Hi all, I was just wondering what everbody uses to set the height on their table saw
blades?
Thank’s, Dick38
Discussion Forum
Get It All!
UNLIMITED Membership is like taking a master class in woodworking for less than $10 a month.
Start Your Free TrialCategories
Discussion Forum
Digital Plans Library
Member exclusive! – Plans for everyone – from beginners to experts – right at your fingertips.
Highlights
-
Shape Your Skills
when you sign up for our emails
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply. -
Shop Talk Live Podcast -
Our favorite articles and videos -
E-Learning Courses from Fine Woodworking
-
-
Replies
Dick, I assume that you are interested in the height when you are not making a cut all the way through the thickness of the board. In that case, I use a height setting gage built from Shopnotes plans from several years back. They sold a kit with wood and hardware. You can buy several different finished gages of this type fairly reasonably. I wouldn't spend much, as any initial height setting ought to be adjusted through trials on scrap pieces.
Dick,
You don't need another gadget to do this.
To start, take the piece you you are going to be sawing, or a scrap of wood if it is more convenient, and using a combination square or a marking gauge, mark the depth of cut you want with an inch or so long line drawn lengthwise on the side of the stock.
Now place the stock on the saw's table with proper face down. Raise the blade to the approximate height you need and place the marked edge of the stock against the side of the blade, with the depth mark approximately over the center of the saw's arbor.
Rotate the blade back and forth a bit and it will be easy to see when you have a tooth at the highest point of the cut, now just raise the blade until the highest tooth just reaches the line and you are ready to saw.
To prevent the setting from slipping, you should always make your final adjustment to the blade height by raising the blade. If you find that the blade is too high, drop the blade until it is a fraction of an inch below the proper height and then raise the blade to it the proper height.
John W.
Edited 2/14/2005 2:01 pm ET by JohnW
Edited 2/14/2005 2:02 pm ET by JohnW
JohnW's method is probably the one I do most.
But there's other times when I don't measure at all. For example, if the depth of the cut I want is for a rabbet - like for a back panel on a carcase - I'll lay a scrap of the back panel on the saw with a couple of business card shims on top for a bit extra. On top of that I lay a scrap piece of something crosswise so that it extends over the blade. Then I raise the blade until the top tooth at center starts to catch on the crosspiece as you rotate the blade. For the rabbet width, I'm usually going after a round number figure like 1/8, 1/4, 3/8, etc. - thicknesses for which I have plenty of scraps lying around and can do the same method.
If I'm trying to cut some kind of half-lap joint, I eyeball the halfway point then alternately flip a scrap board face to face as I make test cuts until the kerf depths exactly meet in the middle.If you build it - he will come.
This is a very simple height guage I came up with in a hurry one day but works very well.
http://www.superwoodworks.com/Projects/HeightGauge.htm
Garry
WoodWorks by Garry
Hi, Garry...Like your gauge concept - excellent for replication and repetition. Gonna fab one myself and add your suggested improvements. I'll probably
add a "formica" indicating surface so's to keep it from getting dinged up. Thanks for the tip!
"I'll probablyadd a "formica" indicating surface so's to keep it from getting dinged up."
I'll ad that to my list also John
Thanks
GarryWoodWorks by Garry
I put a mark on scrap,then move saw up or down till on mark,seems to work.
I usually use a 4" Starrett combination square or my dial caliper. Quick, easy, accurate.
It ain't rocket science, but a little practice helps.
-Jazzdogg-
Whether you think you can, or you think you can't, you're right.
Dick38,
I use my Starretts 6" combo ..watching the bubble to make sure it's level...always dead on.
Hi dick38 you dont say where you are, if you in the UK TREND do a very good gauge for table saws and router depth settings. It retails for around £7. Its horse shoeshaped with a lockable rule sliding along the closed end towards the open end. All the best and happy woodworking. Regards Steve
I'm one of the combination square crowd. I set the square using a set of plastic calipers, graduated in 64ths, try a test cut on a scrap piece and fine tune it as required.
For regular crosscut/rip work, I usually have my blade set so the gullet just clears a piece of 3/4 inch stock. For plywood, I lower the blade so it barely cuts through and use a slower feed rate.
Edited 2/14/2005 6:30 pm ET by wooden splinter
Hi, Dick...
I just spent a little time in writing a Shop Tip, which I'll reproduce here in all its wordy glory. Like everyone else, I use one method or another to point at the desired depth, but I wanted something a little more accurate. Here's the skinny - will have a picture in a day or two if anyone wants it.
================================
Tablesaw Depth of Cut:
I've invested a good deal of time and effort (and a few bucks!) in tuning my 10" cast-iron contractor saw to the point that it's virtually as reliable as a cabinet model. However, I needed an efficient means by which to accurately change blade height in response to measurement of cut depth. (Creating precision fits by the trial and error process of setting and resetting ad infinitum does work, but it makes more sawdust than progress, and I wanted a better method.)
Here's a simple solution:
I reasoned that if I could accurately indicate rotation of the blade height adjuster, and if the change in height were even remotely close to uniform over its range of travel, then I could adjust height from one position to another within very close tolerance by measuring degrees of crank rotation.
(It's important to be able to adjust blade height to values other than those provided by standard height-setting blocks - the exact postioning of a tenon to suit a pre-cut mortise is one example; machining a half-lap joint another; precise dado depth yet another. I focused only on change in height, as precise actual height indication is made an expensive nightmare by virtue of slight variations in blade diameters, absolute height of table insert, and several other factors - that's why industrial machines cost tens of thousands of dollars! Instead, I habitually measure and set intended blade height as closely as is reasonably possible - usually within less than 0.015"; measure the resultant cut depth with a dial caliper; then change the blade height to achieve desired cut depth. It's the old "try-it-and-fix-it" method, but I can now make needed corrections accurately without multiple trial-and-error steps.)
