Continuous-dress creep feed - should we be taking it more seriously?

Creep-feed grinding is just beginning to gain proponents in this country. Deep cuts, if you can pull them off without burning the part, are a much faster route to payday than conventional reciprocating methods. Ten to 20 times faster.

Now, along comes continuous-dress creep feed or CDCF. It offers another giant step forward in grinding productivity--another 10 to 20 times faster deep grinding or parts to shape in a single pass. Cycle times measured in seconds!

But at what cost? That's the big question in the minds of most of us. Who can afford these multimillion-dollar machines and the part-handling sophistication required to take full advantage of a process that is two orders of magnitude faster than reciprocating grinding?

To help sort out this grinding controversy, I talked with two leading CDCF suppliers in this country (there are only three and they are all based in Europe) and with the man who started it all, Dr. Stuart Salmon, formerly with Rolls-Royce, more recently with GE's Advanced Technology Section, and presently president of his own consulting company. And finally, I got a contrary view from a leading domestic grinding manufacturer. Origins of creep feed

It's important to realize how recently creep-feed grinding technology was developed, and that it all started by accident. According to Ted Davidson, vice president and general manager of Elb-GERH USA Inc, it happened in 1958 at his parent company, Elb-Schliff, Babenhausen, Germany. "Gerhard Lang, the son of Elb's founder, was doing some experimental deep-cut electrochemical grinding in the lab when the electrolytic current was interrupted during one test. When he realized what had happened, he was amazed at how good the resultant finish was. So he showed it to his father and they decided it was definitely something to explore, and they developed the first creep-feed grinding machines shortly thereafter."

In the intervening 25 years, creep-feed grinding has become very well accepted in Europe, Davidson reports. "In the auto industry, Mercedes, BMW, Fiat, etc, all grind their rocker arms on Elb creep-feed machines. BMW and Mercedes each have about 20 of our machines, and are also doing steering-gear knuckles and racks. The initial applications are high-performance parts, but the real impetus is precision, good finish, and a good, tight fit between mating parts."

Hermes Machine Tool Co, Fairfield, NJ, is the US representative of Blohm Grinding Machines, Hauni-Werke, Hamburg, West Germany. According to Hermes prsident, George Bard, "Creep feed is definitely here to stay, and regardless of who delivers the machines, they will have continuous dressing."

Although creep-feed technology has become fuly developed in central Europe, there are really no domestic suppliers in the US, Bard feels. "It has not been developed here for several reasons--users don't have the necessary trained personnel, and it's a technology that requires a perfect marriage of grinding wheel to workpiece. But it certainly has caught on in Europe, where, since 1980, it is being superceded by continuous dress."

Elb-GERH's Davidson agrees. "The key CDCF markets in the US are turbine blade and vane production. Creep-feed grinding is still quite new here, so this is a very big potential market for us. I can't think of one creep-feed machine we've sold recently without continuous dress--it's becoming the standard. The technology is no longer in question with any of the bigger user companies." New stability and speed

But this kind of improvement puts greater demands on the machine stability and workpiece stability. "You cannot clamp the part mechanically," Bard continues. "It must be more stable--the limiting factor in productivity is the rigidity of the workpiece capsule.

"In turbine-blade grinding, encapsulation is required for workholding automation, throughout the different operations. It gives uniform gripping capability for either full-capability robots or simple pick-and-place units as the blades move from grinding, washing, drying, gaging, marking, etc. The encapsuled part presents precise locating points for all the processing operations. When the part is finished, it is decapsulated. The major supplier is Fisher Gauge Ltd, Peterborough, Ontario, Canada. So far, everyone in turbine-blade CDCF is using their encapsulation equipment."

The speed of CDCF demands automation, even from the smaller, job-shop-type users, Bard feels. "An operation that used to take 4 or 5 min can now be done in seconds. For a grinding operation to remain competitive, it must adapt to this technology. You can't spend minutes loading and unloading. And with this faster flow, one or two machines now do the jobt of eight or ten creep-feed machines."

Another key element is wheel consumption, particularly the time required to change wheels. "We have also introduced automatic wheel change," Bard points out, "changing a five-wheel set on one advanced machine within 3.5 min, including automatic balancing on the machine without operator intervention. This produces tremendous timesavings.

"You can then tie the entire line together with a computer, monitor where the errors are with feedback gaging, and do all the management control by computer. The net result is the new line replaces many machines and at least 100 people who used to work around the machines.

