Old Mill Drivel (III)

I’d like to address some of the pros and cons to using a milling machine in a wood shop. Before I get into that, I’d like to acknowledge that if your wood shop does the usual work these days, like 95% of the shops out there, then you probably spend your day wrangling sheet goods, vacuum-bagging or pressing to veneers to substrate, and running the typical tools of the trade: wide belt sanders, edge-banders, biscuit joiners, Domino joiners, sliding table saw/panel saw, spray gun and so forth. Because projects in su h shops may involve some solid wood, you will probably see a smaller jointer and planer as well, and maybe a few hand tools too. If that is the realm you operate in, and you never fabricate custom hardware, say, and do next to no joinery, then it seems to me that a milling machine is completely unnecessary and would serve little purpose beyond decoration.

If on the other hand you are one of those oddballs working predominantly or entirely with solid wood and enjoy joinery and often make use of routers and router tables, and/or find the right hardware cannot be obtained and must be custom made, and/or tackle machinery rebuilds or customizations from time to time, working in materials other than wood, then a serious look should be taken at a milling machine. In that realm it greatly increases both the scope and accuracy of what you can build.

If woodworking is only part of what you do and you find yourself doing a wider variety of tasks, maybe fixing bicycles or lawn movers one day, doing various odd jobs, and you have space for a limited number of stationary machines, then it seems to me that a milling machine should be among the priorities, given its tremendous versatility. There’s not much you can’t make with a lathe and a mill.

If you like the results obtained with routers, but detest the high pitched loud noise they make, or find yourself frustrated by having to do router work where you cannot view the cut or are getting sick and tired of making plywood and MDF jigs all the time, only to toss them in a ever-growing pile, probably never to be used again despite your best intentions otherwise, then a milling machine offers a superior and quieter alternative, and you can drastically curtail those particle board purchases.

As for milling machines, there are those designed primarily for metal, and there are those designed primarily for working wood, generally intended for the pattern industry. The woodworking type of mill is comparatively uncommon, and some of these machines are rather enormous (like a Wadkin or Oliver pattern mill), so that limits their potential for use in many shops I suspect.

Some will suggest that for the odd time you require a bit of machining work you can simply head to the local machine shop rather than buy your own milling machine. Seems simple enough in theory, as most towns of modest size will have a machine shop, however I can sound a note of caution with that plan: my experience with machinists over the past few years, with but one exception, has been far from satisfactory.

The other reality is that shops more and more are moving to CNC equipment, and what tends to associate to that are the following:

• skilled traditional machinists are getting hard to find as more and more shops replace manual machines with CNC machines. CNC machining work, via the typical division of labor, is a dumbed-down skillset compared to what is/was done by ‘real’ machinists. Not to say there aren’t passionate and knowledgeable CNC machine operators out there, but they are about as hard to find as the old-style machinist. Shops that do CNC work prefer it if you show up with CAD files.
• CNC machines excel at the production of multiples, and that is where profitability is to be found, so a lot of times such shops simply do not want to be bothered by your little job to make one-offs or a small number of items. If you are lucky to come by when they have a period of down time for some reason they may consider your project, but in most cases they won’t want to bother, and if they give you a quote it will likely be of the ‘priced so high you are bound to decline’ variety.
• even if you find a shop with a few manual machines willing to tackle your small project, and someone knowledgeable to run them, then a lot of times I have been disappointed in the quality of the work and the timeliness of the execution. And this is after trying several different shops in my area. If this were just my experience, that would be one thing, but I have heard similar complaints about machine shop work from several others. The one good experience I have had was with an elderly machinist up in New Hampshire.
• finding a machinist who is passionate about their trade, who relishes interesting challenges and won’t let sub-standard work out of the shop – – well, these types of individuals are, I dare say, rarer to find in machine shops than in wood shops.
• And it can be quite expensive!

Personally, as a result of multiple episodes with machine shops where the results obtained were poor, delivered late, and needed, in a lot of cases, a re-do, my inclination has become to bring future machining work in-house insofar as possible. Maybe you will be luckier than I have been if you need some machining work done, and have just the perfect machinist down the street from you, but I would say it will be the exception rather than the rule.

A few objections to using a milling machine for wood might be raised:

1. wood dust is bad for the ways of a milling machine
2. milling machine spindles operate too slowly to suit wood-specific cutters
3. The work envelope of a metal mill is much smaller than a wood pattern mill, and this limits applicability

Looking at (1), the observation can be made that, if cutters are sharp you should be producing chips, not dust. Nevertheless, wood shops are often dusty, and inevitably air-borne dust will settle on the machine. The sliding surfaces of milling machines, or ways, are precision surfaces often with a ‘flaked’ finish. The flaking provides tiny pockets for the way oil to adhere when the mating part slides back and forth. On some mills these surfaces are further protected by accordion like fabric metal or plastic bellows, while on other machines they may be exposed. Wood dust and chips will settle on these surfaces if they are exposed. And then what happens? Do the ways dissolve into a puddle? Does the machine stop working mysteriously? Hardly.

