Meethod of cutting herringbone gear

crossthread · Jan 21, 2021

Method of cutting herringbone gear

I need to cut a herringbone gear for a repair job. Not that it really matters but the gear is 8" in diameter and over 60 years old. I can't figure out exactly how this was done. I have cut plenty of gears but never one like this. My question is this. Would it be possible or appropriate to cut two gears with opposing angles and then attach them together? I don't see how you can cut the gears without destroying the opposing teeth in the process. I assume they were made with a form tool in a mill but not sure. Thanks.

Milland · Jan 21, 2021

Here's one method, about 2:00 in:

Herringbone gear / Cavuş Dişli - YouTube

If you have infinite time you can use endmills and a fourth axis.

implmex · Jan 21, 2021

Hi crossthread:
Here you go:
Farrel 3-C Sykes Gear Generator - YouTube

Here is what Milland is talking about:
Exceed Gear Manufacturing & Engineering - Herring Bone Gear - CNC Machining - YouTube

And to answer your question about bolting two helicals together...yes you can but you need to be sure the bolts will be strong enough for the load.
Herringbones are used in applications where high loads are encountered and you need to limit the axial thrust of a helical pair.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

Scottl · Jan 21, 2021

implmex said:
Hi crossthread:
Here you go:
Farrel 3-C Sykes Gear Generator - YouTube

Here is what Milland is talking about:
Exceed Gear Manufacturing & Engineering - Herring Bone Gear - CNC Machining - YouTube

And to answer your question about bolting two helicals together...yes you can but you need to be sure the bolts will be strong enough for the load.
Herringbones are used in applications where high loads are encountered and you need to limit the axial thrust of a helical pair.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

I would think dowel pins plus a strong adhesive might be better at permanently "marrying" the two gears together. I'm thinking something like Loctite Black Max but there might be better choices.

Just a thought.

pat pounden · Jan 21, 2021

replying 'in general' to this-are there pics of exactly how sidney made theirs??-thanks,pat

Oldwrench · Jan 21, 2021

Depending on the direction of rotation, the helices might try to push the halves apart—so I'd definitely include bolts along with the dowel pins.
There's nothing inherently wrong with making gears from assemblies or segments; that's how extremely large gears have to be made. If the pitch and pressure angle is a standard, then possibly you can obtain stock steel helical gears and make it from a pair of opposite hand. If you have to make it from scratch, talk to a gear shop.

Dan from Oakland · Jan 21, 2021

You don't indicate if there is a gap between the left hand and right hand gears. If there is no gap, the gear was cut on either a Sunderland or Sykes shaper. If there is a gap, it is possible to hob the left and right hand gears on a single blank. Making the replacement in two pieces would be acceptable. Normally, two piece designs are such that the gear halfs are forced together and not appart by the axial load from the gear mesh.
Lacking a Sykes generator, make it in two pieces and bolt/pin them together.
Hope this helps.

Scottl · Jan 21, 2021

Oldwrench said:
Depending on the direction of rotation, the helices might try to push the halves apart—so I'd definitely include bolts along with the dowel pins.
There's nothing inherently wrong with making gears from assemblies or segments; that's how extremely large gears have to be made. If the pitch and pressure angle is a standard, then possibly you can obtain stock steel helical gears and make it from a pair of opposite hand. If you have to make it from scratch, talk to a gear shop.

Good point re the forces trying to separate the halves.

crossthread · Jan 21, 2021

Dan...there is no gap. That is what is confusing me as to how these are made much less how am I going to duplicate it. I think the only way to do it is in two halves and a ring of bolts to hold it all together. Thanks for the help guys.

implmex · Jan 21, 2021

Hi crossthread:
Did you look at the first video I linked to?
The Sykes shaper makes a helical cut on one half and then immediately makes the corresponding cut on the other half.
So the first cutter has to push the chip right to the centerline but the second chip can fall free into the developing tooth gullet.
That way each shaper cutter can go right to the centerline and only the first cutter has to work really hard at the end of its stroke.
It only gets exposed to the pressure of one depth of cut because the end wall is continually being nibbled away by the action of the second cutter.

Without a Sykes shaper you cannot duplicate this action.
So if you want to make these yourself, you have no choice but to do it in two halves.

Cheers

Marcus
Implant Mechanix • Design & Innovation > HOME
Vancouver Wire EDM -- Wire EDM Machining

crossthread · Jan 21, 2021

That's what I was thinking Marcus. I will make them in two halves and bolt then pin them together.

Rickyb · Jan 21, 2021

Scottl said:
I would think dowel pins plus a strong adhesive might be better at permanently "marrying" the two gears together. I'm thinking something like Loctite Black Max but there might be better choices.

Just a thought.

We don’t know right now but if the angle is high to the axis the separation forces will be also. Forces will be localized at 1 or 2 dowels near the mesh. Any irregularity in the tooth profile or alignment will introduce shear into the adhesive. At a 90 degree herring bone the sep forces will be 70% of the tooth load.

EPAIII · Jan 21, 2021

Your statement, "Normally, two piece designs are such that the gear halfs are forced together and not appart by the axial load from the gear mesh." can only apply to one of the two meshing gears. For any pair of herringbone gears in mesh and transmitting power, one will be experiencing a pair of forces toward the center line while THE OTHER WILL NECESSARILY BE EXPERIENCING A FORCE TOWARD THE OUTER EDGES. And if you reverse the direction of rotation while keeping the same driving and driven gears, then the directions of these forces will reverse on each of the herringbone gears.

So, no matter how they are employed, only one of the gears can have the forces facing toward the center line. And the other must be experiencing forces in the opposite direction.