But at what price? "A single machining system fully automated with subsidiary handling and gaging operations comes to at least $3 million, and the system price goes up from there depending on the degree of automation, how many operations, etc. Robots can replace the conveyor when more than one machine is located in-line, and you can reach a point where automated guided vehicles--robot carts--are more efficient and economical than a transfer-line-type conveyor. The cart would hold many small parts at a time, say 60 or 80 parts/cart." Adaptive control 

Elb-GERH's Davidson feels they have a competitive edge with their latest development--adaptive control. "This controls both the infeed rate of the diamond dressing roll and table-feed rate to maintain the most productive grinding combination. We use a spindle-bearing strain gage sensor to monitor spindle forces--the sole feedback sensor. Using preprogrammed parameters, the control ensures that we are getting the most productive cut while maintaining the part's metallurgical integrity. This tailors infeed (wheel and dress roll) to the specific pgrinding task at hand and avoids eating up wheels unnecessarily. Straight (incremental) continuous dress can be a real wheel eater! With adaptive control, we can maximize wheel life." Aerospace leads in applying CDCF

As is usually the case, the first users of this innovative technology were in the aerospace industry. As creep-feed technology developed in Europe, Davidson reports, "The first big user was Rolls-Royce for the grinding of turbine blades. They have over 500 of our machines, including a fully automated seven-cell line started up in 1981." See Rolling with Rolls, Instant profiles, and Creep in action.

Aerospace also leads the way in applying creep feed and CDCF in this country, and they are doing it with a maximum of automation. According to Bard, "The continuous-dress creep-feed system we're supplying for Pratt & Whitney is a fully automated, computer-controlled system for manufacturing shrouded and unshrouded turbine blades. We have the complete system responsibility for start-up--from encapsulation with robot handling, to the verifier gage, the grinder, a washing and drying system, gaging with feedback to the computer, and laser marking. Delivery will be in about 18 months.

"The Allison Div of detroit Diesel is also considering such a system, another very large installation. In Europe, Rolls-Royce has our latest system, an automated line that is just starting up there now."

This is Blohm's first cell-line installation at Rolls. Elb's Davidson counters that this Blohm achievement was simply a move on Rolls' part to establish a second supplier, not because of any dissatisfaction with their Elb machines. As Davidson points out, "Rolls production right now is way down. They have no new orders for domestic engines. They don't really know what parts they will grind on the new Blohm automatic line. The Elb automatic line produces a blade in 45 sec--a year's worth of parts in four months--so even it is only used on a limited basis. We have always wondered why Rolls bought these new machines from Blohm."

GE's Aircraft Engine Group is also moving into CDCF automation for turbine blades and, according to Davidson, has just picked an automated Elb-GERH line after a tough competition between Elb, Magerle (Midwest Precision Services, Roselle, IL), and Hermes/Blohm.

Blohm, Elb-Schliff, and Magerle have adopted continuous dress because it was a relatively small change for their creep-feed machines. Their marketing emphasis is in this country now because here is where the largest applications are. As Bard explains, "Very few jet engines are being built in Europe now. In the USA, creep feed is being used on some stand-alone machines, but the major initial use of continuous dress will be in the larger installations. This is the real competitive battlefield for us."

Adds Davidson, "IBM has over 200 of our creep-feed machines, making computer parts so proprietary we don't even know what they are. We've been working with Ford on steering applications and several engine-component manufacturers on rocker arms. Yet, there are only three or four creep-feed machines in use today in the US auto industry, and none with continuous dress! The only CDCF installations are aerospace." Future application areas

What future application areas look good right now for CDCF? "Those areas were milling with subsequent grinding operations is done now on through-hardened material," Bard replies. "CDCF would be faster, more accurate, and eliminate secondary grinding. If you now mill and have to harden, this works the workpiece, and you have to grind it again. Now, you can take the through-hardened material and simply manufacture the part in one operation. This eliminates the milling. Actually our process is a very fast and accurate milling process that allows you to fully shape or profile your workpiece.

"Once of the best applications is aircraft parts with sophisticated materials that are very hard and tough to machine--they cannot be milled. CDCF will be used everywhere you can replace milling with grinding. Our machines are doing thread chasers ground from the solid on through-hardened material. We can grind injection-pump rotors of any shape from a solid. For low volumes, they can be done simply one at a time with a flexible grinding system (FGS, see Tooling & Production, July '83), relying on the numerical control ability of the machine to reshape the wheel. But when the process is high volume, we can automate it and add continuous dress with a diamond roll (but minimal shape flexibility)."

