Antique, unsafe shaper knives

Safety and operational issues regarding shaper heads. April 4, 2001

I have a shaper head with two through-bolts in it. When bolted together, it has angled grooves running the full width of the head to hold the knife steel in. The steel I have matches the angle grooves top and bottom and is 2 1/4 x 25" long (slicksteel?). I realize I need to cut the length down, but I also need info on grinding the profile with a bench top grinder (mounting and balancing, etc.).

Forum Responses
Forget about using that head! It's an antique and a dangerous thing to "learn" on. There are reasons that the newer heads were threaded, and still newer heads are corrugated. If money is tight, think how tight it will be without your fingers. You cannot learn all you would need to know about safely using that type of set up from reading these posts. Please think about using a more modern, safe and quick system.

When these knives were used in the sixties, mill workers wore lead aprons to protect their manhood. I've never seen one unscathed or without a story.

This tooling is not for the inexperienced, unless your idea of a good time is a friendly game of Russian Roulette.

All joking aside, the knife stock you have is called beveled edge shaper steel. It is probably high-speed steel with the edges precisely machined to fit the clamp collars you also describe.

This tooling can be very useful, especially for pattern work, due to the ability to infinitely adjust the projection of the knife, small cutterhead diameters, and the availability of clamp collars with an integral ball bearing for following a template. However, it can be very dangerous or lethal, as the knives can slip out at speed, particularly in an unbalanced situation.

Years ago, someone introduced a lockedge variation. This consists of the cutter stock with a series of grooves ground in one edge to engage a worm screw in the clamping collar, similar to a rack and pinion assembly. This eliminates the tendency of the knives to slip and allows for precise adjustment with an allen wrench. With this variant, I can think of no good reason to use the plain edge version you have.

To summarize, don't attempt to use the tooling you have. Corrugated back knife steel and small molder heads for your shaper would be safer. If you must use this style tooling, use the lock edge variety. Charles Schmidt is a good source for this. They can also grind a lockedge on the bar stock you now have. By the way, never use lockedge steel in smooth groove collars!

Before you throw away your smooth groove collars, assuming you have access to a profile grinder, open up the bore to the grinding arbor diameter and use them for grinding fixtures for your lockedge steel.

With any loose knife tooling and cutterhead, it is vital that all safety issues be fully understood and followed--max knife projection, balance, depth of cut and max RPM are a few. Your shaper should also be of a mass and quality to safely accommodate this type of tooling.

As for hand grinding, the “Knife Grinding and Woodworking Manual” by Monnett, available from Charles Schmidt, is a good resource. The makers of the XL or Stetson Ross molder published an excellent booklet on the subject. I think Tyler Machinery still supports those machines.

A corrugated tool is by far the best choice. That is the only type of tool I use for custom profiling. For standards, the use of insert tooling is also a good idea.

Dave Rankin, forum technical adivsor

Although I agree that corrugated tooling is the best choice for most general work on a shaper, I must take exception to the above statement that corrugated tooling is the only type that should be used for custom ground knives on a shaper.

Bevel edge tooling and clamp collars have a number of useful, if not indispensable, qualities for custom millwork and furniture, and with the advent of the lock edge variety, the safety concern associated with the original smooth ground stock was addressed. With a typically smaller diameter, an integral ball bearing, and the ability to fine adjust the knife projection, they are superior to corrugated heads for some operations, such as pattern shaping. Even on operations where the fence can be used, such as panel raising and curved moldings, the ability to adjust the knife to the bearing provides additional support at the point of contact for the work piece, and thus better control.

Regarding safety, any loose knife tooling, including corrugated steel, can be especially dangerous if improperly used.

For years, lock edge tooling was the stand alone model of safety for shaper work. The ease with which one could use an allen wrench to finely adjust the projection was an added feature to the fact that it was almost inconceivable that a knife could be released at speed. Corrugated stock didn't offer anything in small diameters, nor any infinite adjustment, and still doesn't. One is strictly limited to the pitch of the corrugations. Hence the above rub collar example.

