind Regards....

817M40 is excellent material.

As mentioned pl check the Heat tretment condition. To have good fatigue properties get in HT condition (hardened and tempered)

Pl check the corner root radius near the break whether it was nicely rounded ? a smooth radius of atleast 3mm is preferred. Check the word smooth. if machining lines. roughness, chatter marks exist these are recipies for fatigue failure.

Photo please ? did it twist and break or fatigue failed

One month is good enough for a fatigue to start and finish I have seen fatigue failures at a much smaller time.

Check my own experience on shaft failure available in CR4

Thank you all for your help unfortunetely he didnt finish the shaft today so i dont have the pics yet (but should have them by tomorrow!). Some of you are speaking in too many numbers which i dont understand the splines are cut in not rolled as that is the only option there is. I had a word with the engineer that did the shaft as he had previously told me that the shafts were heat treated but when i told him now he said he never said such thing and that they dont do the treatment and I have to take them to the local university to get them hardened and tempered (i guess engineering class) So probably after getting the car back to my garage i will take out both shafts and take them to be hadened and tempered , but wouldnt tempering make them more brittle cos thats what i was told?

but wouldnt tempering make them more brittle cos thats what i was told?

No

thanks ill take your word for that ! im sure you know more than the guy tellin me

Just click the hyper link ( the highlighted "No") and go to the info site for complete satisfaction.

Yes I'm certain

Hardening will increase the strength at the expense of brittleness; tempering will then take back a small portion of the peak strength , giving in return much more toughness.

The hardness/toughness treatment is a balancing act.

Can you supply photos of the BROKEN SHAFT?

I will but later as i dont have the shaft at the moment.

well here are the pics as promised I had to take them with the mobile so dont expect 12mp resolution

This is the broken shaft wher the splines were twisted

and this is the cross section of the shaft where the cracks are present

and this is the new shaft

Thanks for the view

I thought to ask previously but now seeing the result, what type transmission are you employing? If a manual type it's possible the type clutch disc may not be suited for the increase of power on the surface you are using it. For example a clutch impregnated with ceramic or metallic will engage too abruptly on pavement though it is fine on loose surfaces such as gravel, sand or dirt.

Of course improper material spec's will produce similar results or correct material with incorrect material conditioning.

Good Steer!

A comment was made about larger splines being stronger but that just doesn't make sense because this requires a lower number of spline meaning the minor diameter is less than use of the smaller spline with a larger minor diameter.

Am I making sense?

It makes sense to me.

The total cross sectional area is what is going to give the total shear stress that the shaft can cope with.

Improving, (enlarging) the radii will help to reduce the stress raiser effect at the root of the splines if they have to be machined. Ideally they would be forged as this gives a better grain flow and then finished machined.

Wouldn't it be nice to live in a perfect world. Would they need engineers in a perfect world? Maybe not! scratch that last thought.

BAB

he was telling me trhat as there is one side od the shaft that is thicker than the other the thin ner end is prone to break so he took off some more from the middle of the shaft (made it thinner) as he said this would make the shaft flex better.

You need a better consultant.

Again what type transmission you're using? If automatic the difference in power will make a substantial condition when you consider the torque converter stall speed effect.

Though from the picture I would suspect inferior material condition.

well thats what i was thinking and i told him is this en24 ? he told me yes it is and the new shaft is ot thesame material so if the other one failed this one is going to fail too as it is thesame! Im not sure there is anywhere to check the material hardness and thats why they could be off with it sometimes when you pay for material and they give yon another one.

Do you guys thing that if i get my shafts machined out from half shafts it would be stronger?

The halfshafts i've modified seem to be through hardened, & are a *^!!@#!

The shop will probably invent a few words to describe the process.

Can you find something from another model?

Ha!

You're description of the right stuff is accurate but I learned how to turn Stellite and cut it with Carbide or to turn Carbide and cut it with Stellite too.

45 Rockwell is a peach but 52 is harder than woodpecker lips for sure

Try the old 'file test'; at the proper hardness a hand file will take serious work to cut.

Not much different from a GOOD 'allen' (in-hex) wrench

i read ur problem .this is not a big one.The material dat u r using is ok but u have to go for heat treatment b cause ur shaft is twisting it means it is weak for ur required load.Just gor for heat treatment ,take its hardness to 40 then it will b ok

Thank you all with your help it is very usefull (especially BWIRE) i am trying to find somewhere close to Malta to send the shafts to be tempered.

Have same problem , had the shafts hardened tuesday and fitted them yesterday and today took out car for a spin and one of the shafts snapped its like they are made of candy!

now i went to another engineer and he said he will make the shaft out of EN 24T. Is this material b etter than the EN 24 ? as he said the EN 24T has properties like the 4340 chromolly , is this true?

looks like these were a bit too hard.

also looks like fracture originated at the root of one of the grooves, at the runout.

necking the shaft slightly smaller than the root diameter, & blending into spline major diam. with a smooth fillet works wonders to reduce peak stresses in this area.

Hi nexia201,

I agree appears too hard or lack of temper.

I do not have specifics of EN 24 or EN 24T other than description of structural materials.

EN 24T is comparable to 4130 a structural grade perhaps though I'd opt for 4140

well i think he said he is going to use the en 24t and leave it as it is (no temper no hardening as he said its stronger) i will ask him about the 4140 if it is available here (would it be good to leave raw or it will need tempering still?)

Forgot to say thatshaft that broke had the splines already twisted from where you said it broke (had turned about 3mm) and i told him before tempering if i should change it but he said take a chance with it !

Re: alloy selection= here in the States i'd use 4340 in preference to 4140, that bit of Ni seems to help on machined parts. EN24 (based on above analysis) should be every bit as good, provided it has the proper %age of carbon.We're looking for ~ 0.4% C less is lacking in strength, more start to get brittle. i probably would not go 'harder' than Rc 40-42 (we're looking for properties other than true hardness, but within an alloy these correlate w/ hardness, & that is relatively easy to check)

The extra strength of medium alloy steels really stands out only after heat-treatment.

Shafting made from pre H+T materials are usually straighter; also have a uniform strength gradient.

It should go without saying that, once installed, an axle should always in the same primary rotation; the reversed twisting will be a very short thing.

Hardening followed by tempering are steps within the process with enables classification of a material for service. Which weak link do want to create?

En24T is the same material as En24 but it is supplied in the hardened & tempered condition, it should be in the range of 850-1000N/mm². It is equivalent to a Chromalloy steel 34CrNiMo6.

so im back with a question , i was told that all the en24 we get here is en24T now im getting another shaft done from en24T and i am changing the cv joint as i found one with 22 splines instead of 19 splines and the shaft measure at the splines would be 23mm anstead of 21mm my question is would that 2mm in the diameter do the difference as the shaft is going to be 2mm thicker?

Hello nexia201,

Every little bit helps