The shaft on my height adjustment crank turned out to be 5/8" diameter (yours may differ - just measure it), and it showed just over 1/2" of projection between the case and the back of the crank. Hmmm - room enough for a 5/8" setscrew-mounted shaft collar. I found an inexpensive, clear, transparent plastic degree wheel accessory sold by Quint Measuring Systems for its "The Original True Angle" - http://www.compoundmiter.com - with individual degree markings around a 4" diameter circle. It's available at Home Depot. By simply gluing the degree wheel to the shaft collar and mounting the collar close to the case (behind the projecting depth crank), and scribing a reference line in the case on a line perpendicular to the shaft centerline, I've enabled measurement of height-adjuster shaft rotation in increments of not-more-than 1degree.
For my particular saw, I was delighted to discover that rate of depth change is very nearly linear over the range of 0"-3" cutting height, expressed as (very close to) 10 turns per inch of height. You'll want to experiment to learn the values for your own saw - how many turns to raise from 0 to 1"; 1" to 2"; 2" to 3" as compared to the number of turns from 0" to 3". (Because the actual adjustment mechanism turns on a pivot at finite radius, it can't be truly linear, but it's probably quite close, and, as you will see, very forgiving when crank position change is measurable in single degrees.) A series of carefully controlled experiments will determine a tiny "fudge factor" that might be applied for smaller sets of increments, e.g., 0"-1", 1"-2", etc. However, since it's a little silly to strive to work with wood in increments much less than .002", there is actually little need for development in tenths of thousandths and thus the single 0"-3" count is likely to suffice nicely!
Back to the system - because my particular crank moves 10 revolutions per inch of height change, it will deliver about 0.100" change per revolution. The degree wheel , with 360 easily-read marks turning with the crank's shaft, is thus capable of indicating less than 0.0004" (that's four tenths!) of vertical blade movement for each degree of rotation. Expressed differently, 0.001" of blade height change requires about 3.6 degrees of crank rotation. If I allow 0.001" of slop, then it's pretty easy to mentally calculate the number of degrees to turn the crank to achieve whatever height change is desired, and with considerable accuracy. If the marks are at one degree, then you should be able to interpolate to 1/2 degree or less with tolerable accuracy, and thus be able to accomplish calculated change of less than 1/5 of one thousandth of an inch, should you really wish to make yourself crazy doing that. (The real point here is that if you're off a degree or even two, it don't make no nevermind in practical terms.)
For your own saw, measure the number of turns required to lift the blade 1", then divide by 360 to learn the resolution you might expect with a degree wheel. Double-check the range 2" - 3" and compare, just to be sure that there isn't a glaring difference in rise/revolution for extremes of blade travel.
NOTES:
- I enlarged the degree wheel's tiny center hole to 5/8" (as needed for my particular saw) with a sharp 5/8" Forstner bit in my drill press.
- Peel off the blue surface protector sheet.
- If there's inadequate depth for a shaft collar, you may find it expedient to attach to the backside of the crank hub. In any case, it's useful to minimize viewing parallax by mounting the wheel as closely as possible to the scribed reference line. If the wheel can't be brought close to the case, a raised bar with the reference line can be glued or screwed to the case.
- Don't worry about how many actual degrees are registered in use - the only thing that's of interest is how many degrees of change are needed from the present position, wherever it may be.
- Because there's some measurable amount of backlash in any such mechanical system, it's wise to make corrections "coming up" - that is, when raising the blade so that the system is preloaded as the new setting is approached. Thus, if the needed correction is downward, e.g., minus 0.008", then it's best to calculate the desired new wheel position, back down perhaps 30 degrees too far, then carefully raise to the correct new mark.
=========================================
Lee Valley makes a nice guage in aluminum alloy. Heavy and stable enough to stand on the table over the blade. It does not measure the height, it has preset notches for setting the height. common settings. i.e 1.4, 1/8 1/2 etc.It is $38.50 Can funds and can also be used for routers, fences etc. (page 93 in their catalogue) or http://www.leevalley.com
Keith
Hi, Keith...Thanks for the steer, but I've no problem with measuring and adjusting rough blade height. (To me, "rough" means anything more than +/- 1/256th".) The real problems begin to appear when trying to adjust to a max clearance of, for instance, 0.004" for a tenon. Duplicating in thousandths can be daunting, and a lost setup can mean a lot of time in getting back up to speed. By actually being able to predict blade movement in units less than 0.001", I can now very quickly adjust with remarkable accuracy.With upgrades to my saw and application of a Freud SD508, GlueLine Rip, etc., I'm getting really exciting performance and reliability, and want to tune up my measuring techniques to use those new capabilites to their full potential.Next project is a microadjustable tenoning jig....
Whew .........................!!!I use a slightly more elaborate version of John White's idea -- a 6" board with horizontal lines at various heights -- 1/8 ", 1/4", 3/8", 1/2", etc. Drew the lines with fine ink pen. Heck, I even varnished the board and drilled a hole in it so's I could hang it on the wall.