So the future trend in CDCF technology will be the combination of high-volume and low-volume grinding capabilities in the same machine--creep feed with continuous dress for high-volume work, and for those workpieces for which you don't have a diamond roll, you will use the system as a flexible grinding system, sculpting that form into the wheel using two-axis CNC.

"The CDCF process is not only for high-performance parts," Bard points out. "Other materials we can grind include high-speed steel that could not be creep-feed ground previously. In most cases we don't have to use expensive diamond wheels for dressing, because the continuous dress allows us to have an open, fresh grinding wheel at all times. For lower-performance parts, we use conventional creep-feed grinding."

Elb's Davidson also sees a competitive advantage in being able to merge creep-feed and conventional reciprocating grinding. "If you wnat to be able to reciprocate and creep, you do not want a ball-screw drive because of the reversing wear reciprocating causes on both the ball screw and nut. So we have developed and patented a precision rack and worm that can handle both with no adverse effects." Doctor of continuous dress

Dr Stuart Salmon studied the creep-feed grinding process in the early '70s as a young aprentice at Rolls-Royce, Derby, England, and then at the University of Bristol under the sponsorship of Rolls and the British Science Research Council. He developed the continuous-dress technique and was awarded a PhD for this work in 1979, the same year he joined General Electric's Aircraft Engine Business Group. While internationally known for his CDCF work, he is still relatively unknown in this country, even by leading grinding-machine builders.

As he explains, "In the early '70s, Rolls was heavily into creep-feed grinding. The characteristic problem with pushing creep feed to its limit is that there will always be thermal damage to the part. It's not grinding-wheel breakdown or machine vibration--the process limit is always thermal damage. The notion that there is some kind of controlled thermal benefit--for example, the idea in the Dec '83 Modern Machine Shop article by Ken Gettleman and Professor Shaw that creep feed heats up the workpiece material ahead of the wheel--is fundamentally and absolutely wrong! There's no thermal advantage. In fact, that's the worst possible case! You don't want to generate heat at all. (But to be fair, the two MMS articles by Mark Albert on CDCF are excellent sources of basic information: "Taking The Creep Out Of Creep-Feed Grinding," Nov '82, page 80; and "Getting The Most From Continous Dressing," Dec '83, page 70.)

"In the early '70s, Rolls-Royce had over 100 conventional creep-feed machines for doing deep-cut, one- or two-pass grinding of aerospace materials and superalloys. They were burning parts and creating scrap. I was charged with researching creep feed at the University of Bristol, trying to identify why the part burns and where all the forces lie. That creep feed worked had been accepted, but nobody knew the real reasons why.

"We were always using a very porous, open wheel, typically 60 to 80 grit size. This was obvious from the beginning--large pores are needed to carry the cutting fluid around the arc of the cut. So we measured the grinding forces, and identified that with water-based cutting fluids there is a transference in the thermal conductivity of the cutting fluid as it changes from nucleate boiling to film boiling. As the wheel dulls, we create more frictional heat than cutting energy, and the heat goes into the arc of cut. As the wheel moves further down the arc of cut, it takes with it the fluid to supposedly cool the workpiece, but the fluid warms up, burns out, and becomes one long bubble. Instead of nucleate boiling, all the heat goes into the workpiece, and it burns.

"This research continued through the '70s. Continuous-dress technology was not developed by a machine-tool builder. It was entirely a development that I did in Bristol as a part of a research program funded by Rolls-Royce and the British government.

"Because of this dual funding, no patents were taken out, and the rights were up for grabs to the British machine-tool industry. Unfortunately, Britain didn't have any decent creep-feed machines, so we had to go to the Germans for the hardware, where the world goes for creep-feed machines today.

"The 168 or so Rolls machines in the late '70s were mostly Elb machines, so Rolls went to them first for continuous-dress modifications--essentially, adding the infeed mechanism for the dressing roll, and devices to keep the wheel parallel with the bed movement of the machine because not only are you feeding the diamond roll into the wheel, but infeeding the wheel spindle into the work. It must compensate continually for the wheel getting smaller. This is a big leap in sophistication.

"Another difficulty was the machine rigidity required for that very fine dresser infeed rate--between 10 and 80 millionths/revolution of the grinding wheel. Because continuous dress boosted stock-removal rates (over creep feed) 20 times, Rolls was forced to automate. Their first CDCF installation in 1981 was one 12-machine, seven-cell line.

