Shaft steels

[agpowder]

Hi, I hope someone can help me.
I have designed some specific application machinery and one of the processes is a twin screw extruder for making powder coatings, I designed this machine two years ago, made it and have been using the machine now for over a year. originally it was only meant to be a prototype just to see if it worked, ie did it make powder coatings, anyway it did and I have developed a business around this extruder.
The problem is now the gearbox, the shafts last about 2 months and then fail. I recently redesigned the mechanics basically beefing up the shafts diameters etc. My question is: Previously (whilst living in England) I always used En24T, turned the shaft, installed and ran and I always had good luck with it. Here in the States I have had trouble with steels, Does anyone know a decent equivelent to En24T (is it 4340 - chemically similar but is it equiv in physical properties?) I also used 1144 on advise from our steel stockist, but I think he had another motive. OR - does anyone know where I can get En24T from in USA?

 

[TVP]

agpowder,

Several members like CoryPad have link to FAQ's in the signature, so it appears in all of their responses, and is not directed at any one individual.

 

[agpowder]

OK, I have found a supplier of 4340 with heat treated yield strength of 115K PSI and 112K PSI for 1.75" diameter and 1.25" diameter respectively. The metologist (sorry if spelt wrong) is on vacation until Tuesday.
I wanted to check two things here:
1. Is the steel still machinable?
2. Can the shaft after machining be case hardened and ground at the bearing and gear seats?
I have spoken with the CNC shop next door (they will do a better job than me) and they are happy to machine up to 44 Rockwell but the steel stockist salesman doesn't have this info, (he refers to the metelogist)

 

[unclesyd]

Checkout Castle Metals for preheat treated 4340. Normally on preheat treated bar stock the hardness is in the range of Rc 28-32 and very readily mahined.

http://www.amcastle.com/productFormsBar.aspx

It doesn't matter who you get it from you need the E4340 or "Aircraft Quality" material.

Here is another material that we used extensively for parts such as yours. It is a little pricey but we have had excellant results and the Associated service is/was next to none. We used this material as universal repalcement for all low alloy steels.

http://www.associatedsteel.com/pdf/09-10.pdf

 

[EngineerTex]

1) You can machine anything. It's just a matter of how long it takes and how much effort you have to go through.

2) For 4340, a hardness of 44 Rockwell (c-scale, I assume) is somewhere around 180,000 psi - 185,000 psi yield. This would probably be the upper limit of any of your candidate materials.

3) The hardness of 28-32 Rc is around 115,000 psi - 130,000 psi. The range of 115,000 psi and above will definitely require a heat treat, which is no problem -- you just purchase it in that condition. As a side note, it's the temperature of the temper that determines the final hardness of the heat treating of 4340.

4) I'm not so sure that you would need to case harden the bearing and gear seats. Did your machine shop tell you that you would need to? How are they planning on hardening the part?

-T

Engineering is not the science behind building things. It is the science behind not building things.

 

[agpowder]

Thanks Engineertex.
Yes the shop next door said they could send away to case harden the shaft and then they could grind to size. I think he just said he could do it if I wanted it rather than really suggesting that I needed to do it. I get the impression that you lot know more abot the metal properties than he does. He did also say that case hardening would improve the seats life or durability without taking away from the shafts application requirements. What do you think? case harden or not?

 

[EngineerTex]

He is probably correct that it will improve the seat life and durability without taking away from the application requirements. My view is that it's debatable and I can't definitively say to do it or not.

The best thing to do is go to the SKF, INA or Timken website, check out the engineering section and see what hardness the bearing manufacturer says you need.

The experimental side of me says not to do it at all and see what happens. If your shaft breaks again, then the surface hardening was either wasted on something that failed elsewhere, or the hardening was the cause of the failure. So my experimental philosophy would say not to change too many things at once. I could be wrong about that and I'm sure that someone here probably has a good reason why I am.

Is the hardening to be done by flame hardening? Carburizing? Shot-peening?

-T

Engineering is not the science behind building things. It is the science behind not building things.

 

[unclesyd]

I would not carburize the part in question. The shaft material, if 4340, is capable of supporting a bearing with no problems. There is a caveat in that the bearing has to be operating within it's design parameters.