Understood - that's my "rough" setup. Now whaddaya do with a piece of foreign hardware that doesn't quite fit careful English layout dimensions? You (er, well, I, at any rate) measure the piece with a caliper and adjust to suit. If I find it's .021" off, for example, I can change settings by exactly that amount by simply dialing in 75 degrees of crank rotation - no squinting, no guessing - just change and go. Assuming, that is, that it's blade height that needs to be adjusted...! This (from my viewpoint)is about "the fine tuning as required" mentioned in another posting above. Fine tuning can take several passes before it all comes together. Precision dial-in just saves time so I can get to the fun stuff, like putting it together and finishing. Every thousandth I can save in sanding or scraping is worth a goodly amount of precious time.
Edited 2/15/2005 3:15 am ET by John
I don't have any quarrel with your guage -- none at all. It's just that you lost me in the complexity of your description. I never, ever criticize devices that other people come up with to solve their problems. If it works for you, I'm fine with that. Out of curiosity , though, what type of hardware application are you using where TS blade height settings is an issue?
Hi, again..Yeah - I know I tend to rattle on - it seems to take at least a thousand words to describe one simple picture. Will post a photo as soon as I have it. (That's another story. My poor old PS2-350 computer was state-of-the-art when I built it, but it groans and rumbles with new apps. Neighbor took the digital pic, will transfer to disc, and give it to me tomorrow.)I agree with you about others' solutions, though I really like to dissect others' stuff and try to learn something along the way! I certainly didn't mean to sound as though my feathers had been ruffled, and apologize if it seemed so.As to applications - gosh, I've had flashbacks of (too many) years of trying to achieve perfection the hard way. A simple example might be that of insetting a piano hinge. Small cross-sections are diffcult to get exactly flush, and inability to adjust hardware with sanding or scraping its surface makes fitting a little dicey. The hardware you may have in stock is of course nothing like what you buy tomorrow, even though their specs and appearance may seem to be identical - virtually every piece must be pre-purchased and individually measured in an extreme case. Worse, the same stuff in the same bin may vary quite a bit from piece to piece.My approach in a case such as that is to figure about .006" finish build after rubbing, set the dado blade to cut exactly the hinge thickness less .006", and tailor that cut depth according to differing thickness measurements from each intended piece of hardware. It works, but it's a bear to try to adjust in thousandths with a ruler graduated in 64ths! (I could fit it all strictly by hand, but I've too many projects and too few years left - gotta help myself as best I can...)Another example might be fitting a large post and apron, such as for a workbench. Since ya can't very easily stand a 6' apron on end to make a tenon, it's likely that you'd create the mortise in the post, then use a dado blade to crosscut the apron's tenon to suit. The tenon's thickness is critical, no matter how thick it might be. If you're shooting for uniform front/back offset, and .004" tenon clearance, then it's awfully handy to be able to measure one trial cut and dial in corrections. In the "good old days" of my long-gone youth, I enjoyed fiddling with a chisel to pare a tenon to "perfection" (except when a whisper was a skosh too much and I had to start over...!) On the large stuff, though, it just takes too much time, and I prefer to get as close as possible with machinery before finally hand-fitting.A bit of a departure...
I deeply and honestly enjoy the feel and song of a plane as it cuts, and similarly enjoy the feel and crrrick of a razor-sharp chisel in fine paring, and so on - there's a visceral quality to the experience of handling and shaping wood, just as there is in sharpening and then applying the tools at hand. That emotional connection becomes a burden for me when I've deadlines to meet (even those self-imposed), and I find that I'm less stressed if I can efficiently use my machinery to speed progress, and perhaps thereby put a little time in the bank for going back to handwork now and then for the pure joy of it.I've enjoyed our exchange, and hope I've not bored you to death.
You're the first woodworker I have encountered here that works in thousandsth of an inch.But if that kind of precision gives you pleasure, that's great. You have to be the one to judge if the frustration level ever outweighs the pleasure you get from the craft.
There's a difference between precision and accuracy. If you care whether your table is 6' 2.002" long, that would be accuracy measured to thousandths; woodworkers don't bother with that. On the other hand, if you care whether you have fit a tenon snugly, and would use a shoulder plane to remove a shaving .002" thick to make it fit better, you are working at a precision measured in thousandths, though you don't care absolutely what the thickness of the tenon is to that accuracy.It sounds to me as if John is saying he's worked out a method to change the height of the blade in increments this small, reasonable for precision of fit. He's not trying to set the height of the blade to an extraordinary level of accuracy.
Hi, AlanS...pleezta meet you, too...You got it exactly. There's no point in trying to make a cabinet that's 48.001" wide, and thousandths have no place there. But adjusting its joinery for good fit is crucial to its strength and integrity.Hand-fitting more or less "guesses it in", and I've yet to meet anyone who can plane or chisel to .001" accuracy over an entire work face. If a machine will permit such accuracy, then joint strength is improved markedly, and requisite final planing/shaving at assembled joints can be minimized. ... Not to mention that a useful dial-in approach is *fast*
John,
A good solution, but I would probably build it with a shop made disk with one hundred divisions around its perimeter, that way it would read .001 directly, avoiding the degrees to inch conversion. Some way of zeroing the scale would be nice also.
Some of the benchtop saws, and the Ryobi BT-3100, have truly linear height adjustment mechanisms where this approach would be absolutely accurate through the entire height range.
John W.
Hello, JohnW...Pleezta meetcha...You're right, save that all raising mechanisms don't necessarily register exactly 10 turns per inch. If not precisely 10, then the 100-mark wheel would require the same arithmetic to determine "marks per thousandth." Also, 100 divisions is much more coarse than 360 - the 360 division wheel is 3 times more accurate than one with 100 divisions.It's all relative - once you determine the value of change per division, it's all downhill, regardless of the actual hardware in use.
Your system is going on my saw. Perod! Thank you very much!