Corrugated knife stock was "fathered" by Weinig in the seventies and the first few US Weinig Dealers had a tough time converting customers to it.

"Slick steel" was cheaper by far, quicker and easier to grind, and the cost of "cutterblocks" was 4 to 1 in favor of thin steel. The lock edge tooling filled the void between the two choices and is still the popular choice of many for safe, economical, clean, and quick shaper runs.

I do agree that lock edge collars allow for smaller OD of tooling. I prefer corrugated tools when the diameter allows. If you are going to use lock collars, be sure to properly secure the tools and run the RPM at no faster than the recommended speed.

I will not run lock collars. As a safety engineer, there is more opportunity for inexperienced operators to get injured using this type of tool. Many of my students use lock collars regularly for applications where the diameter of the tool is critical.

By using a machine that uses a collet instead of a spindle, a router bit designed tool can be used. This tool can be made with corrugations or with inserts. Granted, this may require a special "spindle" for the shaper, but the increase in safety is well worth the minimal expense.

Dave Rankin, forum technical advisor

It is naïve to suggest going back to a "router designed tool" with a 1/4" or 1/2" collet from a 1" diameter plus shaper with weight and swing capabilities 5-10 times the colletized machine.

Router work is router work and shaper work is shaper work. Where the hobbyist gets into the most trouble is when tooling companies suggest they run heavy cuts in raised panel doors, for example, with colletized tooling.

Besides, name a shaper that will accept colletized tooling. I can think of only one, it's 1500 lbs., and was intended to work this way in non-ferrous, not wood.

Remember, engineers agree that a 1" spindle shaper running tooling with a "body" (not including the wings) of 2" diameter effectively becomes a 2" diameter shafted machine, for purposes of strength and stability statements about the spindle.

Tooling that has a built-in spindle can be made in a very small size. This will allow this type of tool to be used in many applications.

I did not mean to mislead anyone into thinking that a hand router should be used.

I have retrofitted several shapers with standard collet units, 3/4" and 1" size. In fact, several large double end tenoner companies use collets on their coping stations. Great Lakes Carbide Tooling has made many tools for this purpose.

A "router bit" designed tool is an industrial grade tool. Without a doubt, the correct tool must be used, and a qualified engineer or technician should be consulted.

I am still a strong believer in avoiding the use of shaper collars if at all possible. This type of tool has an increased risk of failure due to lack of understanding by many operators. Many people use them without any problem, but a little lack of understanding can cause severe injury or worse.

As a safety consultant, I make it a policy to avoid the use of collars, if possible. The "router" style of tool is a valid option for many applications.

Dave Rankin, forum technical advisor

The only reason I can think of for using a smaller diameter tool for shaper applications is to get more knife projection. I have run the split collar lock edge tools and I would question the amount of knife including height and length projection which can be safely run at 6,000 RPM. What exactly are the limitations? I understand the corrugated type have limitations. The tooling for these are preset in thickness and length projection. With split collar type knife projection, limitations can be easily exceeded with the ability to cut the stock to projection length. The weight and swing capabilities of a shaper are great compared to a router. The abilities for the tooling to hold up to the shaper's abilities are in question. What are the limitations of split collar lock edge tools? Factors to consider: projection of knife, profile requirements, stock removal, spindle RPM. The ability to be able to run the split collar tools the wrong way can be fatal to an inexperienced operator. If the opportunity to exceed limitations of tooling are obtainable to the industry, you can bet it's going to happen.

Were there an answer, this debate would be unnecessary. Unfortunately, the cost of such tooling is prohibitive for the custom millwork, reproduction, or furniture shop, not to mention the serious hobbyist. With loose knife tooling, one can economically and quickly produce a wide range of custom or one-of-a kind profiles, not to mention the fact that knives can be ground using a small bench grinder, if necessary.