 

[agpowder]

Ah. now that we are on load of bearings. The original small dia shaft gearbox bearings never failed, however I always changed every bearing everytime I changed a shaft ~ every two months ish. Now with the new big shaft gearbox I have had to change to a smaller OD/ID difference dimension, I mean the OD stayed approx the same but the ID got larger. The dynamic load ratings for the two deep groove ball bearings are 6890N and 8840N for the small shaft g/box but the Dynamic load ratings for the ball bearings on the big shaft g/box are 6370N and 4360N. This could be an issue. BUT I have been more worried concerning the bearings about the width of the bearings, the new bearings are 7 and 9mm and the old bearings are 11 and 10mm.
Now this is an uneducated guess but if the under bearing race bruising was from nocking or a whiplash load then the shaft damage would presumably be at a single point on the periferal of the shaft and now that I have reduced the width of the bearing does this mean faster damage?
However I think with the new shafts properly tolerenced (slight interference fit) and better machining quality and better shaft material and bigger dia then hopefully there wont be a problem?
I am in the process of ordering everything to make 4 additional g/boxes, two for new machines and two as spares on the shelf. I really would like to get them as right as right can be.
The thrust bearings load rating has pretty much doubled from the increase in shaft dia and so can probably be not regarded for this part of the discussion, or maybe I should worry about "not enough" load.
I forgot all about castle metals, I used to play cricket on the south side of Chicago with one of the supervisers there, small world isn't it.
Oh by the way I machined 0.4mm off of the screw and kneader diameters after you commented about the loss of tolerence on your WP and this really has reduced the load and made the machine sound smoother without any noticable dispersion affect - thanks.

 

[agpowder]

Here is the photo of the damaged shafts.

 

[agpowder]

Here is a plan view of the gearbox with the top taken off.
For scale the diameter of the input shaft is 40mm where the pulley or sheave sits.
The gearbox output is 5/8" Hex.
This is the "new" gearbox after 6 hours operation.

 

[agpowder]

Here is a picture of the intermeshing screws of the extruder. These are an old set but very similar to what is running now. The powder coat premix is intraduced into the auger flighting at the far right hand side and then conveyed towards the kneader blocks on the left side. Only the left side of the extruder is heated really just to start the melt. Once conveyed the premix is forced through the kneaders and is worked on to disperse all the raw materials. Once running the work the kneaders impart on the product self generates sufficient heat that we actually have to chill the extruder barrels to cool them down.

Although the individual components are tightly clamped I did notice when I turned the overall diameter down (between centers on lathe) the screws where definetely whipping, I think it would be a reasonable assumption that they try to do this in the machine also.
The only support that these screws have is the gearbox at the right hand end and the floating natue of the screws in the molten plastic at the left hand end. This is why they chatter and bang around when they are new or have just been cleaned. It takes 20 seconds or so for the epoxy or polyester resin to fully support the kneaders.

 

[unclesyd]

You definitely have fatigue involved though it might not be the whole picture. The number 3 and 4 (left to right) shafts have definitely failed from fatigue, evidenced by beach marks.
The large shaft 3 is interesting as a failure in the keyway area usually indicates a torsional component in the equation.
The number 1 and 2 shafts also look to have fatigue as a major component of the failure.

Considering the age of the gearbox it appears that there is a severe wear problem with the gears. As a first pass it looks like they are way under designed.
There also appear to be heat tint everywhere. You will need to change your lubrication to an oil with a higher temperature limit and very high EP load.

At present you have what I call a basket case. you have to resolve the problem piecemeal. Like fix the lubrication, get a harder/tougher gear and work on getting a better material for screws.

Do you remove any components prior to failure?

Have you had any problems with the couplings?

Do you have any connections to a Metallurgical Lab?

 

[desertfox]

Hi agpowder
Thanks for the photographs.
Well the shaft on the far right with the keyway looks like it failed from fatigue and the fact that the failed face isn't flat suggested that it failed under combined torsional and tensile loading. The other 3 shafts look to me anyway to have failed in torsion quite rapidly looking at the surfaces, on the those 3 shafts there seem to be a machined shoulder just at the point of failure.
Now interesting if I am right and the 3 shafts have failed in torsion ie by shear stress then according to my book "Machine Design" by Paul H Black and O.E.Adams Jnr Fatigue cracks do not grow under shear or compressive stresses but I am happy to be corrected if my failure mode is incorrect.
The 3 shafts that I felt have failed in torsion where abouts is that failure when looking at the gearbox picture?I agree with unclesyd its looks like the design needs beefing up and this is where the number crunching comes in, it may pay you to get someone to analyse your design before making anymore parts. Also as suggested get those failed shafts to a laboratory and they should give you a better idea of what the problems are.

