View Full Version : Corse spline cutting

11-12-2007, 08:44 PM
My buddy showed up at my shop today with a 60 year old transmission input shaft for an old Massey Harris tractor. I found this thread and decided to try to duplicate the idea in the second post: clicky (http://bbs.homeshopmachinist.net/showthread.php?t=22842&highlight=splines)

Here is a quote from the thread, hopefully including the pictures I copied.

I used a hand ground bit in a 1 in bar (kind of fly cutter style)held in collet in my Cincinnati 2MI horizontal mill ( would work just as well with a vertical). used dividing head with tailstock, ran fairly slow feed and speed to keep vibration down. Came out like a factory job, fit clutch splines exactly like original. Spend the time grinding the single point cutter to exactly match existing good part of spline, be sure to have tip relief and side relief and make chips. I made about 4 passes to full depth of spline and did not have to touch up bit once...

Here are some pictures of the tool holder and bit that I made.



My buddy ground the HSS lathe bit to match the existing splines. He put more relief on it that I would have thought was needed, but neither of us have ever ground our own cutter, so we were going off descriptions, and playing it by ear. I don't know if it makes any difference, but I was using this in my vertical mill, I don't own a horizontal mill.

Anyhow, the problem was this, the cutter seemed to be knocking, no serious vibration, but it wasn't making chips. For whatever reason, it was just pounding the work until the tip of the cutter wore/broke off. We built the damaged section of the shaft up with weld and then turned it down on the lathe with carbide inserts. I don't know if the problem was due to the weld being too hard, or if it is cutter geometry. We are going to have another go at re-grinding the cutter tomorrow. Any advice?


11-12-2007, 08:54 PM
I think you know the answer . My guess is the weld to hard. But befour you start again Rough out the slot with a small end mill the mount the fly cutter tool and rotate it by hand and check for clearance after regrinding your tool bit.you may slow the RPM down if the shaft is tough.I think you are on the right track just take it easy to start you can always speed up after you see what is happening.

Michael Moore
11-12-2007, 09:14 PM
A single-point tool like that isn't going to be very smooth cutting. It is just like doing an interrupted cut on the lathe where the tool slams in, springs back when it cuts free and then repeats the process. You'll want to take light cuts.

You may also want to try taking a file to the weld to see how hard it is. You may need to try a carbide tool (which will probably be more prone to chipping) or anneal the welded area to soften it. As lane says, anything you can do to remove excess metal with an EM (and you may want to use a carbide EM if the shaft is hard) first will be a big help.


11-12-2007, 09:24 PM
It was welded with ER70-S6, which should not be much harder than mild steel, but who knows what the 60 year old base metal was and whether it would have effected the weld area hardness much or at all.

I just had another thought though, is it possible that the point of the cutter is to far from the center of the holder, causing it to enter the work at to shallow of an angle? It seems that maybe closer to the axis of the holder may make it slap less. I have done interrupted cuts on the lathe before, this thing literally sounded like a wood pecker...

My original plan was to make a holder that took an insert to do this, but I didn't have any inserts close enough to the right shape/angle and nothing to try to grind them with so I decided to try the HSS.

Thanks for the advice so far.

Michael Moore
11-12-2007, 09:53 PM
Jason, the farther the tip is from the center the more flex you've got in the assembly. I think that making it all as short as you possibly can would be a good plan.

Did you make sure that you didn't get the cutter tip positioned so that it is entering the cut too high or low (like having a lathe tool too far above or below center)?


11-12-2007, 09:54 PM
It was welded with ER70-S6, which should not be much harder than mild steel,

Agreed on the filler: ER70S6 is just mild steel wire with extra silicon as flux for dirty steel -- it shouldn't be able to get very hard. Did you weld the entire spline in one shot, or did you do it in sections, and let it cool in between?

I can't tell from your picture, but did you grind a top rake angle into the form-cutter, so the bit isn't hitting the work flat?

11-12-2007, 10:04 PM
I've never used that wire myself, but it's a high carbon (0.07) and manganese and silicon alloy. The carbon content alone suggests that it would harden if cooled quickly.

Maybe John S. could suggest a wire that is easily machined, since he seems to often use MIG to build up damaged shafts?

I've learned the "hard" way that welds almost always have to be annealed before machining, when using 7018 or even the humble 6011.

Typical Wire Chemistry
C 0.07
Si 0.80
Mn 1.45

Typical Mechanical Properties (As Welded)
Yield Strength, psi 73,000
Tensile Strength, psi 90,000

By comparison, mild steel is 36,000 - 40,000 psi.

11-12-2007, 10:13 PM
ER70 is .07% carbon -- that's considered mild steel (0.05–0.29% carbon content).

Medium carbon is 0.30–0.59%, and high carbon is 0.6–0.99%.

Yield Strength, psi 73,000

That's the "70" in ER70, or 7018 stick electrodes: the yield strength of the weld. ER60, or 6061 stick electrodes, have 60 K Psi yield strength.