"At this point, the Germans and the Swiss (Elb, Magerle and Blohm) developed machines for the continuous-dress creep-feed market. You can now buy stand-alone CDCF machines, or for manufacturing operations such as turbine-blade applications, with workpiece encapsulation."

According to Salmon, no grinding manufacturer in this country can make an adequate CDCF machine today, and take advantage of this new grinding-market opportunity. "The impetus in the US is from the aerospace people because of their difficult-to-grind materials and the high accuracy they require. However, there are many milling and broaching applications that could also be benefiting from CDCF. This is at a matter of convincing the industry that they must not simply consider changing over from present grinding applications. They should be examining all their milling, grinding, and broaching operations.

"We're looking at applications in pneumatic pump-motor rotors, deep grinds through fuel-injection rotors, deep-groove gear froms, typewriter racks, and other deep-ground, accurate-form, or broaching applications.

"CDCF can remove enormous amounts of stock and we're looking at broaching a slot 1.5" deep, 0.5" wide on the ID of jet-engine rotor discs (to hold the blades). They normally rough it out and then finish broach it, but with CDCF you can eliminate all the broach toolign and do it in a single form grind. There are also many applications in broach-tooling manufacture--for those re-entrant features that you cannot grind.

"Another advantage of CDCF is that it is continuously sharpening the wheel so that you can grind any length of part, whereas in creep feed, you can go only a couple of inches into a deep grind before you get some thermal damage. With CDCF, you can grind until the wheel runs out. This is ideal for long broach tooling or a setup of a number of parts to grind in one pass, such as rocker-arm cam tops." Creeping into creep feed

So why isn't everyone rushing to CDCF? "We're still in the early days of CDCF. The conventional creep-feed market is now increasing here in the US at a rate of 20 percent annually, according to an independent survey of wheel use by Norton. That's the foot in the door. Now it's a matter of saying, 'Now that you've accepted that creep feed works, take a look at continuous dress.'

"Of course, there are people out there who have tried creep feed 10 years ago--with the wrong wheels and wrong technology--and decided it was not the right way to go. Yet now, they are seeing everyone doing it in Europe and deciding it's time to take another look.

"But continuous dressing is even more difficult to accept. It's an accelerating technology. Grinding was stuck in the doldrums for about 100 years, and all of a sudden creep feed came along--something really new. Now, just when people are making commitments to a new creep-feed grinder, they're thrown into a new quandary. They're saying, 'Now what do I do? I've just justified a $400,000 machine, and someone is saying there's another machine that's 10 to 20 times better.'"

So it is difficult, but as Salmon explains, "You don't have to go for the $3-million highly automated CDCF systems that the aerospace people are using. I've seen a couple of small job shops that do subcontract work now using CDCF machines and getting enormous productivity, and they are quite pleased with them. They can manually load the machine because they are competing with other job shops using much older technology. They don't need robot loading to get big productivity benefits." Resolving areas of controversy

There had been some talk that CDCF sacrifices surface finish to get productivity. Not so, says Salmon. "The CDCF process can be tuned to get the surface finish you need. The lowest specific energy--the lowest amount of heat or energy used to remove a unit volume of stock--yields the worst surface finish. This is not one that looks like a dog clawed on it, but one that might well be acceptable for some uses.

"A compromise must be struck between speed and finish. When we're running at 2 micrometers dresser infeed rate and a 0.8 ratio of dresser speed to wheel speed, you get a relatively rough finish. To improve surface finish, we hold back on dresser infeed, and because we know that we will start to get burnishing action, we back off also on the stock-removal rate. We can achieve excellent surface finishes and still get a quantum leap on stock-removal rate."

What about the arguments that CDCF chews up grinding wheels at a horrendous rate? "When you look at grinding wheels per parts machined, you will use maybe 1.5 times as much wheel with continuous dressing compared to intermittent dressing. It's not 10 times as some have claimed. This is hardly an enormous increase in wheel usage.

"Because we're machining 10 to 20 times faster with continuous dress, you obviously must change wheels much more often. If this is done manually, it should take theoretically about 15 to 20 min, but in practice, it's typically in hour or two.

"So automatic wheel change is very advantageous. The Blohm cell can change a wheel in 2.25 min (they conservatively claim 3.5 min), but this is a very complex and expensive machine tool. For someone to stagger into this complex machine and try to manually change the three wheels that are mounted on a single spindle is difficult. This leaves you with a high risk of ruining the machine. So automatic wheel change has two main benefits: speed improvement and fixture protection. Yet, it is only essential for the highly automated cells." CBN?