That leaves us with corrugated and bevel edge (lockedge) steel. Clamp collars are no more complicated to set up than corrugated heads and properly set up, they are just as safe. In fact, there are more issues to be considered for corrugated heads than clamp collars, such as blind wedge, 60/90 degree corrugations, proper tensioning and balance of gibs, empty slots, etc. One could only argue that because corrugated heads are more widely used in this day, they are assumed to be more widely understood, a point I will concede, but a small one nonetheless.

The bottom line is that practical considerations should rule. For an operation that uses both molders and shapers, corrugated tooling would minimize duplication and would be a primary choice, with the exception of some specialized operations noted in my earlier posts. For “stock” profiles there is a wider range of off-the-shelf knives available and again, corrugated would be the best choice. For the custom or reproduction shop, needing a versatile tool, particularly for pattern shaping, the lock edge tooling would be the best choice, and last but not least, personal preference may rule.

Any tooling system can be dangerous if used improperly, and there never seems to be a shortage of people willing to do stupid things and injure themselves in the process. If the safety of a given tool is going to be judged on the possibility that someone may misunderstand it or use it inappropriately, then I think the rest of us would be left to stand around and stare at bare spindles.

I use lock edge knives all the time where I work. In over 12 years, I have never seen anybody throw a lock edge knife (although, I imagine it's done).

We find it easiest to have our knives professionally ground and balanced by our sharpening service. The price is reasonable and the knives match perfectly. They can grind off a sample piece from a sketch or a DXF file that you e-mail to them. They use thick steel, not the thin, cheap stuff.

Our time is better spent making product, rather than butching out tools by hand on a dry pedestal grinder. Don't waste your time and safety grinding to a line by hand. A great knife, ground to within a thousandth or two, is easier and safer to get ground.

I was confused about clamp collars and lock edge shaper heads (never having seen one), until today. I went to look at a shaper for sale (big, cast iron and driven by a 3" wide belt) and I can only describe the way of holding the knife as clamped by the edges. There were definitely no grooves on the sides of the knives, so I'll guess this is a rather old machine. The spindle is 1-1/4" by at least 4".

The shaper you describe might be babbit or a conversion, circa 1920. The spindle elevator is probably done with mitered gears and the assembly has a lock handle on the heavy v-ways.

The motor is stationary and the flat belt self-adjusts by riding up or down the pulley face when raising or lowering the spindle. Danger, don't convert to v-belt.

Parts (fence, mitre gauge, guard, etc.) can be made or borrowed from other machines.

Following are some guidelines I have used regarding beveled edge shaper collars. I recommend you verify this and any other info you receive on a forum such as this. I consider Charles Schmidt Co the best source for this type of tooling and information on its safe use.

1. Balance the knives. Unbalanced tools can cause finish problems, bearing wear and, in the case of the slick steel you have, can cause a knife to be thrown. A lot of the old timers used to grind 1 profile knife and fill the other slot with a "balance knife", usually an old knife with no profile that weighed the same. As the projection of the 2 knives cannot be the same in this situation, this guarantees an unbalanced assembly, even though the knives are the same weight. Avoid this kind of "logic".

2. Choose the correct thickness of steel for the job. Beveled edge steel is available in 1/4”, 5/16”, and 3/8”. Maximum safe projection of the knife beyond the collar is 3-4 times the thickness of the stock (verify with supplier). There is nothing wrong with overkill in this area, but remember the thicker the steel, the harder and more time-consuming the grinding.

3. Inspect the grooves on the collars for dirt or damage, to ensure a good seat for the knives. Discard collars with any sign of damage or excessive wear.

4. Don’t mix steel of different manufactures in the same pair of knives. Some say not to mix different bars from the same manufacture, but I would not go that far.

5. Cut the stock long enough to fill the slots at least to the center of the spindle/cutterhead. I consider this a bare minimum and usually cut to fill about 80% of the groove.

