Worm rack problem: for your consideration

[Forrest Addy]

In looking through Fred Colvin: "Planing, Shaping, and Slotting" a Lindsay Publications reprint of the 1943 original I ran across a novel table drive. It was a worm and rack design where a motor driven worm engaged a rack attached to the planer table; rotating the worm against the rack resulted in an axial motion of the table.

Here's the novel part. Instead of the rack being cut to the helix angle of the worm and the worm's axis running parallel to the rack motion, the rack is cut straight across and the worm is angled across the rack following its helix. The angle permits the driving shaft to exit the side of the machine at an angle so the motor tucks neatly behind the column.

Five advantages accrue to this arrangement: a large drive reduction, intrinsic stiffness, compact design, low parts count, low manufactured cost. Disadvantages are: highly sensitive to continuous lubrication, expensive oddball worm, and substantial modification of the machine structure is required.

Thanks to the angled worm, tooth contact progresses across the rack avoiding concentrated wear. The progression of contact distributes wear and permits higher loading; the line of tooth bearing is constantly moving. Planer drives are typically 12 to 18 revolutions of the motor per ft of table travel. Thus large axial forces are produced with a single reduction consisting of only the lead of the worm instead of a train of expensive gears.

Since the worm is inclined to the rack the worm's normal pitch would have to be cut to suit it meaning an oddball worm and compound gearing the machine that cuts it. Since we have a helicoid contacting a rack tooth, tooth loading would be line contact where the axial thrust is taken by the tooth flank more or less divided by the number of worm threads engaged. Thus lubrication under heavy load would be sliding contact in boundary mode. Big potential for scuffing.

Strictly as a mental exercise I've been evolving the drive to suit my Rockford planer. I've been assuming the following preliminary design parameters.

- Direct drive from a 1200 RPM 20 HP motor rated 1.15 service factor

- 2 diametral pitch rack

- 5" pitch diameter worm

- 14" length of worm engagement with the rack

- 12,000 lb of thrust at 80 ft per minute by overloading the motor 20%. (note duty cycle)

- 150 ft per minute max table velocity

Thus about 8 complete threads will be engaged. Dividing thread engagement count by total thrust equals 1500 lb per each thread engaged.

- Rack material: 4140 HT to Rc 35

- Worm material: 8620 carburized 0.030" deep HT to Rc 58 .

- Worm inclined by the the helix angle (about 5.7 degrees).

- Flood lubricated with an EP worm gear oil.

- Max duty cycle would be 50% averaging 15 strokes per minute for up to 20 minutes at a time.

- Annual utilization would be 1000 hours where 400 hours would be at max duty cycle.

The manufactuing data and actual machining of the component parts are routine machine shop technique. The drive design is straight forward. The rack can be purchased to most any gearing class affordable from a dozen sources as can the oddball lead worm if it's not made in the shop. Installation and alignment requires judgement and care but, again, nothing out of the ordinary.

We have line contact not area. So the question I've been pondering as an exercise narrows down to this: how do you calculate drive longevity?

Forgot to add: control over direction and speed of the motor would be via a VFD and PLC logic. Adjustable limit stops used in the tradtional manner would set table stroke.

[This message has been edited by Forrest Addy (edited 02-20-2003).]

[Al Messer]

Go with a modern hydraulic system: easier to control and less complex. Won't keep you awake at night trying to figure it out. Uses stock materials, fittings and controls. Rig it up like a surface grinder controls.

[Forrest Addy]

It already has an excellent hydraulic system. My posting was intended to be pose a mental exercise in machine modification employing novel (for these days) design features.

I confess I have a powerful prejudice against hydraulic drive. They're noisy, inefficient, and sooner or later they leak.


[This message has been edited by Forrest Addy (edited 02-20-2003).]

[jdk]

Way over my head, I'm still trying to etch some lines in a dial. But maybe someday, guess I can dream.

[Spin Doctor]

Hydraulic Systems, Disadvantges
1) Noisy; Yes they can be but some of the systems we are using today are a lot quieter that the older types. One thing that has helped a lot seems to be the fact that most modern hydraulic systems today are using pumps that are not required to suck oil due to their being below the tank

2) Ineffiecent. That's a though one to beat. To damn much power gets converted to heat.

