Broaching tries for a bigger bite.
Broaching's speed has always been attractive to the auto industry. They are well aware that the broaching industry has been preaching productivity for years.
But large gears like flywheels have been relegated to relatively slow hobbing processes because available broaching machines didn't have the muscle to do 12 to 20 gears and show the quality levels they can achieve.
Now one broaching company, General Broach & Engineering Co, Mt Clemens, MI, has put its money where its mouth is. They've invested $1/2 million in development funds to build the largest tonnage pot-broaching machine ever made. It is demonstrating to the auto industry that 16 flywheel ring gears can be punched out at impressive rates of 380/hr--from ring blank to finished gear form in 3 sec of actual cutting time, plus 6 sec of loading/unloading time. And General Broach is challenging the industry to replace 15 of their existing flywheel-gear-hobbing machines with each super pot broach.
Obviously, the auto industry is very interested in this development aimed at the high-volume auto and small-truck flywheel market, but there is more to this story than just promises of productivity. The hobbing people are not taking this lying down. They have a good case to present also. But first, let's look at the General Broach development.
The General Broach machine is rated at 80 tons and 275 hp, and is the result of a three-year, independently financed development program. The largest previous pot-broaching machine, also built by GB, was 35 tons and 75 hp.
According to an enthused Robert E Roseliep, GB president, "We expect our new machine to revolutionize flywheel production. Pot broaching is tremendously superior in accuracy as well as speed to current methods of machining the ring gears that are welded onto the flywheel in a two-piece construction. Even more important, the pot-broaching process will now permit one-piece flywheel designs with the gear teeth cut directly into the flywheel, at rates of 500 to 600 parts/hr without an operator. The reduction in fabrication steps plus high broaching speeds make this a dramatic productivity breakthrough in engine design and manufacturing.'
Automakers have been working for some time on a single-piece flywheel, looking at a variety of ways to form the gear, including roll forming, stamping, and hobbing. But the stamped or formed gear still has to be hobbed, and the one-piece fabrication savings are offset by the cost of cutting teeth--so far yielding no net manufacturing benefit.
The current ring gear is 1060 steel bar stock that is roll formed into a circular shape, butt welded, coin flattened, coined or machined round, and then hobbed. The gear is then welded at several points around the periphery of the flywheel.
That's a lot of manufacturing steps, and the performance of the gear has on a problem area for a long time. The consistency of the hobbed gears causes mine assembly problems, requiring ?? adjustments of starter position some cases. Starter noise, alignment, ?? life, plus instances of broken welds ?? the gear (and the gear flying loose at ?? rpm and sawing through the transission case, car floor, and beyond), all suggest that this is less than an ideal design to manufacture with consistency. It has been said that none of the starter ring gears in use actually meet design specifications fully.
The broaching promise
Bob Roseliep feels his broached gear could solve a lot of these problems. "The pot-broached gear is machined with pitch-line run-out well within 6 sigma ( 3) tolerance. We have delivered sample broached ring gears to some automakers, and they have run them through their gear lab, static testing, and dynamic engine testing, and have pronounced them far superior to anything they ever hobbed.'
Roseliep also claims other user advantages over hobbing. He feels his broach is a simpler machine with reduced maintenance requirements and no need for periodic tooling adjustments. He cites the "hidden' costs in the hobbing process for repairing and maintaining the arbors that hold the stacked gear blanks--hidden because they are lumped into plant maintenance and not added to hobbing tool costs. Yet, he does not claim any net tooling-cost advantage over hobbing.
Another comparative advantage is that the hardened portion of the ring blank at the butt weld sees only two of the 144 broach teeth at any one time and, because the weld position is randomly distributed, wear is distributed uniformly over all the broach cutter teeth. In comparison, hob teeth have to plow through every weld on every gear.
In short, the advantages claimed over existing hobbing methods are higher speed, a payback of $1 to $2.50/part, less use of floor space, reduced downtime, and a superior gear.
High mileage, low noise
The General Broach machine design and tooling arrangement provides rigid metal-removal action that eliminates the noisy vibrations of column-type pot broaching or conventional ring-type broach tooling. Conventional pot broaches cantilever the pot and hydraulic ram on a column. General Broach mounts the pot on a horizontal base with the hydraulic cylinder directly below and concentric to equalize the cutting forces. With no sound abatement, broaching noise is less than 80 dBA.
They expect a tool life of 20,000 pieces between sharpenings. They've run 12,000 ring blanks on the experimental broach and seen no appreciable tool wear. The broaching elements are simple vertical bar shapes, 72 of them, one for each pair of teeth. They are elementary enough to be sharpened or even made by any broach-tooling supplier. The vertical positioning permits staggering of the tooth pitch to maintain an even thrust on the workpiece as it moves up through the pot. This design also allows shorter broaching strokes, lighter pots, and faster production rates.
The key element in General Broach's is the proprietary toolholder that secures these vertical tool elements. It is made of a special zero-thermal-expansion coefficient material that is cast to fill the gap between the broach elements and the pot itself. This requires machining precision mandrel to make the casting, and this is the long lead-time item (40 weeks) in building the broaching machine.
According to Roseliep, the approximate $1/2-million machine price tag breaks down into $35,000 to $50,000 for the broaching elements, $100,000 for the holder, and $375,000 for the remainder of the machine structure, hydraulics, and controls. This does not include automation for loading the blanks into the machine's bottom slideway. With some hydraulic modifications, they feel this machine could produce gears up to 22 dia. The quality level of the gears produced was quoted by General Broach at AGMA 6.