One more thing. Any 360* protractor would work. the bigger the better. Alot of them are plastic and work fine since they are used for drafting.
Edited 2/16/2005 12:08 pm ET by david
Hi, David...
You were one of the few who understood the implications of the micro-adjust method for table saw blade height, so I'm trying to catch you up to date with where I've gone since that flurry in the Forum.
Idiotically, I only saw a protractor when I looked at and purchased the protractor I listed. I finally saw that it read at +/- 180 from 0, rather than a full 360. Ratz. Found a very useful substitute here, if you're interested: http://store.artcity.com/cth-255.html
I've found the smaller size to be entirely readable, and manageable for other applications where placement of a large one might be awkward. Just finished a micro-adjustable mortising table for my drill press mortising attachment, using that small protractor for fore/aft positioning.
Best wishes,
---John
Thanks John, I'll folow up on that 360* protractor. It was thoughtfull of you to post back at this late of a time. You don't see that much these days. Thanks again.
use a ruler
Good idea!!!!!!!! :-) Gosh you all make something that is the simplest of the basics tooooooooooo hard and complicated. One thought above somewhere noted the age old idea of raising the blade to a mark on the wood you are going to cut and steelheader said use a ruler.......Folks, those two ideas is all you need.
For those of you who have spent money to get to within a gnat's #### of perfection have wasted a lot of bucks. However, you are keeping companys that have some good stuff in business so for that I thank you.
We here in the Very Deep South do not understand any thing past the most simple basics:-)
Dick C.
Starrett or B&S combination square. Set the height, check for squareness at the same time. Also handy for setting 45° angles. Oddball angles for the mitre guage, a Starrett or Mituyo (sp??) machinist's protractor on a combination square blade. For the router, a B&S 6" dial caliper.
If you're looking to purchase any of them, check e-Bay and your local pawn shops.
Regards,
I use a Fowler height gauge from J&L supply. It is way overkill but does a great job on table saw and router table. Has a hand wheel and mechanical digital counter and dial indicator. Just make sure you catch the top edge of the bit or blade tooth, take backlash into condideration and cuts will be dead nuts. Link below.
KDM (Tool Junkie)
http://www.jlindustrial.com/catalog/product.jsp?origin=SEARCH%3ACATEGORY&id=FOW-74212K
Product Detail
Edited 2/16/2005 10:24 am ET by duke-one
You guys are so far ahead of me in precision that if we were going in a circle, I would soon be ahead, as you come around the second time. Anyway, this is NOT a smart-#### question, it is real.
If you are fitting joints this close, what do you do about glue thickness? Do you build in a space for it or can you get enough by just wetting it? I have heard of "glue starved" joints but they are never a worry for me. I usually need a good gap-filling glue. I am sure I will never reach the point this is a worry for me, but I am curious.
The tolerances you are talking about are many times thinner than a human hair. I have watched Tool and Die makers fit metal parts in moulds to .001 inch and unless it is surface gound (essentially polished) they won't go together.
"If you are fitting joints this close, what do you do about glue thickness?"Keith,As I believe another poster already noted, the purpose of the jigs, fixtures, and other devicss that allow such remarable precision is NOT to achieve joinery with a precision of three decimal places, BUT to enable repeatable machine set-up that reduces trial and error methods commonplace in workshops worldwide.There may be folks who precisely cut their mortices and tenons and use feeler gauges, micrometers, and black magic to determine how many microns to allow for glue thickness, but they are probably being housed in sleeping quarters with padded walls and being visited at regular intervals by folks who wear white lab coats while dispensing potent medications.-Jazzdogg-
Whether you think you can, or you think you can't, you're right.
O.K. got it -finally.
Keith
Not as much precision as ease of reading. My dial caliper has a nice, large dial to read. Easier on 50-ish eyeballs.If I'm a few mils off of 3/8ths it's not a big deal. Wood moves.Leon Jester
Hi, Keith...Here's my "take" on it...You're right - a thousandth is right next door to an "interference fit." It isn't that you want .001" matching - it's that you (I) want to decide on the desired clearance and then achieve it quickly and accurately over the entire area of the mating surfaces.A tenon, for example, will require about .003"-.004" freedom to fit properly and allow some space for glue. It should dry-fit with moderate to heavy hand pressure only, and should stay in place when inverted - that is, the tenon won't slip out of the mortise under the weight of the workpiece alone - ya gotta work just a little to get it out.You can get away with considerably sloppier joints, but they'll suffer for strength. Similarly, if the entire mating surface(s) don't enjoy the same clearance, then there are bound to be some (many?) spots with lower strength. You can get darned close with a bullnose or shoulder plane on the cheeks, but you can't get it as nearly perfect by hand as you can with properly set up machinery, nor can you do it by hand as quickly. A part of the argument has to do with whether you're trying to get a job done, or spend time enjoying the woodworking process. For enjoyment, I love nothing better than all hand woodworking. However, I have not had the privilege of a 20-year apprenticeship (the medieval equivalent of grade school through advanced degree) and have a hard time making sufficient progress with hand work alone. Therefore, I try to use my machines to quickly achieve the "perfection" that I'm still trying to develop in my hands.That said, understand that with precision measuring/adjusting you can create mortise and tenon joints (for example) with table saw and drill press mortising attachment and quickly arrive at perfectly flat and square tenons that match the mortises with exactly the desired clearances on all 4 sides. Accurately controlling placement of the tenon across the workpiece thickness will cause faces to align perfectly - even if the workpieces aren't the same thickness - and save a lot of time and effort in trying to get the joint scraped and sanded flat after assembly. Here again, thousandths count.Think about gluing. If you make a simple glue-up, you mate two flat faces and clamp them, right? Flat to start with, and then, fer pete's sake, you clamp 'em tight. Not much clearance, eh? It takes little glue to form a maximum bond if the surfaces are properly prepared. Since you can't clamp a tenon, then there's something to be said for accuracy in its cutting.Could you do a good job by hand? You bet. Could you do it faster and more accurately with your machines? You bet.Here's an example of the benefit of speedy setup:
