Reference Straight Edge's and Machine Calibration

[Paul Canaris]

What do you use for reference straight edges for calibration of your equipment?

I am aware of Starrett and similar offerings, but wonder if anyone has had good success with less costly straight edges or found a source that offers similar accuracy as a better price.

Also, if you did break down and drop the big bucks Starrett wants, what did you get?


In my particular case, I own a long bed jointer…16” width by 118” length and a 10’ sliding table saw that will need periodic calibration....

Any suggestions are welcome..

[Dennis Peacock]

Just go to your local Machine Shop Supply and get you a couple of nice Machinists Straight Edges. They are perfectly close enough for working with wood. The price is cheaper and the tooling is better than many of our machines ever will be. Remember, it's how well the cuts are on the machine that tells you how much setup you really need to do.

[Hoa Dinh]

Get one (or more) of these from Lee Valley (http://www.leevalley.com/wood/page.aspx?c=2&p=50074&cat=1,240,45313).

http://www.leevalley.com/images/item/Woodworking/PowerTool/05n6301v1.jpg

http://www.leevalley.com/images/item/Woodworking/PowerTool/05n6301a.jpg

[Paul Canaris]

Hoa Dinh.............50" is the longest Lee carries, too short for my Jointer. I really need something like 72".

Dennis Peacock.....I'll call a machinist supply shop in Austin and see what they have. I am attempting to beat the going rate of $400 to 450 for a Starrett 72" non-beveled, non-graduated (accuracy quoted as 1.2 thou over entire length)......I am currently getting inaccuracy in jointing....several good quality SS rulers edges I am using all indicate error in the same direction, but I won't use those for calibration...

Also looked at Pinske as an alternative, if I understand their copy, a 69" would be about 2.5 thou over length....and a third of the price...

I admit to being a bit anal about accuracy...

[Tim Sproul]

......I am currently getting inaccuracy in jointing....

A good analysis of the the inaccuracy often tells you what is wrong with the jointer set up.

Folks have been setting up machinery to do accurate woodwork for far longer than highly precise/accurate straight edges have been made affordable.

[John Scarpa]

Paul,

You have a real challenge with a jointer bed that long. It's going to be expensive to find a 6 foot reference straight edge. I have 48" Starrett but would have gone with a Lee Valley were it available at the time. You may be able to get a 72" piece of stock machined for you. I imagine that if you have seveal thousands of an inch drop over 72" you are still in good shape. IMHO I would go with the longest Lee Valley item and only worry about it if you experienced problems with your finished stock.

Good luck!

John

[Don Baer]

I have a large (72") machinest level. It has a very flat steel base and is dead on flat. Look around for one of these you can lay it on the outfeed table and set the blades till they just kiss the edge. Then crank up your infeed to match and your tuned. Back off the infeed to your desired height. I use a feeler gage at this point.
In a past life I used to use this for aligning motors to machine where the alighnment had to be perfect.

[Jamie Buxton]

You might consider straight edges intended for flooring installers. For instance, http://www.tools4flooring.com/carder-139-75-in-professional-straight-edge-p-679.html is a straightedge which is 75" long, with a quoted tolerance of only .003". And it costs $44! If you don't believe it will survive UPS to you, look elsewhere on the same site for a spring-steel straightedge. They just roll up, so they're nearly impossible to damage. Crain's got a six-footer with .003" accuracy for $75.

[Mike Cutler]

