1900 ERA HELICAL GEAR SET 2

[RBPrice]

I am trying to reverse engineer some missing gears on a 1900 era antique marine engine. Two of the four gears that drive the camshaft from the crankshaft are missing. I have a cartoon of the AS-MEASURED dimensions of the crankshaft gear and the gear on the end of the Cam Shaft attached. I know the Diametral Pitch of the gears, the Pressure angle and the number of teeth for the two existing gears and the two missing gears. Both sets are 90 deg. crossed shaft arrangements. I have assumed that both mating sets were created using odd helix angles to get the center distance to be what was needed to connect the drive and achieve the 0.50 Ratio; i.e., the camshaft rotates at half engine speed. The engine is fairly large, 25 HP and is not meant to run in reverse. It was a typical trick of the old time designers to mess with the helix angle in order to get a desired Center Distance.

My gear book library gave me some formulas for finding the helix angle if one knows the Center Distance and all of the above parameters. The calculations for the 12T - 9T set on the crankshaft came very close to the existing dimensions but the calculations for the camshaft gears does not seem reasonable.

I have looked in Dudley, Jones, Buckingham, and some others but cannot seem to find what I am missing. Jones and Buckingham have the same equation.

Anyone have some suggestions?

Thanks very much

Bob Price

Bob Price
A*G*M

 

[gearcutter]

After running some numbers on this problem; it seems that 8DP isn't going to work.

How do you know it's 8DP, 14.5PA?
Have you measured the base pitch to confirm this?

What are the helix angles, root diameters, measurements over balls of the gears that you currently have?
How did you measure the helix angles; protractor or on a gear cutting/measuring machine?
Remember that measurements over balls are taken in the transverse plane.

 

[RBPrice]

Hi Gearcutter - as you will note from the dimension sketch, all the information came from the guy with the engine: he is in Tennessee and I am in NY. He measured the teeth with some gauges and got the OD of the two existing gears with a caliper since both existing gears have even numbers of teeth. The distances from one shaft center-line to the other he got with a some straight edges, adjustable gauge blocks etc. Since the engine was manufactured in New Jersey it seems safe to assume that the gears are not Metric. And since they were made in 1900 or so, it is most likely that they are 14.5 Deg. PA.
I have attached a photo of the assembly which shows how he got the measurements.

My problem is that I got numbers for the Crankshaft pair that seemed quite reasonable: that is the Pitch Diam. and Center distance for the pair of gears were close to his measurements. The numbers for the Camshaft pair did not come even close to what he sent me and I suspect that the measured distance from the Camshaft to the vertical shaft is wrong.

This is not the first time I have worked on a problem like this. I successfully figured out a helical crankshaft gear for a 1911 3 cylinder diesel that had a 58.8 deg. helix angle pinion that drove a 31.2 helix angle gear to drive the Camshaft gears. That too, was a 90 deg. crossed shaft drive.

 

[gearcutter]

I suspect that the measured distance from the Camshaft to the vertical shaft is wrong.

I'm sorry; then I'm not fully understanding what your original post is for.

If you've proved your theories on the first pair but those theories aren't working on the second..............you've just answered your own question.

For me, or anyone else for that matter, to help you with this; I'm going to need you to answer my questions, otherwise it's just 'guess work'.

 

[RBPrice]

OK - I guess that answer confirms my suspicions.

Thanks for your help.



Bob Price
A*G*M

 

[tbuelna]

RBPrice-

Are you trying to do a historically accurate restoration, or does it just to need to appear similar and function well? If the gearset does not need to be a 100% accurate reproduction then I would suggest designing a new set of gears using the existing tooth numbers and shaft centers. Based on your description of the engine it does not sound like transmitted torque will be an issue for this cam drive. However, there will likely be significant torsional oscillations produced at both the drive end (crankshaft) and driven end (camshaft), and timing between the crankshaft and camshaft is also important. So you'll want to control backlash at each mesh and minimize axial play of each gear.

If you design a new gearset you need to pay close attention to normal contact ratio. Select your pressure angle and helix angle so that the mesh normal contact ratio is >2.0. I'd recommend starting with a 14.5deg normal pressure angle since it is commonly used and low pressure angles improve contact ratio. Based on your center distances and tooth numbers the teeth should have a relatively large whole depth which will help. But the small number of teeth on the 6T pinion and 9T gear may present some difficulty. Start with a helix angle of 45deg and adjust as needed to get the normal contact ratio 2.0 or higher.