I was working on a jig last night (for the same reasons that people buy Starrett tools, I believe that jigs have to be very accurate) and needed to plow four rabbets exactly 1/8" deep. That's .250", not 1/8 +/- 1/64th or even 1/256th. I set blade height with a precision rule, made one trial cut; used micrometer calipers to gauge its depth; adjusted blade height with the degree wheel; and made a second pass that measured at less than .001" off - all in less than one minute! There are those who argue that an accurate depth-setting gauge could be used instead - just raise the blade into it. Well,maybe. Now let's say you wanted to leave exactly .500" *above* the blade, and the stock was some (as usual) weird thickness coming out of the mill? (Similar argument for setting cutting depths for a half-lap joint.) Ain't no height gauge made that'll do it as well or as quickly as a degree wheel. A setting gauge is a start, the degree wheel (or other device capable of rendering tiny increments) the tuning device. The method also absolutely does not require that the blade touch a steel surface.For myself, I'm currently contructing a mortising fence with micro adjustment fore and aft and micro-repeatable end stops, and a tenoning jig with micro adjustable placement and width control. Saw blade height is already controlled to less than .001" Need it as well for saw fence placement and router table - er, see "Incra Jig" - several hundred dollars, or a simple adjusting screw with a degree wheel?Once again, it ain't about working in thousandths - it's about *adjusting* in thousandths for the sake of strength and stability.
put a mark on scrap,then move saw up or down till on mark,seems to work
Ditto. But with over-40 eyes, I put on a pair of Opti-visors as well.
And in general, I avoid getting into situations where absolute accuracy in height-of-cut is necessary. Relative accuracy, yes (i.e., all pieces need to be cut the same). but I just cut all the pieces at the same time to ensure that.
You have some good ideas John. I know I will never get that kind of accuracy (precision?) but that does not mean not trying for it. I am curious about your idea for a micro-adjusting tenoning jig. I just bought a Delta and the pointer was so coarse as to be useless. I filed it to a knife edge, but that does not help with the coarse scale it points to. I have been thinking about mounting a dial indicator on the otherwise stock jig. The fine adjustment screw on the jig seems to work well eough and I polished the mating parts before assembly so it slides fairly smooth. What are you planning to build?
Keith
Hi, Keith....Thanks for your kind remarks - and, yes, you *can* get that kind of precision - just gotta keep fighting with it 'til it works!(BTW - I like the short and sweet responses, but some of these thorny issues aren't so easily dismissed, so I rattle on a bit ...)I haven't yet designed the tenoning jig; it's just a flock of conceptual approaches floating around in my head (la-la land?) at the moment. I'll share some thoughts, and want also to discuss a little about the Delta - I'll do that first.I've not held your Delta jig in my hand, but I suspect that its bones may be good, but execution lacking here and there. If it's sufficiently rigid, then your observation about the adjustment screw is encouraging. That bit of polishing or lapping here and there can work wonders on oherwise so-so motion. Fit of bar in table slot is important as well, and it might need some kind of "stabilizing outrider" addition to ensure against lateral rocking.The dial indicator sounds like it might work, but I envision some limitations, such as the fact that very few of them will indicate over an inch - suggesting that you might experience terrific accuracy over the thickness of the tenon (most of them, anyway), but not be able to precisely control/indicate its width when it's turned 90deg. Also, depending upon how and where it's mounted, it's entirely possible to see enough flex in the assembly that the gauge's accuracy could be marginalized - that's to be fiddled with and overcome with a little field work before finalizing the design approach. You have an adjusting screw - uhh...why not consider attaching a degree wheel?
------------------------
Just for the sake of kicking around ideas, here's where my head's going on jig design:[ Point of order: Machined mortises are made first, and with the "front" or "good" side of the workpiece against the fence, so that all are offset exactly the same distance from the workpiece "good" side, regardless of variation in stock thickness. I have a standing self-imposed rule that requires every piece of stock to go through the planer - unbelieveable, but I actually got tripped up on some "primo" stuff whose faces weren't parallel! These days, the commercial stuff ain't what it used to be... Besides, planing helps to ensure uniform stock thickness for an entire project.]Arrrrgh - a flash of insight tells me that the age-old "cut-'em-vertical-on-the-saw trick" may not be the best approach for tenons - verticality alone is a challenge with long stock, and made worse if it's heavy stuff - how the hexx do you hold a long skinny stick perfectly vertical so that even vibration-induced movement isn't a factor? (Not a problem on small projects, but a 4' span in a large case makes for a pretty unwieldy workpiece.) Then we're gonna shove on the whole assembly to make it pass the blade? Goofy. I've seen huge fence-guided box jigs that make a pass at the stability problem, but they're pretty awkward. Really large tenons that have to be dado-cut flat on the saw table are just as bad. How do you move a 6-foot-long 2x6 oak workbench rail across the blade using a miter gauge..??? You and Atlas and all of his friends, maybe?Gonna look really *hard* at a router machine/jig for tenoning.
