Extruded Aluminum Fences, Design vs. Execution

  • Advertise with us
Blog entry by easelman posted 07-07-2009 07:59 PM 3223 reads 0 times favorited 5 comments Add to Favorites Watch

I am a retired machinist/mechanic and have always dabbled in woodworking, first as a hobby and now commercially. Over the years, my pet peeve with small, non-industrial woodworking equipment is that they sadly lack most forms of x,y,and z axis indexing commonly found on metal working equipment, most notably mills. Fine adjustments to table saw and router table fences, blades and bits were usually a seat-of-the-pants feel, no dialing in a few thousanths to get that last bit of accuracy. Yes, there are many gadgets and guages to aid this process, and the advent of products by Incra, Kregg, etc., has helped move woodworking to a higher level of indexing ease and accuracy.
Even though it is a time-consuming process, I get very good results with seat-of-the-pants micro adjustments by using machinist squares and various guages. I also have a heavy duty router table with a very accurate lift, so this solves some of the z axis indexing. The weak link in my shop was an indexable miter fence, so a number of years ago I purchased an Incra Miter 1000SE that I mostly use to cut angles away from 90 degrees. (I have a custom made cross cut fence used for 90 degree cuts.) Out of the box, the 1000SE looked impressive, if not a bit fussy. It has a very good angle indexing system, and that feature was what I needed most. Incra touts their products as highly accurate devices that can make micro fine adjustments. I believe Incra’s design of the 1000SE is capable to deliver such accuracy. Unfortunately, Incra’s execution of their design, the production of their product, lends itself to the very inaccuracy their design tries to overcome.
Out of the box, the 90 degree bracket, (a raw stamping) that mates the angle adjustment plate to the extruded aluminum fence was not square, which tilted the fence rearward 2+ degrees. Not a huge problem for 90 degree cross cuts, but any cut away from 90 was a slight compound angle if the stock was held tight to the fence. The angle bracket is a stamped piece of steel, no further machining by Incra, and out of the box, it’s vertical face sat 92+ degrees to the table saw bed, (a Delta X5). Upon machining the bracket square, I noticed the aluminum fence’s front face was not plumb with it’s rear face. When installed to the now square angle bracket, the fence’s front face tilted back about 1 degree from 90. After close inspection to the front face of the fence, the extruded ridges were found to be off a few thousanths in various places, no big deal there, but the rear face’s lower rail was way off. Rather than cutting it down, I took more stock from the angle bracket and solved the problem.
Just recently, I’ve started a small company to produce and sell products that utilize extruded aluminum pieces, so I had a local company make an extrusion die to my spec’s. What I’ve learned about extruded aluminum has led me back to the Incra fence problem. Most extrusions are fairly accurate, plus or minus a few thousanths from spec’s when the die is new. Of course dies wear, and complex profiles such as the Incra fence extrusion have the possiblity to wear the die unevenly over time. If Incra wants to deliver on their design concept of high-end accuracy, maybe they should include a few extra production steps to machine and thus true the various critical surfaces.
I don’t want to be totally negative about Incra. Once I fixed things, I’ve had great results and I really like the angle adjustment system. My word of caution would be to not assume any manufacturer’s extruded aluminum fence product to have a machined level of accuracy out of the box. Extrusions and raw stampings are not used to build accuracy into expensive metal working or wood working equipment. Accurate machines are built with accurately machined parts, there’s no gettin’ around it.

5 comments so far

View PurpLev's profile


8534 posts in 3612 days

#1 posted 07-07-2009 08:48 PM

amen to that. thanks for the info and the insights. not being a machinist myself but very interested in the field, I found this to be a very interesting read.

I agree that we as consumers should not blindly accept products as accurate without verifying them out of the box first. I believe that every company is subject to certain degrees of inaccuracy from time to time, and over certain products. and as the end consumer that pays for it- we have to make sure we got what we paid for, and not what the company delivered in the box.

with all that said, and I am all for higher accuracy and perfection – woodworking is very different then metal working in the sense that wood is alive and moves, a perfect cut to 0.0001” today, will change vastly tomorrow when the wood absorbs/loses moisture. a perfect tight joint today, might (not might- it most certainly will) not be the same tomorrow. a flush joint WILL raise/drop tomorrow – just the nature of working with wood. although – indeed a square cut should be square.

-- ㊍ When in doubt - There is no doubt - Go the safer route.

View jlsmith5963's profile


297 posts in 3312 days

#2 posted 07-07-2009 11:09 PM

There was a post a few days ago that expressed a similar concern over hp ratings. Having spent a significant amount of my youth working in my father’s tool and die shop I am quite familiar with the issues related to the making of dies and machine parts and their associated tolerances. I didn’t follow (directly) in my fathers footsteps but became an architect (that also builds custom built-ins and other architectural woodworking). Taking in the total range of my experience I have worked, (or supervised work), with tolerances ranging from .0001” to 1/2”. What I have learned is that each material/manufacturing system has it’s own low end tolerance limits. Trying to apply these limits broadly across different materials or manufacturing techniques will only lead to frustration. If I acted like a tool and die maker in my wood shop (or vise versa) I might just end up driving myself crazy. (In other words I agree with PurpLev)

-- criticism: the art of analyzing and evaluating the quality of an artistic work...

View Derek Lyons's profile

Derek Lyons

584 posts in 3532 days

#3 posted 07-08-2009 06:08 AM

“Fine adjustments to table saw and router table fences, blades and bits were usually a seat-of-the-pants feel, no dialing in a few thousanths to get that last bit of accuracy.”

That’s because that level of accuracy is all but ludicrous in wood – as the wood can move more than a few thousands over the course of day, let alone seasonally.

-- Derek, Bremerton WA --

View Don Butler's profile

Don Butler

1092 posts in 3359 days

#4 posted 07-08-2009 03:55 PM

For me, adjustable miter guages are unacceptable. I don’t trust the indexing and I don’t like adjusting the guage.
I need only a few angles, notably 45º, 60º, 22.5º, etc. I would never need something in between, such as 25º or 44º, so I have fixed miter sleds hanging about and when I need a miter (not my favorite joint, anyway) I pull one down and slide it through.
Two 45º sleds are hanging directly above the tablesaw, one for the left hand cut and one for the right hand cut. They both can be on the saw at the same time and I can work freely with them with never a fine adjustment at all.
But, that’s just me. I have no criticism concerning the use of guages like the Incra one.


-- No trees were damaged in posting this message, but thousands of electrons were seriously inconvenienced.

View Dan Lyke's profile

Dan Lyke

1519 posts in 4089 days

#5 posted 07-08-2009 10:12 PM

A few years ago I read the Dave Gingery series on building your own machine shop from scratch. The series starts with a simple charcoal fired blast furnace, then works up to a lathe, eventually to a milling machine. It’s not really scratch, you buy the motors and the threaded rod and get old aluminum ladders and such for the casting material, but its a great series of books for thinking differently about the jig creation and tool using process.

One of the lessons of that book was that if you want a flat surface, it has to be machined. And this is why you can buy a $15 3 foot level and a $100 3 foot level.

And, yes, a lot of this level of accuracy is irrelevant for woodworkers, because even clamping pressure can sometimes make that miter fit better, and cutting away wood also releases pressures that change the shape of the wood. But a lot of that is excuses for not getting our setups and jigs better in the first place.

-- Dan Lyke, Petaluma California,

Have your say...

You must be signed in to post the comments.

DISCLAIMER: Any posts on LJ are posted by individuals acting in their own right and do not necessarily reflect the views of LJ. LJ will not be held liable for the actions of any user.

Latest Projects | Latest Blog Entries | Latest Forum Topics