11-12-2007, 10:21 PM
Agreed on the filler: ER70S6 is just mild steel wire with extra silicon as flux for dirty steel -- it shouldn't be able to get very hard. Did you weld the entire spline in one shot, or did you do it in sections, and let it cool in between?

I can't tell from your picture, but did you grind a top rake angle into the form-cutter, so the bit isn't hitting the work flat?

My buddy did the welding while I was turning the fly/spline cutter, so I am not sure how much he welded at one time. There was top rake, however, one thing I did not account for was where the tool would be in relation to center. The hole for the bit is drilled on center radially through the holder. So, this does put the cutting edge ahead of the center line of the tool holder. There was still enough relief ground that there would be no rubbing of the heel, but perhaps we need to regrind with the cutting edge on the center line, or as close as possible. It did cut a very light cut for about an inch before it just started making noise. I suspect part of the problem is the massive amount of relief he put on the heel, it probably weakened the cutting edge a lot, but I really think that there is a geometry problem here as well. I just don't have the experience to know what it is. If anyone could give better ideas with more pictures/angles of the tool, I can shoot some more.

As for the welding filler, if I am not mistaken, the tensile strength of the filler or steel does not have a whole lot to do with it's hardness. It was not cooled quickly, I know that for sure, we didn't want to take a chance warping the shaft. It could need annealed, and if that is the case we will try it tomorrow.

Thanks again for all of the help so far.

11-12-2007, 10:24 PM
Maybe I will try the cutter on a chunk of CRS just to give a test of if it will cut something that hasn't gotten hot... I will hit the shaft with a file next time I take the dog out and let you all know the outcome. I have filed my welds numerous times though, so I kind of think this is an unlikely culprit.

11-12-2007, 10:31 PM
I've done some shaft repair in the past on projects like yours, I would be willing to bet that your shaft is made of something between 1045 and 1080. The rod mixed with the parent metal can be quite hard because the shaft acts like a heat sink and rapidly pulls the heat from the just welded area, just like a quench. Try a file and see how hard it is. If it is hard you will have to anneal it before machining. If it isn't hard then you will have to check the geometry of your cutter. Make sure that the top of your cutter is on the centerline of the fly cutter or just slightly below center. If it is to far above center then it takes a lot of clearance to make it cut thus making the edge weak, and then it won't cut well.


11-13-2007, 04:37 AM
Having done the original pictures, I will go a little further in :)

Use only enough back rake as needed, too much will really weaken the poor cutter, which is getting the $$$$ pounded out of it ...

My job was into the original non welded Model A Ford Input shaft ,which was pretty hard, I would say around 40 Rockwell C or higher.

You could make a larger diameter tool holder with hole offset and less tool stickout. Flex is your enemy.

Tighten unused table locks..

Try to start at about 1/4 of total depth, Some steels will work harden if not aggressive enough on feed (tough with single tooth cutter tho)

The file test listed above should be done (If you can't cut weld with a file, HSS won't either...

Annealing and a reheat treat may be required (or even that might not be possible, due to whatever happened during welding).

Another messy option, is to dress a thin grinding wheel to proper form and grind the bugger...

I also used a 5500 lb beast with 50 taper tooling :) still sounded like hell.. I was amazed edge held up for whole job...

11-13-2007, 05:59 PM
Well, I worked on this again today, all I can say is I have got to be doing something seriously wrong with the fly cutter. I reground the cutter, moved it way in, put the cutting edge on center, and it cut better, but once I got fairly deep in the spline, it started to rub, and with the cutting edge ground so far down into the bit I didn't have a lot of room to grind more relief. If I ground more, then I snapped the tip off. I was also never able to take much of a pass with it, it is just plain ground wrong. I will educate myself further in my spare time though.

Here are pics of the cutter when I gave up on it:



I ended up making this out of a HSS end mill. This was my original idea, until I say Bguns post. I tapered the EM to fit the spline and then I cut tapered flats 180 opposite each other. Put a bit of radius on the tip and shallow relief in the center of the "spade" on either flat. I was kind of surprised that it cut at all, I am sure it is no where near correct geometry, but I was out of ideas. It actually cut quicker than the fly cutter. Made chips and a passable spline. I had to re-sharpen it several times, and was about out of EM by the time I got the 10th spline done, but it did a satisfactory job, and not a real terrible finish either.


Anyhow, here's the results. I have to clean up the OD of the splines a bit. I let my buddy try his hand at turning the weld down, and he got it a bit cocked up, all in all, I am anxious to see if the clutch disc fits after all of this work.



11-13-2007, 07:33 PM
I do the weld repair of shafts ALOT! 1 st rule is turn the shaft down before welding so you later machine JUST THE WELD! The weld bond to parent metal section is what gets VERY hard! Even in mild steel or scrap iron.


11-13-2007, 07:40 PM
Thanks Mark -- that's a good tip. Sir John mentioned that he turns the shaft down by 1/16" before he MIG welds it, but I think that was to keep the weld from delaminating from the parent material (i.e., so you're not milling past the weld line).

By the way Jason, it's too late now, but it occurred to me that you might want to try the form cutter on a piece of mild steel to make sure it's cutting well before you go for broke with the weldement.