Obviously, wood dust and chips land on an oily surface and then soak up a portion of the way oil. As far as being a damaging material, one only need to compare wood chips to other commonly machined materials to see it is rather benign. We can look at some other machinable materials to compare.

Cast iron detritus, for example, is renowned for adhering to ways and quills, then, via being dragged back and forth against a seal with machine movements, can readily conglomerate into little tiny hard balls of cast iron, which can then work their way past seals and permanently damage machine surfaces.

The dust from carbon fiber is conductive, and can migrate into electronics and short them out. Hence you must vacuum the dust frequently when machining to avoid causing damage to the machine. How do you think it is to breathe carbon fiber dust?

I have found that brass chips have an uncanny ability to find their way into all parts of the machine, getting stuck seemingly with ease under seals and leading to oil leaks.

If a shop runs a surface grinder, then the grinding swarf, which is highly abrasive, can settle on the ways of nearby machines if they are not protected.

It is common for many shops to use coolant when cutting certain materials, and there are two basic types. One is oil-based, which tends to make the machine quite filthy after a while if not cleaned up. The other is water-based, which tends to promote corrosion. Some machinists in fact feel that a rusty machine is an acceptable price to be paid for using water-based coolant. By comparison, cutting wood requires no coolant, and the chips don’t really hurt anything on the machine.

And this is not to mention the other hazards inherent to many metal machine tasks: the chips, if long and stringy, can give you lacerations upon handling, and if short, are akin to red hot little knives spit out at you while you cut. By comparison, wood chips are rather more benign. All you have to do deal with is the fact that they will get on machine surfaces, and absorb way oil. The solution? Wipe the machine down after use. If you won’t be using the machine for a while put a blanket or other cover over it.

Now this – wiping a machine down after use – is going to be an alien practice for a lot of wood-workers. The same can be said, based on observations, of many machinists in this country, judging by how absolutely filthy a lot of machine shops are in general. I’ve been in some really oily filthy pits visiting some shops. Lest you think this is an inaccurate assertion, perhaps some of the commentary in this thread may bolster my point: worst machine shop thread

There is one principle hazard that associates to wood chips and dust on a metal working mill. If you don’t wipe the machine down with regularity, then the dust and chips will form a gummy paste with the way oil, and over time this could lead to the machine ways not receiving proper oiling (which means no maintenance is happening either) and getting dried out, at which point wear on the ways will accelerate leading to a machine going further and further out of spec with each use, and the travels getting sticky to operate. Keep the machine clean and the problem never arises.

A second main objection that might be raised is that milling machines have spindle speeds too low for woodworking. Putting aside modern vertical machine centers and the like, which can have spindle speeds over 10,000 rpm, most older manual machines will top out at 2000 rpm. Bridgeport mills are an exception, as they top out around 4500 rpm. My Zimmermann mill’s top spindle speed on the primary head was 5600 rpm, however it is a pattern mill. I would say that before the mill arrived I too had apprehensions about the spindle speed being too low, but I have found in practice that 4000~6000 rpm is just fine for all but the smallest (i.e., under 6mm/0.25″) router bits. And the smaller bits will work, just not as well as they could, and feed rates need to be dialed down accordingly. I’ve since learned from a conversation with another woodworker (Brian Holcombe) that Whiteside Machine, a famous US maker of woodworking router bits, tests their router bits on a Bridgeport. If it is good enough for them….

Cutters are optimized, in a given material, to move over the surface at a particular rate, termed SFM, or surface feet per minute. This is obtained by a combination of spindle rpm and feed rate. With smaller cutters, if the rpm is insufficient, then the cutter can not be run at the proper rates of feed. In such a case, your material removal rate will greatly slow and you will be inefficiently using but a fraction of the machine’s available horsepower. This will be much more of a factor when you are making multiples of a given part.

So, let’s say you don’t have a Bridgeport or higher-speed pattern mill, but another make of conventional mill with a top speed in the 1900~2400 rpm zone. I would say that, yes, this top speed is a bit on the low side for all but the largest router bits. But, there are several solutions to this challenge – machinists face the same issue themselves, as certain materials either cut better at higher speeds, or, can be cut at higher speeds and rates of feed, which means operations can be completed more quickly and thus more profitably.

Some solutions in terms of wood cutting applications:

• On some Deckel-style mills, there are optional high speed heads, or optional fine boring heads, both of which give a 6300 rpm top speed.
• many people have fabricated aluminum brackets to mount a conventional wood router, trimmer, or engraver head to their machine. Especially if you have an older machine, perhaps with a worn main spindle, and your intention is to use it only for wood, then scrapping the main head and adapting on a router head can be a good solution, albeit with the attendant router motor howl. Machines with a rotatable ram, like a Bridgeport, can have the high speed router mounted to the opposite end of the ram from the primary milling head. My Zimmermann has this double-ended feature too, the secondary spindle giving over 17,000 rpm.
• there are devices called ‘spindle speeders’ which mount to the quill of many milling machines, and which use gearing or belt drives to produce far higher output rpm, taking the 2000 rpm input and giving 12,000~15,000 rpm at the cutter, an increase of 4x to 7x of spindle speed being achievable, depending on the model. A trade-off with spindles speeders is that they take away headroom between cutter and work table.
• if you want a really fast spindle, there is another type of spindle speed increaser which is air-driven rather than gear-driven. With these, depending upon input rpm, speeds up to 90,000 rpm can be obtained.