When asked his estimate of the potential market for machines to broach automotive flywheel gears, Roseliep said 20 machines worldwide, with half of those in the US. Obviously, there are many other application areas for a broaching machine with this capability.
For more information from General Broach on this development, circle E4.
Does this mean the end of hobbing of flywheels? Not exactly, as we found in discussing this development with Sidney Ashcroft, product manager, parallel-axis systems, Gleason Machine Div, Gleason Works, Rochester, NY.
"Flywheel ring gears is an area where broaching is a definite threat to traditional hobbing,' Ashcroft admits. "Ford has mentioned to us that they are considering the pot-broaching alternative. This is an ideal part to broach, and we readily acknowledge that there are other areas where broaching offers time-cycle advantages provided the gear-quality levels are low enough to permit broaching, such as splines and synchronizer teeth on transmission gears.
"I don't think General Broach can establish an AGMA rating for the gear they are producing. This would be very difficult for anyone to do--define an accurate overall gear-quality figure for this type of gear--because each gear parameter could be labeled with a different rating or quality level. It's a question of what tolerances to place on various gear parameters.'
"The common method of making starter gears today is to machine them in a stack 6 to 9 high, using a conventional relatively low-quality (Class C) hob cutting a high number of threads (five to seven) simultaneously for the optimum productivity. The hobbing process is limited by the rate of rotation of the work past the hob, and the speed and feed of the hob entering the part. A large diameter means a wide face width with lots of teeth and long cycle times. The hob cuts continuously around the stack of 20 to 30 gears in a spiral fashion, and it can take as much as 14 min to hob a stack of 22 gears.
"So our answer to this problem in 1979 was the Gleason G-Trac --essentially a tracked hob of infinite diameter. (See Tooling & Production, Sept '81, pg 92.) With it, you can simultaneously cut teeth in the whole stack height in 1 1/2 revolutions of the hob track and complete stacks of ring gears in 2 1/2 min.
"We have two of these machines operating in Germany, but we are no longer offering it on the market here. It was a very expensive machine that demanded very high part volumes. You could not afford to change it over often because of setup time and tooling costs. We found that the volumes to make this machine practical in this country were just not common. People here need to change over very quickly from Gear A to Gear B, and the volumes between changeovers have been shrinking. But the people using them in Germany love them, and one of them just bought a second machine.'
The latest gear hobber
"Conversely, the standard hobbing machine today is very versatile; it is easy to change it over. We have just supplied a new hobbing machine to Lloyds Ltd, West Midlands, England. They had an English hobbing machine and one from Germany, each capable of hobbing a stack of flywheel ring gears in 15 min. Our machine is hobbing the same stack in less than one third the time.
"So the question is, what are you comparing? If you compare your new machine with an old hobber that's been running for 20 years and has limited productivity and cannot use the latest tool technology, it is easy to show improvement. With a modern, heavy, rigid machine like ours (and some of the German models), you can make radical improvements in productivity, even though the basic process is the same, and the method of handling the parts is the same. The only difference is the capability of the machine and the capability of the tooling. The machine we sold Lloyds hobs 15 ring gears in 4.1 min, or 16 scc/part, which is 219 parts/hr. The only faster hobbing machine is our G-Trac.
"The tool is a titanium-nitride-coated hob that is 5.5 10 and costs about $1800/tool in volume. With an average of 15 regrinds, it gets about 18,000 pieces/hob. Tool costs are a very small part of the process costs, about 60 cents/piece. The machine cost is about $300,000, and this is fully automatic, loaded with a magazine.
"The reshaping benefit claimed by broaching is comparable to the benefit from our own machine. The concentricity of the gear teeth with the bore was the major concern of the man from Lloyds. The blank bores are never very good, the cleanup on the welded joint is very inconsistent. He would have accepted anything with 0.010. On the 2000 rings we did here for him to test out the machine, the maximum eccentricity was 10 times better, or 0.001. This is because the Gleason fixturing, a series of radially expanding pads, reshapes irregularities in the ring blanks.'
For more information from Gleason on their latest hobbing equipment, circle E6.
"At any rate,' Ashcroft concedes, "we're aware that favorable comparisons on starter ring gears are being made between existing hobbing equipment and the pot-broaching approach, but it would be more proper to make the comparisons instead with the latest state-of-the-art hobbing equipment. Yes, there are many areas where pot broaching is considerably faster than shaping, but not many where pot broaching is clearly superior to hobbing. Certainly not in gears of much higher quality than starter ring gears. The problem would be in maintaining accuracy, in transmission gears for example, where the AGMA quality rating is in the 10 area.
"General Broach will be faced with some of the same problems we had with G-Trac. A high unit machine price, complicated and expensive tooling, a need for great care in tooling maintenance, and the need for very high product volumes. A lot will depend on the tool life they can get. If they can guarantee that the machine and tooling will stay up for long periods of time under production conditions, that would be a significant plus. Certainly there are applications at Ford and GM with the necessary volume to justify this machine. But you have to be concerned about the flexibility of retooling sometime down the road, and it's much easier to retool a hobbing machine. I would guess that they will have a similar degree of response to what we had with G-Trac. They won't sell the 20 they would like to, but something less than that.'
Photo: Their new process takes 9 sec to transform this 16-dia blank into a flywheel gear with superior dynamic characteristics, General Broach reports.
Photo: Top-mounted gear ejector mechanism rotates to the rear for access to the pot broach and its 72 vertically stacked cutters.
|Printer friendly Cite/link Email Feedback|
|Author:||Sprow, Eugene E.|
|Publication:||Tooling & Production|
|Date:||Oct 1, 1984|
|Previous Article:||Grinding machine innovations boost productivity.|
|Next Article:||Squelching stamping noise.|