Dan from Oakland said:
You don't indicate if there is a gap between the left hand and right hand gears. If there is no gap, the gear was cut on either a Sunderland or Sykes shaper. If there is a gap, it is possible to hob the left and right hand gears on a single blank. Making the replacement in two pieces would be acceptable. Normally, two piece designs are such that the gear halfs are forced together and not appart by the axial load from the gear mesh.
Lacking a Sykes generator, make it in two pieces and bolt/pin them together.
Hope this helps.

EPAIII · Jan 21, 2021

In your case the original set of herringbone gears may have been made with no gap between the two sides by using a special machine (Sykes?) there is no reason why a mating or replacement gear has to have that "no gap" feature. I dare say that in 99.99% of the cases, any slight decrease in power handling ability that adding a gap between the sides would cause would be completely negligible and of NO CONSEQUENCE. Any difference in the life of the gears could be made up with better alloys and/or lubrication.

I would consult with the customer first, but I would make the new herringbone gears with a gap to allow for cutter clearance and chip discharge.

The dedicated gear cutting machine in the video (a Sykes?) has the answer and although it is making a herringbone gear with a gap in the middle, it would seem that it could also make them without such a gap. The cutters in the film can easily cut all the way to the center. All that is then needed is a way to remove the chip before the cutter on the other side reaches that point. Or, perhaps with just a few thousandths of cutting past that center point, the chips would be completely free and just fall out.

I also see no problems with making them in two halves and bolting them together. The separation forces could be calculated and a sufficient number of bolts that are large enough to handle that force could be employed. I do not think that just pinning them would be a good idea.

crossthread said:
Dan...there is no gap. That is what is confusing me as to how these are made much less how am I going to duplicate it. I think the only way to do it is in two halves and a ring of bolts to hold it all together. Thanks for the help guys.

EmGo · Jan 21, 2021

implmex said:
The Sykes shaper makes a helical cut on one half and then immediately makes the corresponding cut on the other half.
So the first cutter has to push the chip right to the centerline but the second chip can fall free into the developing tooth gullet.

It would be nice if it worked that way, but in reality each cutter bottoms out on the opposite tooth. They get away with it because the feed is so light that the chips are very thin. A Sykes is really slow. Sunderlands also, but they are a touch more interesting since they use a rack-shaped tool instead of a pinion-shape.

Yes, one cutter sort of clears the way for the other (in both directions), but in the end they both bottom out on solid metal. It's a rough deal for the cutters.

Sykes and Sunderland are the only two methods to do a continuous tooth.

So if you want to make these yourself, you have no choice but to do it in two halves.

3d printing

Practically speaking, except for a few cases where you need a continuous tooth, most people have gone to double-helicals (have gap) instead of herringbones.

That's not going to work for him either, though, unless he has a Maag. Hobbing leaves too big a space between the teeth to practically make a substitute, shaping would leave a narrow enough gap but then he'd need 30* guides (herringbones are almost universally 30* helix angle) and a 30* cutter, that's $$$$ and a long time, Maag is the only practical answer and if you go that far, may as well hunt down a Sykes and do it the original way. There were some people in Texass who had them, Lufkin uses them, and O'Connell up in New York.

Or make it in two pieces, yup

Milland · Jan 21, 2021

EmanuelGoldstein said:
Sykes and Sunderland are the only two methods to do a continuous tooth.

Is this correct? You can do a full herringbone profile using a ball endmill and a 4th axis, right? There will be a small radius at the inside junction, but either relieve it there a bit if you have to mate to a existing "sharp" gear, or radius the peak of the outside face if you're making a set.

Or if you insist on the sharp internal profile, hell, EDM it as a finishing op.

EmGo · Jan 21, 2021

Milland said:
There will be a small radius at the inside junction ...

But then it wouldn't be a "real" herringbone

Some real early cast ones had a radius insted of a sharp corner ....

Or if you insist on the sharp internal profile, hell, EDM it as a finishing op.

That'd be fun ... but even less practical than just sending it to someone with the correct equipment

Two-piece is where he's going to end up on this, or sending it out. Everything else is way beyond reasonable price-wise.

pat pounden · Jan 21, 2021

so sidney gears were made by a sykes machine??-thanks,pat

EmGo · Jan 22, 2021

pat pounden said:
so sidney gears were made by a sykes machine??

Yes. At first glance herringbones look great but the downside is, no way to grind them. So you win some, you lose some ...

Since then they have found that double-helicals with a gap actually work better, as the teeth of a real herringbone don't flex equally across the face, so you can transmit more power more accurately with the gap, but Sidney was a long time before we had tools to measure things like that.

One place they are still favored is for pumps, where you obviously can't have a space between the teeth and helical is better than spur, less pulsing. Other than that, and for mud pump drives* (Lufkin is the biggest manufacturer of continuous-tooth herringbones ? used to be, anyhow) where they are replacements, continuous tooth has mostly gone away.

*Someone tried straight helicals as replacements here, but the side loads pulled the bearings right through the housings. Very low speed, high load.

Dan from Oakland · Jan 22, 2021

"Your statement, "Normally, two piece designs are such that the gear halfs are forced together and not appart by the axial load from the gear mesh." can only apply to one of the two meshing gears.

I took statics almost 40 years ago and consider myself pretty good at the subject even now. . . thanks!

Forget the glue, forget the ball end mill and EDM BS, make a left and right hand pair and bolt them together. Throw a few dowel pins in if it makes you sleep better at night.

Meethod of cutting herringbone gear

Titanium

Diamond

Diamond

Diamond

Cast Iron

Titanium

Titanium

Diamond

Titanium

Diamond

Titanium

Cast Iron

Diamond

Diamond

Diamond

Diamond

Diamond

Cast Iron

Diamond

Titanium

Similar threads