"Get 'er down, nail'er in position, and move the cutter." I'm guessing that it may require an extra trial cut to get it dialed in to perfection, but the time saved and improvement in accuracy would be well worth it. Today's routers are already microadjustable for cut depth, so face registration and cut depth can be pretty straightforward. (Gang-cutting several pieces may be possible, but I'd predict problems with clamping and face registration because of variation in workpiece thickness - we'll see how accurately the planer really works for "absolute" thickness unformity - doesn't matter *how* thick - just *same* thickness. With a desired limit of .003" to .005" room in the joint, planed thickness needs to be uniform within about .002" or so.)Thoughts are gonna jump around a bit - be patient - sharing them in the hopes of initiating fruitful discussion ...Boy - my head's in overdrive. This thing will look very much like a 12" planer, but instead of the usual cutterhead, it would have workpiece clamps and an overhead router with dual cross-travel --- on Thompson ball bushings (even just Oilite bushings on a pair of polished 60-case shafts would probably do for many years of service. ?? Or maybe even well cut and waxed oak sliders...??) See the new Jess-Em fence-mounted precision miter gauge for a picture of the ball bushing concept. It only needs to cross-travel perhaps 6" to be able to make a huge tenon - maybe an 8" travel spec will make it bulletproof, plus "cross-cross" travel of up to 4" to span a really l-o-n-n-g tenon. Vertical travel to clamp and work a max thickness of perhaps 4" would be more than plenty - 6" for bulletproof - oooops - stock width makes that at least 8". With that range, it would do *logs*!Let's get practical about the ball bushings. You can make great dados by just riding the router on the surface. Seems to me that if the router simply slid on a flat, polished and slotted platen, that at least one direction of motion could be solved without the need for fancy guides.A locking planer-style raising mechanism permits vertical movement to accommodate variation in stock thickness. Once it's set down to the workpiece, the router height adjustment would control cut depth. Since the router will adjust at least an inch, carrier height adjustment isn't micro-critical, so long as it's reliably parallel.Depth of the actual cut measurement is based simply on a vertically mounted, spring-loaded finger or plunger attached to a dial indicator (or somesuch) - - push down to register the stock face, slide or swivel to the cut and push down again to register the cut face, then subtract to learn actual offset in thousandths - should match the "good face" mortise offset within a few thousandths. Affixed to the router slide, it works with the router in place, so the workpiece need not be removed for offset measurement on the first cut.I'd knife and then pre-cut crosscuts on the table saw to define tenon edges - the kerf shoulders can then be used to locate the tenon in the clamp mechanism.Now I gotta work on the edge width cuts .... Same problem with dial indicator range. The Fowler double-beam height indicator that Duke-one suggested is along the right lines, but perhaps a bit more than would be needed.Would like to think in terms of making as much of the thing in the shop - and of wood - as possible .... oak's pretty reliable stuff. Add plenty of tuning adjustments. There are likely to be a few(?) parts that will of necessity require machine-shop services. Someone observed that "wood moves" - that's right. But fine-tuning adjustments can be devised so that daily or seasonal checking and tuning are relatively quick and simple.I trust this has you as dizzy as I am...??? Hope it sparks some fried brain cells on the part of yourself and others considering the same problem, and that y'all might share ideas as time passes.
You sound to me like you get at least as much pleasure out of inventing as you do from woodworking. I do too. The following suggestion may spoil the fun, but for what you want, you should get ahold of a metal workers milling machine. The cutter speed is a bit low, but you have perfect control. Clamp the work to the table and crank the cross slides. You usually also have a digital readout accurate to .0001 or better. If you found one with a burned out motor (cheap) you could modify it to carry a router, a fixed base would be fine, all the adjustment is done by the milling machine. They are available with either vertical or horizontal axis. I think you would want a horizontal for your big pieces. They are not real expensive used, as most shops have now gone to CNC equipment. Yes, I guess you would have to mount a router as these machines are all usually three phase power and often 440 volt. By the way I am the guy who posted the "crazy" idea about using a Radial saw for the same thing, but some writers rightly discouraged me because the carriage on these saws are pretty loose. Not problem with a milling machine. With milling machines the head is stationary and the table cranks up and slides in two dimensions, so mounting the router solid should be easy. All you need is a fair bit of room in your shop. While thinking about set up, I recall the Tool and Die makers made a lot of use of "joe blocks" for initial set up. Precision ground and polished, they are dead accurate and never lose their setting. There are lots of things the metal boys use that are applicable when talking the degree of precision you are after, Try to get a guided tour of a Tool and Die shop, not a machine shop. The degree of precision is an order of magnitude up from a regular machine shop. Even if you don't use a milling machine (I suspect you won't) take a look at one just to get some ideas.
Edited 2/19/2005 6:10 pm ET by Keith
Hi, Keith...Time seems to pass. We'd briefly discussed accurate jigs, and you requested info about an anticipated tenoning jig. I'm not there yet (overcome with other opportunities!), but I've cobbled together a micro-adjustable rig for a drill-press mounted tenoning attachment. (Nice to be able to achieve dead-center, or any desired offset. A second real advantage is that its guide system ensures that the chisel remains square to the workpiece if the fence is moved.)Let me know if you'd like to see pix.Regards,---John
Hello John,
I don't know about Keith , but I am keen to see any pictures of rigs, jigs, tools etc made by w/w's for same-so don't hesitate.
Gotcha. Will take a day or two to process decent pics, which I'll make accessible online. Will post when they're "up." Made aux fence for table saw, too, with replaceable inserts. (Finally getting around to cobbling up some of the stuff I've needed for years.) "Can't take time to sharpen the axe - I've got all these darn trees to cut down..."Regards,---John
Thanks-must be early morning where you are?