Paul. I use a 6' Starrett machine rule, affectionately referred to as" Nasty Stick" at work.
I was extremely lucky, mine was free. It was headed to the dumpster after it had lost it's NIST paperwork, and the cert's had lapsed,and could not be recovered.
I use this for everything. machine setup, layouts, all of my tape rules are measured against it. It is the 'standard" in my shop. I probably reach for this tool more than any other single object in the shop, with the exception of a pencil. It is capable of some very accurate layout work.
Using this tool, it is entirely concievable that you can hold the tolerances to <.125mm easily across any project( pencil line width), the tolerance could be even tighter if you use a marking knife.
It is not just a straight edge for aligning machines.
I have no idea how much one would cost that was "commercial grade", but you might want to give Starrett a call, or do a net search. I'm sure that there are straight edges that will suit the specific purpose of aligning jointer tables, but it may be worth considering everything else that this tool can perform.
Additionally, the ends have a "square acccuracy", and the edges have an accuracy spec on parallel, real nice for accurate transfer of layout lines, and incremental spacing.
They used to make "giant drafting tee's". This being the computer age, I don't even know if they make "old time drafting tools" any longer. These used to be pretty darn flat and accurate along the length. Even though they were made of plastic, they were still kind of expensive also.
Good luck.

[Greg Mann]

A good analysis of the the inaccuracy often tells you what is wrong with the jointer set up.

Folks have been setting up machinery to do accurate woodwork for far longer than highly precise/accurate straight edges have been made affordable.

I'm with Tim on this one. Even if you invest in the straightedge and find a few thou out-of-flat what do you do with it? Do you really want to start pulling tables off to be ground? That would make the investment in the straightedge seem pretty small and that's assuming whoever does the grinding is competent. I would do everything I could to tune the machine as is.

Greg

[Paul Canaris]

Thanks for all the advice guys, got a bit of thinking to do now…

[Mark Carlson]

Tauton Press has a video on setting up a jointer. I havn't watched it in awhile but it does have a section on creating your own straight edge to align a jointer. It involved using a number of long boards and screws. At the end you ended up with a board with screw heads at the same height. I just tried to find the video and remembered I had lent it to someone. Maybe someone else can remember exactly how this was done.

I just found the video online. Its called "Tuning up a Jointer" with John White.

[Paul Canaris]

Looked like a good solution and I appreciate the effort it took to find the product....Got a hold of the manufacturer of that Carder straight edge. It's +/- .003...or .006"....won’t work<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>

It's annoying when vendors post in-accurate information....<o:p></o:p>

[Jamie Buxton]

Looked like a good solution and I appreciate the effort it took to find the product....Got a hold of the manufacturer of that Carder straight edge. It's +/- .003...or .006"....won’t work<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>

It's annoying when vendors post in-accurate information....<o:p></o:p>

What accuracy you believe you need?

[Stu Ablett in Tokyo Japan]

Tauton Press has a video on setting up a jointer. I havn't watched it in awhile but it does have a section on creating your own straight edge to align a jointer. It involved using a number of long boards and screws. At the end you ended up with a board with screw heads at the same height. I just tried to find the video and remembered I had lent it to someone. Maybe someone else can remember exactly how this was done.

I just found the video online. Its called "Tuning up a Jointer" with John White.

Or buy the book, I did, it is called

Care and Repair of Shop Machines : A Complete Guide to Setup, Troubleshooting, and Maintenance (http://www.amazon.com/exec/obidos/tg/detail/-/156158424X/002-8800773-4082412?v=glance)
by John White (http://www.amazon.com/exec/obidos/search-handle-url/index=books&rank=relevancerank&field-author-exact=John%20White/002-8800773-4082412)

Worth every penny.

Cheers!

[Paul Canaris]

I would personally be comfortable with .002 to .003 over a 72" length....I would rather know that my setup is dead-on based on accurate measurement tooling, that way any deviations encountered later in usage are clearly a result of technique or materials...it just lessens the variables I have to look at...and as I said, I admit to being anal about this kind of thing.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>

May be that I just need to plunk down the cash and get a Starrett.... <o:p></o:p>

[Tim Sproul]

Thanks for all the advice guys, got a bit of thinking to do now…




Mind letting us know what is the inaccuracy?

[Jamie Buxton]

I would personally be comfortable with .002 to .003 over a 72" length....I would rather know that my setup is dead-on based on accurate measurement tooling, that way any deviations encountered later in usage are clearly a result of technique or materials...it just lessens the variables I have to look at...and as I said, I admit to being anal about this kind of thing.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>

May be that I just need to plunk down the cash and get a Starrett.... <o:p></o:p>

So in your view, .003" would be good enough, but +/-.003" is not good enough? Mind walking us through the thought process which leads to that conclusion?