What about the conflicts between proponents of CBN (cubic boron nitride or GE's Borazon) wheels and your reliance on soft wheels? "The Borazon wheel is for different materials and different applications. Borazon will eventually wear and burn the part when the wheel dulls, and it still doesn't really have the stock-removal rate and the finish capabilities of CDCF on superalloy materials.

"CBN is suited to extremely high-volume applications with stock-removal rates of reciprocating-type grinding processes. High volume is necessary because the wheels are quite costly. Once you've dressed and trued these wheels--taken a lot of time to set up the process-you want the benefit of a very long run. You don't want to have to match this type wheel to a number of different part runs and configurations. And the problem with CBN is not that the wheel wears out. Somebody either drops it, or it gets abused and enormous costs result. Wheel wars

"CDCF wheel materials are primarily fused alumina and silicon carbide. The vitrified aluminium-oxide wheels are used on the steels, tool steels, and superalloys; whereas the silicon carbide works exceptionally well on titanium, aluminum if you really want it, and the more chemically reactive-type materials."

The wheel must be the very high porosity creep-feed design, of course. Even though you can buy creep-feed wheels made in this country, Salmon feels that they are not good wheels. "It is difficult to get the high-porosity structure homogeneously throughout the width of the grinding wheel. You can get spots of close-packed grit and uneven porosity. It's a problem in how the wheel is formed or pressed in this country.

"Two European wheel manufacturers have the creep-feed grinding-wheel market in the palm of their hands--Elbe (not the same as Elb-Schliff, the grinding-machine company) in Germany and Tyrolit, in Austria. They clearly have the best wheels. Now please understand that in creep-feed grinding, the wheel is more critical than it is in CDCF, which is more forgiving.

"The 'wheel war' is far from over. Some of the research work I've done shows a normal noninduced-porosity grinding wheel doing just as well, with the same forces and stock-removal rate, as a high induced-porosity wheel. This is because continuous dressing is negating the wheel hardness. The bonus we get with the very porous wheel is that in the limit of the process, the very fragile bonding breaks down ahead of the continuous dressing, and this protects the part from thermal damage. The wheel shatters before the part can burn. A harder wheel can load up and burn parts."

For example, in experiments with a relatively porous WA 60 80 FP2V wheel (3 mm DOC) on C 1023 material, Salmon found that it could be creep-feed ground at speeds up to 60 mm/min before the part burned. With continuous dress, top speed leaped 25-fold to 1060 mm/min and the failure mode was wheel breakdown, not part burning. With a harder wheel, a WA 60 KV (again 3 mm DOC), the creep-feed limit was 10 mm/min (burning) which leaped a hundredfold to 1060 mm/min, but the failure mode was burning. Thus, if you can live with only a 25-fold speed improvement, you can use the softer wheel to eliminate all possible thermal damage.

What other risks are there in running a CDCF process? "It's much easier to screw up parts--through misuse or ignorance in using the machine--with creep feed than with continuous dress. Besides thermal damage, with creep feed you risk not getting the stock-removal rates that you should expect from the process, because you're not doing it right. With CDCF, you really can't make a mistake once you've set it up--the process is in control all of the time!

"Whem compared with milling, CDCF offers not only much higher stock-removal rates, but a much better quality finish. A lot of companies that make pump-type rotors for pneumatic or vacuum pumps will mill a slot down the rotor and get a lot of scoring marks that later wear on the vane that fits into that rotor. This makes for a poor quality pump. But with CDCF grinding, you can get a much higher qualify finish with no sacrifice in productiviey, and the product lasts much longer. What's wrong with the grinding industry?

To Dr Salmon, too much of the US grinding industry, or the whole machine-tool industry for that matter, is still turning out machines that look like they came from the Dark Ages. "The sad part about the US grinding industry today is that even companies like Cincinnati Milacron still want to sell machines right out of a catalog. Yet both CDCF and creep-feed machines are very specialized pieces of equipment.

"If you look at the Blohm or Elb-Schliff catalogs, you'll see that no longer do they sell standard machines. They ask, 'What do you want? What are your products? What horsepower do you need? Fine, we can build you a machine with exactly that size spindle and this assortment of table sizes, etc.' They have explosive views of their machines that enable you to modularly design the machine yourself to your specific needs.