6. Make sure corrugations are engaged (if using lockedge) and knives are properly seated in the groove before tightening the spindle nut.

7. Do not project the knife out the open side of the slot (the side the screw threads into) on lock edge collars.

8. Do not use lock edge knives in smooth collars. Note: I do not recommend the slick collars and steel.

9. If you use the 2 cap screws, run them down with the assembly on the spindle for alignment and carefully tighten equally. Personally, I don’t use them unless I wish to keep the knives in place when removing from the spindle for later use. I normally rely on the spindle nut for clamping.

10. Do not over torque spindle nut, as in--no wrench extensions or persuaders--this is not good for the machine and can put unnecessary stress on wide, thin knives as well.

11. Start small and at relatively low RPM and work up as experience is gained. Large cutterheads, small spindles and high spindle speeds on light duty shapers are recipes for disaster.

12. Use feather boards, jigs, etc to guard cutterhead. I generally grind the profile so the heaviest part of the cut is below the work if possible. A power feeder is one of the best safety devices around for a shaper. It will also increase quality and production with a smooth, even feed.

To supplement information on the hand grinding of knives, the following are some additional tips:

1. Transfer the pattern carefully to the knife stock, use the long edge for alignment, and remember there is no independent vertical adjustment of beveled edge steel. I use machinist blue ink and a carbide scratch awl or scribe for this.

2. When grinding, always hold the knife parallel to the wheel, the grind marks on the knife should be perfectly vertical and parallel on the bevel. If you don’t observe this rule you will end up with a mess!

3. I grind to my layout line at a 0 degree clearance angle, as in no bevel. You will grind quicker and cooler this way, as less surface area is presented to the wheel. Once the profile is perfect to the layout line, adjust the tool rest for a bevel of about 45°. Grind bringing the line of the bevel to the top edge of the knife. At this point, you no longer need the layout line, you simply need to bring the bevel to the top edge of the steel, forming a cutting edge and turning a minute burr. If side relief is needed, I stop short of the cutting edge at points where side clearance and back clearance intersect, grind the 5° side relief first and then finish up the back clearance. Rub out the burr and the bevel and you are done.

One final note: don’t let anyone tell you it is impossible, unsafe, inaccurate, or even impractical to hand grind your own profiles. It is a skill like anything else and takes practice and commitment. For the serious amateur or small shop professional it is well worth the effort.

The comments below were added after this Forum discussion was archived as a Knowledge Base article (add your comment).

Comment from contributor W:
Some additonal notes on lock-edge collars vs. corrugated heads:
I run both systems, preferring collars for curved work, due to their ability to adjust to templates, and for deeper profiles. Recommended projections are greater for collar knives, as there is generally more steel, both in thickness and in the length cantilevered back into the groove. I certainly don't recommend it, but I have seen old bevel-edge knives swinging 3 inches plus out of the head. Corrugated steel is better for taller profiles, as none of the spindle length is taken up by the head as in collars.

The question of collar diameter is interesting, and another area where collar set-ups are more versatile. Small diameters have an obvious advantage in following inside curves of templates. Larger diameters will take the load of larger knives. In the past one could order sets with, for example, a six-inch bottom collar and three-inch top collar for large, sloped profiles without having excessive projection, a trick impossible for other systems. I have seen old collars (without the lock facility) down to 1-7/8" diameter with a 1-1/4" bore. It seems logical that the very small sizes would grip better on very small knives, which could often be seen as small as 1/2" high by 1" long.

Lastly, some words about the knife bevels themselves. The standard angle today is 60 degrees, but if you buy old knives you may find some that are 62-1/2 degrees or possibly other bevel angles. Do not use these in modern collars! Make sure the bevel fits very snugly in the groove, and is clean and rust-free. Knives must be of precisely the same height, a factor even more important than proper balance. Do not attempt to grind or alter edge bevels on a bench grinder, or use old knives that have had their edges altered.