[Greg Mann]

Looked like a good solution and I appreciate the effort it took to find the product....Got a hold of the manufacturer of that Carder straight edge. It's +/- .003...or .006"....won’t workffice:office" /><O:p></O:p>

It's annoying when vendors post in-accurate information....<O:p></O:p>



+/- .003 is not .006

It is within .003 of perfection in one direction or the other. Big difference. I still wonder what you are going to do with the info once you have it. FWIW, we have granite surface plates re-surfaced from .002 to .003 OOF to within a few tenths over a 2 by 3 foot surface area for $96, plus about another $100 travel time. You could probably have the tables lapped professionally for less than it would cost to disassemble the machine, especially if you pay cash. ;)

If you were to do this, you could save the cost of buying the straightedge in the first place.

Greg

[Sam Blasco]

I've got a Starrett 6'. I forget how much it cost, but I'm sure I cried at the time. You could bring the jointer down to my shop and borrow it ;}.. Remember you are just milling wood here, don't get too crazy on dialing it in. A good 4' level can give you nice results, even on long bed jointers, combined with test cuts, will tell you all you need. Use 5' boards for test cuts. Concave edge: Outfeed table too high. Snipe at the end of the cut: Outfeed table too low. Wedges - tables not co-plane. Even with rudimentary gauges and measuring devices, a little bit of trial and error will get you excellent results with regard to setup.

[Paul Canaris]

John thanks for the idea of contacting a machinist.

I am going to take some Hot Rolled rectangular steel I have; (two) 6’ by 4” by ½” sections to a local machine shop. This guy has a milling machine with a 20 foot table travel, so six feet is easy for him. He will mill two flat edges on each piece, to less than a thousandth, over the length for $250. Better than spending what Starrett wants for the same thing and it satisfies my concerns about accuracy. Also, it is long enough to calibrate the jointer and can be used for calibrating my slider as well. Thanks to everyone for the feedback. I appologize if I got anyone upset, but we all do what were comfortable with.

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p> </o:p>

Also thanks to Stu Ablett and Mark Carlson for the book suggestion. I just ordered it from Amazon for $15 including shipping, good deal.

[Jamie Buxton]

John thanks for the idea of contacting a machinist.

I am going to take some Hot Rolled rectangular steel I have; (two) 6’ by 4” by ½” sections to a local machine shop. This guy has a milling machine with a 20 foot table travel, so six feet is easy for him. He will mill two flat edges on each piece, to less than a thousandth, over the length for $250. Better than spending what Starrett wants for the same thing and it satisfies my concerns about accuracy. Also, it is long enough to calibrate the jointer and can be used for calibrating my slider as well. Thanks to everyone for the feedback. I appologize if I got anyone upset, but we all do what were comfortable with.

[/size][/font]

I've been fiddling around with a sagulator program I downloaded a few years ago. It was intended for wood, but I can plug in numbers for steel. If I've tweaked it correctly, it says that if you stand that piece of steel on edge and support it at the ends, the middle will sag about .003" from its own weight.

[Paul Canaris]

Everyone else can ignore this as it is a bit over the top. :D

Jamie, what you describe is not how a straight edge is typically used. If a longer straight edge is used (typically anything over 36") as designed, there are two support points about 1/3 of the way in from each end that allow you to suspend the straight edge so contact occurs with the intended surface, but the full weight does not rest on the straight edge or the intended surface, therefore vanishingly low deflection.
Most of us will use a straight edge by supporting its weight on for example the outfeed table of a jointer with a portion extended over the infeed table. So there will be deflection, but not as much as in your example.
Also, a straight edge that starts out at .003 given the same support conditions and geometry as a straight edge that starts out at .001 will incur a larger amount of total error due to what was already present in a static state.
The actual equations involved in determining the sag are as follows:

Equation for moment of inertia of a rectangular cross-section = I = 1/12*b*h^3

Equation for maximum deflection in a beam supported at each end with the W, or "mg" of the beam calculated as a concentrated force at L/2, where L = length of the beam:

Y_max = WL^3/(48EI)

Using the approximate weight of the beam of 50 lb (22.68kg), the Young's modulus of the steel as 207 GPa, and the beam's cross section is a rectangle with a base of 0.5" (12.7mm) and a height of 4" (101.6mm), I calculated a deflection of 12.577 MICROMETERS (10^-6), or 0.000495 inches.