Regards

desertfox

 

[agpowder]

Thanks Guys.
I was concerned about the gears as well and I stripped the box down again with intention of changing them out.
I agree and am going to impliment everything that you guys have advised.
For the short term I have to get the machine running for production. Hopefully the NEW (today) gears will last a week for me to get all the stuff prepared. I am going to get the 125K 4340 and machine a set of shafts really carefully paying attention to the fits and finish. I am going to relook at the design and see if I can increase the width of the gears and get them specially made by a specialist gear maker and take their advise on the steel etc.
After getting the machine rigged up to last a while I do need to find a company that can develop the gearbox.
Do any of you know how I could possibly find an engineering company that may want to make and market the machines? They really do work good at making the product.

 

[agpowder]

Sorry desertfox I didn't answer your question.
The three shafts in question did fail at a shoulder and the two bigger ones (the input drive shaft) failed at the other end to where the pulley (sheave) is.
The smaller shaft failed at the right hand side of the gear - the other side to where the hex is.
These shafts are from the old smaller gearbox.
These shafts represent the three types of failure.
The bearing seat diameter reduction damage is the least common. the shoulder fail and the keyway fail are about equal in occurance.
I think the basket case description is correct, I think I need to make and assemble this whole box perfectly with regard to fits and finish and see what happens.

Excuse my ignorance again, when you say a very high EP Load are you refering to the AGMA Grade. I am using an ISO grade of 46 which I think is about 1EP are you saying I should use say 7EP which has an ISO grade of 460?

 

[Artisi]

The actual breakages are now well covered, so my comment is regarding the machined finished on the shafts especially in the region of the failures- the surface finish is, if you will excuse an Australian colloquialism,"bloody awful" - from what can be seen on the photo's a lot more attention is required to surface finish - I would not be surprised if one aspect of the failure is "stress raisers" which have been machined into the shafts.

 

[unclesyd]

Yes I would go with a 7EP until you can get things under control. Here is an oil that I've used for many years both as crutch and a standard lubricant. One thing it will do is track the gears and is very slow to drain off on shut down. You never startup metal to metal.
There are other products that have similar characteristics.

http://www.le-inc.com/products/documents/0604-0609_tdb.pdf

agpowder can you give us some information on the input to the gearbox and maybe a little information on the gears, motor hp and rpm and sheave sizes? That is one helluva an input sheave.

desertfox,
I was thinking along the same lines as you but when I went to a better monitor I believe I see beach marks on #3. The edges of the fractures appear to have the look of multiple origin fatigue. The final fracture would be shear in either case. It gets tricky since the shafts are so overload. I would like to see the fracture face with oblique lighting.

 

[desertfox]

Hi All

Artisi makes a very good point regarding surface finish.

Unclesyd
You may well be right I tried zooming in on the pictures and
the shaft next to the one with the keyway and it may have some slight beachmarks but I couldm't be sure, I could certainly see some river lines pointing from bottom right toward top left indicating the source of a crack.
All the failures look quite brittle and it would be interesting to know how long this items lasted in service and are they all made from the same material.

desertfox

 

[unclesyd]

desertfox,
As you posted I was sitting here thinking about my old metallurgical lab with 2 Axiomat Metallurgical Microscopes, 2 high power stereo microscopes, and all the accessories. Just around the corner from the lab there are 3 SEM's. All this equipment is going unused as they have no one that is interested. I surely would like to have the fractures and about one hour.

agpowder I will check tomorrow if I can find the name of the company that built some multi output gearboxes for us.

 

[carnage1]

Better surface finish and a stronger steel will help you I am sure.
One specific design change you should do is to change the shape of the keyway. your current keyway design is a constant depth and ends with a radius equal to its width. that is not a really bad keyway but you can do better.
make your keyway the width and depth you have it, but at the end it should continue straight as the depth of the keyway shallows up along the radius of the cutting tool you use. this will greatly increase the fatigue strength of that portion of your part.
The machine shop will have no problem doing this or if you choose to do it you can use a woodruff cutter cutting in from the side (assuming a vertical mill)

http://www.abmtools.com/Cutting_Tools/Woodruff_Cutter_Key_Way.htm

the picture on the left in this is what I meant by the shape of the keyway. (took 15 pages of Google images to find a good drawing)

http://www.me.metu.edu.tr/me307/Standards/keys/TS 147-ISO 2491.pdf