"Finally getting round to cobbleing up...."- I have recently had a good run of that:made a knife setting jig for "difficult" surfacer, a set up to grind my machine knives using the sliding table of my shaper and some marking tools based on the business end of the common throwing dart.This was not done during normal working hours.
Keep well.
You're inspiring me. Would really like to learn some details (pix?) of your approach on knife sharpener. One of my "to do" projects is a rig to allow me to flat-grind hand chisels and plane blades on 8" bench grinder - and hopefully to do a decent job of it. I can (I think) come up with a usable slide system for both cutting and dressing, but am getting a little sideways over accommodating depth adjustment without destabilizing the setup.(Am in WA State, so have not quite seen the witching hour here.)---John
O.K, John, I will take some photos tomorrow and send them that night if possible. I have done this specifically for my surfacer and planer knives,but it could do any blade that can be suitably mounted.
"you're inspiring me...." Whoa there John-I have just read this post from #1 . I now think that you may find my set ups a bit primitive as I rely on the creep-up approach-but from then onwards the "thing" must work accurately, smoothly,consistantly, stay set and so on. I make mistakes reading scales, dials,micrometers etc so rely on "other' ways...When doing engineering work I then go into the precision measure and dial it on mode-but not for woodwork.
Anyhow, the basis of tthe knife grinder is there and certainly lends itself to the refinements you are good at, so I will send details soonish.
I'm not sure why people are so obsessed with measurements. The measurements themselves don't mean anything. They are just an over-complicated way of transferring a dimension. From what I've read here people are over-thinking things. Don't focus on the number, focus on the fit. You are not making a set up to produce 10,000 parts. Make a jig, or template, if you need to duplicate a piece at a later time. Woodworking is not nuclear engineering. If you are off by .001" your piece isn't going to melt down, and contaminate the neighborhood. I find that I have to make things "idiot-proof", because most of the time I'm the biggest "idiot".
It either fits, or it doesn't. If it's too tight, plane it down. If it's too loose, cut another piece.
The more you measure the more chance there is for error.
Relax. No one has a corner on ideas, and I'm always very grateful for the inspiration found in considering others' concepts. I too do a lot of creeping up, as specific needs for versatility have a way of revealing themselves just after "the grand design" is finished...Your "accurately, smoothly, consistently" approach is exactly what I shoot for. Doesn't have to be pretty or elegant - just functional and reliable.Lest have the wrong impression of my approach to dimensioning -
I do NOT work in thous, but I DO find it very handy to be able to adjust with precision. That is, to measure a desired change and then dial it in quickly (that's *change*, not actual.) Thousandths-versus-degrees-of-rotation is a natural association for home-built adjusters. In too many cases with the "skosh more or less" approach, the process is more like a damped response - a little over, then a little under, then over, then under - always in smaller increments until the right setting is attained. With very simple angular dial readouts for setting changes, I can very quickly get great results with one adjustment, maybe two depending upon criticality. Not elegant or fancy, but cetainly efficient and reliable. Not always necessary - I find that I use the dial on my tablesaw height adjuster perhaps 10-20% of the time - but it's a life saver when I need it.Looking forward to seeing details!---John
Hello John,
I get the concept: would like to have it handy if and when I get a decent circular saw...
Herewith some pics of my grinding set-up.Note- it is the concept I am touting here, as you will see that the components are one-offs in some cases.This does not mean that it is impossible to do-if you have spindle with reasonable sliding table you could mount the stone on the spindle arbor itself rather than use an independent motor as I chose. I just had all these things on hand and cobbled them up to make the set-up more versatile.
The only cost was that of getting another four guide rollers made up for the table- but I always thought it could benefit from this anyway.
The device holding the knife is made up of the milling table from my Emco Metal lathe, also the green angle block. The knife is clamped in a piece of brass angle, which I milled grooves into and drilled and tapped to suit and that is bolted to a mild steel bar that was lurking around here (i.e it did not come from Zimbabwe).
The most important item is the little table feed that the motor is bolted to-got this out of my store of things called "this is a good thing-will keep until I find good use for". It is ideal ,with dove tail guides fully adjustable and a fine enough feed-something like 1.5mm per revolution of the hand wheel and no back lash.
The motor is .5hp 3 phase-I resisted painting it for now, but replaced bearings and removed wasp nest(vacant) which came from Zimbabwe.
P.S-can someone please tell me why I can never attach pics the same way twice, and why this time the quality is all blurry???
Edited 5/31/2005 5:26 am ET by mookaroid
Edited 5/31/2005 5:29 am ET by mookaroid
Mook, picture quality seems fine to me. #49 might be just a little out of focus, but not really blurry. Do you have a web accelerator program running while you're browsing? forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)Another proud member of the "I Rocked With ToolDoc Club" .... :>)
Hi forestgirl,
Thanks for your response. I can't answer yout question because I don't know what a ' web accelerator program is'.I use Ffox after I.ex. became idiotic.Please tell me all....
A web accelerator program is provided by some internet service providers (Earthlink, Netscape for instance) to help out us poor dial-up customers get pages to load faster. My understanding is, they would store some graphics and stuff on your hard-drive for faster loading. The other thing they do, and why I asked, is they compress pictures for faster loading.