3) Leakage. Seals are getting better yet using smart design the problems can to a large degree be controlled.

Advantges

1) The ability to transmit power into tight places and package it in fairly small devices.

2) Controlabilty. With the advent of modern electronics a lot of engineers seem to of forgotten that hydraulic systems can be extremely versitile. With the deleopent of the position sensing cylinders they will do damn near any thing a ball screw will do.

But this is all probably a moot issue now because of the advances in linear motors.

[darryl]

That angled worm drive sounds interesting, however, there must be a high wear factor, and play developing quickly, not? And forces distributed in many directions, leading to energy losses and inaccuracies. Maybe this is one of those times when the good old threaded rod is hard to beat, or the good old crankshaft, connecting rod setup. jmhoitwhotm.

[Forrest Addy]

Spin:

It's very difficult to run hydraulics at very low speeds (like milling feeds) with out surging and stick slip. One of the reasons for the electric drive is to convert the machine to a planer mill where shifing between planing and milling can be done at the flick of a switch.

Darryl:

Well, you have to consider the scale of the machine. 20 HP. 7 ft stroke. 12,000 lb max table thrust. High speed lead screw drives have limitations.

Rear axle gears have much the same kind of sliding/rolling load bearing as this worm rack I'm thinking about. It's worked on Sellers planers for a couple of generations. I managed to get a drawing of the Sellers worm rack drive posted:

http://www.practicalmachinist.com/ub...ML/001485.html

You can see from the drawing the tooth bearing progresses across the rack as the worm turns and the load never concentrates at a single site.

Multi-stage geared reduction meshing with a bull gear and rack would be superior but far more expensive to execute in a machine not designed for it. If I had my druthers I'd have an electric drive G. A. Gray planer and the Rockford would be long gone.

By threaded rod I assume you mean accurately machined lead screws and bronze nuts, not the rolled thread stuff you buy at the hole center to bolt together a planter.

A lead screw capable of transmitting 20 HP at 150 ft per minute would have to be 4" diameter quadruple lead Acme fed with oil under pressure. It wouldn't be any more efficient than hydraulic drive but it would be stiff if the bearings were properly anchored to bed structure.

I considered a lead screw drive but making an 11 ft (circumstances force this) heat treated alloy steel lead screw with 9 ft of thread would be a big project for my equipment.

[This message has been edited by Forrest Addy (edited 02-23-2003).]

[Oso]

Seems like the slow speed problems of hycraulics are more from pump capacity and running too slow. If the fluid moves smoothly, the ram is gonna move, that is the nice thing about hydraulic equipment.

So if you have a smaller volume capacity pump you can run at normal speed, you get slow speed on the system.

The rack drive has the problem of no high speed, it is all low speed, unless the worm spins awful fast. Maybe you don't want a high speed return, but....

What are the reasons the above isn't true?

[Spin Doctor]

One of the worst mistakes I've ever seen in regard to hydraulic systems controlling machine tools is the use of the control device on the working side of the actuator rather than the exhaust side. We're using hydraulic actuators on machinery performing every thing from cutting woodruff keyways in crank and camshafts to precision boring operations with no problem whatsoever. Is it easier to dial in the exact tool advance that you want per revolution with electronics, sure. Besides replace the cylinder with hydraulic servo drives and it's possible to creep the feed at incredably slow speeds.

[darryl]

Forrest, I was referring to leadscrew/bronze nut, and not threaded rod as you assumed. You make many good points about the worm rack drive, not the least of which is the fact you pointed out about rear ends having a similar tooth action. That analogy goes a long way towards suggesting the longevity of the worm/rack, esp. at the power levels, 20 hp, or greater, as in vehicles. I had a thought, how well does diff lube (I'm assuming you'll be using some similar high pressure lube) do it's job at very slow speeds, will it just squish out, allowing metal to metal contact, and accelerated wear? I had another thought (guess I'm still alive) could the worm shaft have a universal joint on it, to allow the worm being lifted clear of the rack, to allow some other mechanism to do a rapid reverse. Maybe that's too much add on trouble.