"Nobody looks in a catalog today and says, 'Oh, I want to buy a Model 2' or whatever. Yet, that's what the industry over ehre seems to still want to do--churn out the same old grinders or milling machines year after year. Until they see someone else develop a market for something, they're not wiling to stick their necks out. And then it's too late." The jet-wheel alternative

Obviously, Dr Salmon is not too impressed with what he's seen over here for the last five years. And many in the grinding industry, in turn, are not yet too impressed with his work. The standard conservative reaction is, "Who needs continuous dress? I'm doing fine without it." For some, this is entirely true, and for others it's a statement of ignorance.

Here's one argument for looking before you leap. Frank Ridgeway is president of Sheffield Div, Bendix Corp, and he sees no great need for continuous-dress technology and sophistication. Their jet-wheel-cleaning option continuously cleans the wheel with a high-pressure water jet, loading ti with coolant and extending the grinding time between dressings. It works with creep-feed as well as conventional grinding techniques.

As he explains, "This is a patented technique, used mostly as an accessory with crush-wheel grinding machines for many years. Unfortunately, it has been treated primarily as an afterthought.

"It was developed over a decade ago. A very high-pressure (5000 psi to 8000 psi) jet of coolant impinges on the wheel's periphery, blasting loose any swarf embedded in the wheel. The jet is 0.010" to 0.012" dia for safety in case the hose breaks.

"With the jet cleaner, you can use very hard grades of grinding wheels that maintain a more accurate profile, provide even wear rates, and hold tolerances of 35 to 40 millionths of an inch. In some applications, the wheel can run all day long without redressing. Workpiece material can range from aluminum to Waspaloy. It is presently being used in creep-feed thread grinding on our Model 109 grinding machines.

"This approach is so simple that we are now raising the flag on it, compared to the sophistication required for constant wheel dressing. I don't know how or where this constant wheel dressing got so much inertia lately. It's all baloney! Continuous wheel dressing is simply not needed to preserve an accurate profile on a grinding wheel!

"We're one of the biggest promoters of creep-feed grinding because it is faster, no question about it, simply by virtue of the amount of arc embedded in the workpiece and doing the cutting. When you can bury 5 degrees of wheel arc in the work instead of 1/2 degree, it's 10 times faster!

"The problem is that the profile is not going to be the same when it comes out of the work as when it went in. So you need to add a jet wheel cleaner or continuous-dress function. Either accomplishes the same thing, and the jet wheel is one fourth the money and requires nowhere near the sophistication.

"The jet's high pressure loads the wheel with coolant; it actually is absorbed into the wheel pores to depths of 1/4". This occurs about 120 degrees from the cutting zone and carries the coolant right into the grinding zone. It is not spun out. We can plunge about 1/4" into a part in about 1.5 min, take the part out, and hand it to you, and it's room temperature.

"I'm not really familiar with Dr Salmon and his work. We've been promoting creep-feed principles for 20 years on cylindrical parts. We shipped four machines to Saginaw Steering Gear that are doing a 4"-wide cylindrical part, complete form, holding four different diameters at the same time to [plus-or-minus] 6 sigma part distribution at 50 percent of the part-print tolerance. This means overcoming wheel-wear problems to hold accuracy of profile. We dress about every 8 or 10 parts, taking off a tenth or two just to be sure. There is no special sophistication required to do this!" The doctor rebuts

Here is Dr Salmon's reaction to the comments of Sheffield's Ridgeway. "Yes, he probably doesn't know much about continuous dress. Take the Sheffield diamond-roll dressing unit--the rigidity isn't there, the accuracy isn't there, etc. Much more precision is demanded by the continuous-dress process, than the normal diamond-roll used on any process, creep-feed or conventional OD grinding.

"With continuous dress, you now have the process in complete control! You're not just dressing a wheel and hoping for the best when you send it through a workpiece. We're dressing it and conditioning it all the time, throughout the cut. The level of sophistication can be mind-boggling to someone who still hasn't accepted creep feed."

Adds Elb's Davidson, "We use the Sheffield jet wheel cleaners on our machines; it's absolutely required for CD--you must have a clean wheel. The thing that CD does that the jet wheel cleaner cannot is maintain the wheel profile that is constantly breaking down. To compare the two is completely in error!"

So you can decide for yourself what you want to do about continuous-dress creep feed. For more information on CDCF machines from Elb-GERH, circle E20, and from Hermes/Blohm, circle E21. For info on the Sheffield jet wheel cleaner, circle E22. And for abrasive-machining information from Dr Salmon's organization, Advanced Manufacturing Science and Technology, circle E23.