In any event, nice job on the repair!



11-13-2007, 09:13 PM
Thanks Mark -- that's a good tip. Sir John mentioned that he turns the shaft down by 1/16" before he MIG welds it, but I think that was to keep the weld from delaminating from the parent material (i.e., so you're not milling past the weld line).

By the way Jason, it's too late now, but it occurred to me that you might want to try the form cutter on a piece of mild steel to make sure it's cutting well before you go for broke with the weldement.

In any event, nice job on the repair!



I don't know a lot about machining, but I do know a fair amount about welding, and if your weld ever "delaminates" from the parent metal, you did not weld it correctly. The filler and the parent metal should totally fuse and become one metal. You should never see a joint or a differentiation between parent metal and filler. If you do, you have a cold weld and it is not fused with the base or parent metal.

I could not turn down and remove the existing splines and start fresh. I set up off the existing splines for every cut. I do not have a dividing head or any other way to index, and so I had to do it with V blocks and clamps. It wasn't incredibly easy to set up, and I wouldn't try it on something that wasn't destined for scrap if this attempt failed.

As for trying the form/fly/spline (what are we going to call this thing?) cutter on mild steel, that had crossed my mind, but, the weld was easily cut by file, so it would have had to been a real fluke to have it be too hard to cut with HSS.

I know for a fact that something was wrong with my grind on the original cutter. You can see in the above pictures where it got nice and shiny from rubbing. Originally I reground it to fit the splines while exactly perpendicular to the shaft. I think that for one I need to make a new holder with the whole off center so that I don't have to grind so much bit away to get the cutting edge on center. Then I think it really needs a lot more heel and side relief than I thought. My buddy probably had it ground closer to what it needed to be when we had the edge way above center. Once I got to 2/3s the depth of the splines it quit cutting all together and just started rubbing and galling the metal. I tried a couple more re-grinds to put more relief in it, but in the end I decided to figure that stuff out later and go with the end mill approach. Kind of makes me wonder if there are tapered EMs in production that would match the spline geometry. Would make short work of something like this.

It ended up taking me about 4 hours of total time in front of the mill to cut 10 splines. That is to long IMO, but it was for my life long best friend, and he has put up with my projects around his shop for much, much longer than that.


11-14-2007, 04:51 AM
Your tool is ground all wrong....

That top rake is reversed and you are running it as a Negative rake tool which works on a stout lathe with Carbide and non interuppted cut..

The top (As you see it if it was in a lathe) needs to tilt below tip to rear of cutter, yours tilts toward tip from rear of cutter.

Looks like you need more side clearance also ... Hard to describe
angle is exaggerated but once you think about it it will make sense. You are pushing the knife blade sideways at wrong angle into cheese for example...

tilt blade and angle will be right to cut instead of rub...

Here is a super quick sketch with a dull Sharpie....

Of course some one will post a pretty CAD 3D Color sketch.... :)

11-14-2007, 05:38 AM
It did have more of a "hook" shape to it before I managed to round off the cutting edge. I did read alot more about grinding your own cutters tonight, and I am going to give it a shot re-ground once again like you say in just a chunk of mild steel for grins.


11-18-2007, 03:31 PM
Well, I found a 10 spline clutch disc in my parts pile, and tried to fit it on the shaft, and it was a no go. Looked like the EM cut splines were just a hair off on angle. I decided to give the form cutter another go. Third try was the charm on that devil. I read a lot about grinding your own bits the other night. One suggestion was to set your grinder rest at 7 and use it to cut all of the relieved sides. I cut the shape with 7 of relief, and then I used the rest set at 0 and cut the top edge of the bit down bellow the center of the arbor. I read that form cutters do not need any top rake, but I ground a bit just to be safe. I really wish that I had figured it out in the first place, it definitely went faster than cutting with my EM. Anyhow, here are the pictures.




And a picture of the touched up splines. Are the lite chatter marks in the depth of the spline normal with this type of cutter, or does it still show I did something wrong with the cutter geometry?


Cont. next post for picture limit...

11-18-2007, 03:33 PM
I don't think the chatter marks will effect performance, the clutch disc fits and is good and tight. No play in the splines, and I cannot feel a difference between the existing splines and the new cut ones.


Anyhow, thank you all again for your help. I appreciate it.


11-18-2007, 03:47 PM
I expect that the added side relief really helped you out here. It was the flanks that were rubbing badly in the bottom of the cut previously so you were rubbing.

11-18-2007, 05:02 PM
Nice to see you got er done.... :)

The main problem I see is, you have center of hole hole for the cutter down the center line.. I would drill toolbit mounting hole in cutter holder off center.

Off center holder would allow more metal to remain on tool bit which would make it stronger.

The holder could have a simple end milled slot for a square tool and be open on bottom....

I cheated, and used round HSS tool bit stock, so a simple round hole is all that was required....

On my Horizontal Mill I placed a small Jack under middle of shaft to help reduce spring of work, tougher to do on a vertical but you should have had something lightly pushing on shaft on non work side of shaft to reduce spring/vibration. This would have helped with surface finish. :)