Good luck with your project. It looks like fun!
Terry

 

[RBPrice]

Good Morning Tbuelna - thanks for the response

This project is restoration with the intent of running the engine at antique marine engine shows. Probably won't see more than 25 hours use per year. As I mentioned, the gears are 14.5 deg. PA and 8 DP which would be typical for that era. Just as odd helix angles were used to get the center distances to match what the engine layout dictated.

Last night I assumed that the OD measurement of the Cam Shaft gear was probably close to correct since the gear has an even number of teeth and therefore the owner could get a good measurement. Backing into the Pitch Diameter of the gear and then calculating the Helix angle gave me a number very close to 45 deg: 44.8

Using 45.2 HA for the 6 tooth Pinion I came up with a center distance of 1.944" vs the measured dimension of 2.215. So now I have to bug the owner to see if he made a mistake in that measurement. Crossed Axis helical gears are tolerant of small errors in center distance but not that much. Another consideration is that the vertical shaft is 0.750" in diameter (it fits into some cast housings fitted with grease cups) and the calculated PD of the pinion is only 1.061 so that is troublesome. The gears will run in the open so will have to be lubricated with that black gooey open gear lube stuff.

I am going to ask the owner the recheck his measurements (for the fourth time) and see what he comes up with this time.

Again, thanks for the suggestion and comments.

Bob Price

Bob Price
A*G*M

 

[tbuelna]

RBPrice-

Since this is not a 100% historically accurate restoration, then I would suggest simply making a complete set of (4) replacement gears. But as I noted above, the prime considerations with your cam drive are minimizing backlash, end play in the gear bearings, and getting a normal contact ratio as close to 2.0 (or higher) as possible

You could design a new set of gears using the same tooth numbers and shaft centers, while adjusting the DP to fit. You can then adjust the helix angles to optimize the normal contact ratio.

If you are squeezed for space between the small pinion root diameter and the .750" shaft bore, then you might consider changing to larger tooth counts and smaller teeth (higher DP). For example, you could simply double the number of teeth on each gear and reduce the size of the teeth accordingly. Or you could use some multiple of the gear tooth numbers that provides the same ratio, such as 20/15 and 9/24. This would provide more space between the small pinion teeth root diameter and the .750" shaft bore. The stresses in the gear teeth might increase, but I don't think you have a stress problem. And modern materials for this type of gear application (like aluminum bronze) are far superior to what was available 100 years ago.

Regards,
Terry

 

[RBPrice]

Hello Terry - all that is good info but my view is that we have two of the four gears and we know the numbers of teeth in the missing gears and we know the Pitch and the Pressure angle. Plus, there are two cast housings that surround the mating gear that meshes with the Crankshaft gear and a second housing that encloses the gear that is on the end of the Cam Shaft. The challenge is to work out the helix angle for that set of four gears on, what we think, are the center distances dictated by the geometry of the engine. The materials I will use to make the two new gears will probably be a medium carbon steel like C-1080 which will have adequate strength and some wear resistance. Manganese Bronze would be better but is too expensive.

Some further looking in both my gear book library and On Line uncovered some equations that I think will let me get the answers to the question of what helix angles were used to match what we think are the center distances. It is a bit laborious since one has to incrementally change the HA and then finish the calculation to see how close we get to what we need. See page 82 of Jones & Ryffel.

However, my wife wants me to stop fussing with this for a few days and focus on a list of must do things like pick up our daughter at the airport in about 90 minutes.

Happy Holiday to you and yours. Bob

Bob Price
A*G*M

 

[tbuelna]

Bob- With crossed helical gears having perpendicular shaft axes, the sum of the helix angles is normally 90deg. So if you know the helix angle of one gear, you should be able to determine the helix angle of the mating gear. The other variable you need to determine is the DP of the gears.

Due to the small number of teeth used on the 6T driving pinion in your second stage, it may use a helix angle closer to 60deg to prevent undercutting.

 

[RBPrice]

Hello Terry - I knew all that stuff going in; what I didn't appreciate is the extreme Helix Angles needed to make the gears work. See the attached copy of my calculations to see what they turned out to be for the gear set on the Crankshaft. I have yet to do the Camshaft set but expect to get a similar result. Staring at the photos sent to me convinced me that the angle guessed at by the owner of the engine was way off. Further proof that the old timers really knew what they were doing and they did it without any modern gadgetry like iPads etc.

Next step after some drawing creation will be to find a gear cutter that can make the gears.

Thanks very much for your interest and help. This Forum is a great place to go to for conversations that a person working alone has difficulty having.