Clearly, obtaining high spindle speeds is not an insurmountable issue.

One solution some might propose is to use a Variable Frequency Drive, or VFD, to increase motor supply frequency and obtain higher speeds. This can work, however only if the existing spindle bearings, gear train, and tool-holding system is suited to the higher speeds one could obtain. In some cases therefore, adding a VFD to a machine not designed for it may only obtain slight speed gains safely, so it is not as promising a solution as it might appear. Many mills also have multiple motors, and this generally requires multiple VFDs.

It’s important to note that with a milling machine one is not limited to the conventional router cutters available in woodworking. There is a world of tooling beyond the typical ½” and ¼” shank options you see on routers, and in many cases one finds that prices for end mills, even quite large ones, are extremely competitive, and often cheaper than the end mills sold as router tooling. For example, I bought a new magnificent 7″ long ¾” in diameter end mill, 4-flute, for about $50. You can regularly pay $75~$100 for ½” shank carbide spiral router bits by comparison. These now seem a bit of a rip off to me.

Besides conventional solid carbide cutters, there is also a vast array of tools with insert knives. While there is at present a limited selection of insert knife router bits out there on the market, vastly more options present with machine tooling. With larger diameter tooling also comes a lower requirement for spindle speed.

If you regularly work with ¼” shank router tooling or less, then higher spindle speeds will naturally be a desirable situation. With my current mill, I find 5600 rpm fine for pretty much all my tooling, though I haven’t had much call to use my really tiny router bits (1/16″ and smaller) on any recent projects.

Another point: using little router bits – in the ⅛” and under category – with a hand held router or in a router table, presents the chance of breakage due to inconsistencies of position or feed speed. With the same tool in a milling machine, you have perfect control of the piece, both in terms of work holding and in terms of moving the table to effect the cut. Not to say that inattention to feed direction can’t lead to immediate bit breakage, but, well, it is important to be aware of what you are doing in any case.

The last exception that can be raised is that of the comparatively small work envelope which associates to metal working mills. With a metal working machine, unlike almost all woodworking machines, the components need to be very robust to handle loads imposed by the cutter on the material. So, while there are some enormous metal mills with large work envelopes, filling entire rooms and the like, the weight of the machine begins to climb dramatically. A machine with a 1.5mx1m table, say, is going to be a 4+ ton machine. Obviously, such large machines are not going to be in the realm of consideration for most shops.

If you want to mill larger pieces, then it makes sense to consider a table and gantry type of CNC machine. Most of the industry is going this route. These machines, if of decent quality, are quite expensive ($50,000~$150,000 not being unusual), and given that the vast amount of work done involves sheet goods, these machines typically have rather modest ‘z’ axis capacity.

There are a lot of reasons why someone might opt to go for a CNC machine, and I have thought long and hard about going that route myself, albeit along the DIY CNC route. For this post however I have set that consideration aside and only choose to consider old school manual metal milling machines. And yes, the work envelope of most mills means one can’t machine every sort of stick, end to end, that is required in a typical furniture project. But you can still tackle a significant portion of work, as well as hardware and jig fabrication.

There are loads of basic woodworking tasks that are better done on a mill I feel. Let’s say you want to produce a series of holes or slots at a precise spacing. Sure, this can be done on a drill press, but to obtain precise spacing you would need to fabricate some sort of fixture to control part position, so there’s a time and material sink. Or maybe you drill one set of parts, accepting of slight spacing variances – if it is not a critical aspect – and then use transfer punches to a mating part and then drill those. The job is faster and more precisely done on a mill, and no need in most cases to construct any sort of special fixture.

Here’s an example of a woodworking joint – multiple sliding dovetail – that is darn hard to do really cleanly by conventional means, but is not too difficult on a mill:

Another area where I find a milling machine is a huge advantage is the jointing, dimensioning and other cutting of small parts. These present a certain challenge by other methods, namely in terms of safe work holding, but are attended to on a mill with relative ease. A picture from Karl Holtey’s blog shows just one example:

Another area where a milling machine realizes an improvement over conventional milling practices (jointing, planing and so forth) is in the area of working with difficult woods with interlocked grain. Working with curly quartersawn shedua, for example, I could only get as close as 1/16″ (1.5mm) with my Tersa-head planer without risking tear out that would penetrate down to the intended finish surface. Possibly a helical head planer would handle this, or a sanding machine, but I found that putting the piece on the mill, where the cutting was across the grain instead of along it, removed all risk of tear out. It really saved the day for me in some cases.

Anyhow, all of the above is not intended as a sales pitch. I simply wanted to lay out some of the reasons behind my choice to obtain and use a milling machine in my shop. I hope at the very least it is informative and may give the reader pause to consider the various ways in which work can be tackled.

All for now – thanks for tuning in.