Now in the case of laying the edge half on and half off the outfeed table, the deflection will more than if the straight edge were suspended, but considerable less than the .000495”. So for all practical purposes it is negligible.

Again I agree that this is woodworking and not metalworking, but I still contend that inaccuracy is additive, better to eliminate the contributions from your equipment (particularly with Jointers that are multiple pass where error compounds with each pass), after that it’s you or the material that is causing the error, makes for a bit easier focus at that point in determining what went wrong.

Again I admit to being a bit anal about all of this. :rolleyes:

[Les Spencer]

How are you planning to check for flatness today and 6 months from now? :confused: Will your straight edge be normalized?;) Are you concerned about stability?:confused: Have you considered using a machinest level to check flatness?

[Don Baer]

How are you planning to check for flatness today and 6 months from now? :confused: Will your straight edge be normalized?;) Are you concerned about stability?:confused: Have you considered using a machinest level to check flatness?

and don't forget effect of changes in tempeture differances in thermo expansion due to differances in thickness and types of material ....;)

Just kidding. Like I said, a good machinst level will due the trick.

[Paul Canaris]

Never used a machinist level, wouldn’t know how.<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>

[Les Spencer]

I'm still wondering how you plan to check the staight edge for flatness?:confused: I doubt that the machine shops mill is flat w/n .001 for 6' unless it is relative new. Now if you said he was going to grind it after milling and heat treating, I'd say you have a chance. If you plan on using it (the mill) to check for flatness all you will be doing is checking how accurate the machine milled the surface not the flatness. Hopfully they will have a surface plate large enough to check it. A machinist level is accurate w/n .0002/ft. :) I worked in a shop rebuilding machines from mills, grinders to jig borers etc and we were taught by an old time Pratt and Whitney Jig Bore builder how to use the level to maintain flatness. His point was the straight edges just couldn't be depended on to always maintain flatness but the level would always be accurate. Although we had our doubts about his techniques, he made a believer out of me and others with the quality we were able to achieve in the finished machine.


There is a definite reason why Starrett or Enco gets big $$$$ for 6' straight edges. There is alot of work and the use of very precision machines to make a SE. If you forego the heat treat process, you will be creating an eventual boat anchor, although it will be a long one.:D It will not stay flat without heat treating. IMO you are out chasing butter-flies with an elephant gun.:rolleyes: Your finished pieces off the jointer should be all you need to determine how your jointer is. If the table is out what will you do? The earlier suggestion on lapping it flat is not going to work unless you are willing to pay mega $$$. I think he was talking about granite surface plates which lap many times easier than cast iron or steel. All of our cast iron surface plates required hand scrapping. You will need to have the jointer table(s) either hand scrapped or ground to correct any error.:eek: Again, big $$$.

Don't be anal. Be Cool. :cool:

[Steven Wilson]

I use a 4' Starrett straight edge as a reference. I keep it in a case, in the house, when not in use. I also have a 7' Starrett straight edge that use to be 8' long but was damaged. A friend of my dad, who use to own a couple of tool & die shops in Detroit, cut off the broken bit, treated it, machined it flat and sent it to me. It's very straight (at least any continious 4' section) and is just the ticket for long jointer tables and sliders, but way overkill; hey it was free. One thing I use often are a couple of long jointed boards that I checked against my Starrett (a little lapping with sandpaper works wonders). The boards are not only good for a quick check but are usefull as winding sticks.