Bob Price
A*G*M

 

[gearcutter]

Nice working-out sheet Bob.

I've checked your calcs and, while you've been able to mesh the helix angles at the measured centre distance; you're not going to be able to mesh the missing driven (gear) with the existing driver (pinion). Your arrangement drawing shows the original pinion as having a diameter of 69.85mm. Based on your calcs; the new pinion will have a diameter of 78.00mm.

Perhaps the original is some kind of stub pitch.
Perhaps we're dealing with transverse DP instead of normal DP.
Without the information that I've asked you to provide; it's very difficult to answer these questions.

 

[RBPrice]

Hello again Gearcutter - as I tried to explain, the helix angles of the existing gears are not accurately known since they are still on the engine. The teeth were measured with a new set of 14.5 Deg. PA gear tooth gauges and also, I am assuming that the PA is 14.5 because the gears were made in 1900 or so when that was the predominant PA used. What is known for certain are the tooth numbers 12/9 and 6/16 and that combination produces the correct overall reduction of 0.50: i.e., the Camshaft turns at half engine RPM. There are probably several other tooth combinations but that is not what is on the engine. The latest version of the shaft positions is attached.

I have tried to work backward from the information that the owner has provided. He says the Crankshaft gear has an OD of 2.750 so that leads to a Pitch Diameter of 2.500 assuming a normal Addendum of 0.125 for 8 DP gear teeth. But putting that into the equation PD = N/Pitch x CosA and solving for CosA produced 53.1 deg. So maybe that gear is a stub tooth with an addendum of less than 0.125 If the gear is a Fellows Stub then the Addendum would only be 0.100 and the Pitch Diam. would be 2.550 vs. 2.500 That produces a Helix Angle of 54 deg. Even using a Nuttall tooth would only reduce the Addendum by 0.019 in. Did they have Nuttall gears in 1900?

I am going to ask the owner to send me the Camshaft gear since that is readily removable from the engine. I can then make my own measurements. Getting the Crankshaft gear off is problematical since it is probably rusted in place.

Thanks for you comments Gearcutter

Bob Price
A*G*M

 

[gearcutter]

Why not get your customer to send you the whole crankshaft assembly?
At least then you'll be able make more accurate measurements of the driver.

The other thing to consider is that at that time; it was more common to use Circular Pitch than it was to use DP.

 

[RBPrice]

Not sure about dismembering the engine - But how would using Circular Pitch change things?

Bob Price
A*G*M

 

[gearcutter]

But how would using Circular Pitch change things?

What you need to be doing is reducing the pitch to try and bring the diameters down.
There will be a standard CP that is close to 8DP but smaller; 2/5 & 3/8 are close.

I've just tried 7DP and that seems to work well for the given centre distance. A helix angle of 58deg at the PCD of the pinion works out to a tip diameter of 69.84mm (2.7496"). Diameter for the gear works out to 57.68mm.

The final result doesn't look too bad at all.


7DP, 14.5PA by ronvol, on Flickr

 

[RBPrice]

Good Morning - I looked at using 3/8 Circular Pitch and could not seem to get a combination that worked without getting a very large diameter 12 T Pinion. As for the Crankshaft gear being 7 DP, I just checked with the folks at Ash Gear and they do have a 7 DP hob but doesn't that beg the question of how well does an 8 DP gauge fit into a 7 DP gear?

And sticking with the 8 DP and using Helix angles of 55/35 comes close so I wondered if the gears could have been cut with a short addendum on the pinion (Crankshaft) gear and a long Addendum on the mating gear. That, of course, weakens the tooth but there is not much torque being transmitted.

And Nuttall Gear was started 1887.

Bob Price
A*G*M

 

[Occupant]

Something isn't right here: I used 7DP and 55/35HA just like you for the second set (6 and 16T) and x=0.5 profile shift on the small gear and came up with a CD of 2.214 with the large gear at 3.076OD.

 

[RBPrice]

WOW - so maybe all of this is so hard because the gears really are 7DP and not 8?

The owner is a retired machinist and it is hard to believe that he could have used and 8 DP gauge to measure 7 DP gears and not noticed something wrong. Guess I need to talk to him about this.

Thanks Occupant

Bob Price
A*G*M

 

[RBPrice]

Occupant - using 7 DP I got a PD for the 12 T pinion on the Crankshaft of 2.816 which is quite a bit larger than the measured OD dimension. This was using Helix angles of n 52.5 and N 37.5; the center distance calculated to 2.219 in. which is certainly close enuf.

Bob Price
A*G*M