Because the pictures are compressed, they don't have the detail they would otherwise. You can turn the accelerator off when needed to prevent this. Awhile back, we had one member had big problems with his attachments when he clicked on his own post because the pictures were so compressed they were all pixelated. Once he adjusted the web accelerator, they were fine.forestgirl -- you can take the girl out of the forest, but you can't take the forest out of the girl ;-)Another proud member of the "I Rocked With ToolDoc Club" .... :>)
Mook, the photos look fine to me.Thanks for posting them.FWIW, I'm using DSL and Firefox as a browser.Leon Jester
Leon,
Re the photos- is it possible for one computer to show the same photos in adifferent quality to another? John used a good description- said some were 'jaggy'- would this happen if I have rotated them when putting them into Picasa?
Re the photos- is it possible for one computer to show the same photos in adifferent quality to another? John used a good description- said some were 'jaggy'- would this happen if I have rotated them when putting them into Picasa?
################It's possible. Could result from several things: screen size (1040x768, 1200x1040, etc.), program used, how well a monitor is calibrated or an ageing monitor.If anyone's interested, this is the url for an SMPTE‡ test target:http://brighamrad.harvard.edu/research/topics/vispercep/smpte/smpte.jpga set of Kodak grey scales and colour standards for graphic arts (warning, this is a huge file, very high res.):http://www.ezslides.com/4000full.jpga generalized set of test targets:http://www.cinedrome.ch/hometheater/testpatterns/I'll suggest that anyone interested in calibrating their monitor use the SMPTE target. Adjust your brighness and contrast until you see (ideally, depends on your monitor) a small grey patch barely visible within the 95%/100% block and a black and almost-black box within the 0%/5% box. The target itself should be in black & white. If it's got any colour at all, adjust your colours until it's black & white -- with the obvious shades of grey. Most monitors will not show the separation of the 0/5% box. The 95/100% box should be visible.The Kodak one is a high-res scan of their graphic arts grey scale and colour control patches -- which used to be available from Kodak dealers, I don't know if it's being made currently. The grey scale should run from dead black to white, a total of ten steps. It is useful to folks that do graphics, but probably not of much use to others.If you can't get the SMPTE target to be in B&W without colour, your monitor is probably in it's death throes, or about to begin them. ‡Society of Motion Picture & Television EngineersGood luck,Leon Jester
Leon,
I have been looking at the smpt thing-can't understand a thing-you need to be here....
Meanwhile am posting pictures at a hell of a rate, but now notice that the size is creeping upover the 100mpx range. Iset camera to the smallestsize i.e 640x480 and setjpeg at0%. Itried resizing down but the quality goes out the window then....
BTW-have solved the rotating thing!
Thanks for your advice.
The SMPTE targets give both resolution -- which, with a modern monitor shouldn't really be an issue -- and colour.Bring the target up, then adjust your colour settings to get it black & white -- no colour.When it's neutral, adjust your contrast/brightness so that the steps I mentioned show up -- don't be surprised if the dark ones don't show, they're very difficult to do.If there's colour again in the target, go back to your colour settings and adjust them so it's neutral again.If your photo editor has a "post to the web" setting, use it -- it should take the pixel count down to 72. The whole thing works better if you shoot at a higher setting, 1200x1040 forex, then save it as a master, reduce the pixel count to 72 and the size to the smallest setting (mine lets me use inches, I usually use 4x6) and save that one, (this maximizes detail) it's the one you use to upload to the forum.Good luck,Leon Jester
Thanks Leon,
I think I can now get to grips with this using your guidance.
Glad to hear it's working out.Nice assortment of planes, BTW.Leon Jester
Sorry for the late reply,I have been away, but yes I would like a picture of your invention. I just bought a dial indicator that I planned to rig up. I would like to see your idea first.
Regards Keith
Hi, Mook ... and anyone else interested..."The jig is up" .... link at page bottom.Your sharpening jig is an imaginative use of a buncha stuff lying around - wish I had your junkpile! It looks to be stable and useful - just the kind of thing to which I'm personally attracted, as it can save calendar time and puts you in control of your own destiny, so to speak.As to your pix - they're certainly viewable and illustrative. I do see quite a bit of "the jaggies" - lines with jagged edges. That's usually the result of artificially resizing a jpeg image. Is that what happened? (Let me know if you want to get into a separate Email discussion about image processing.) Can't comment re differing posting results - I've not tried direct posting to the forum.I've a start in sharing pix and ideas here:
http://www.primeshop.com/access/woodworkRegards,---John
Edited 6/1/2005 4:17 pm ET by John
Hello John,
Thanks for the link-have bookmarked it as a source of ideas. I am thinking that your 'protractor gauge' would be good on my thicknesser, as even though I have a scale there marked in mm's, when it comes to fine adjustments it is not good- up to now I have got by by imagining a clock sitting behind the handwheel.... Now , instead of the imaginary clock, you have an easy to see 360* protractor PLUS a rotating arm which can be clamped at any of the 360 degrees as a reference.... then you know what you are adjusting to....this is a bit like an engineers dividing head and if I recall there is a similar concept on certain artillery gun laying devices.
I think I am going to sleep on this first.
Re the pics- I did not do any re-sizing, but I did rotate one or two.Or maybe they just got worn out by me moving them around so much when I was (again) working out how to send them when in forum mode!
Hi, Mook...Good - hope you found something halfway useful! I'll tell ya - even at 3-1/2", the 360 is very easy to read, and, for most apps it takes 3-7 degrees to indicate .001 linear movement, which allows for very fine adjustment.Hmmm - you got to link OK. Must not be using MSIE for browser...? I was embarrassed to have forgotten to check HTML against IE, and succeeded in hiding the place from nearly everyone - got it fixed.
---------------------
---------------------
If pix posting isn't of interest to you, there's no need to read further ....DIGITAL IMAGES:
Don't know where yours are coming from ... scanner or digital camera? Rotating can be the source of distortions, particularly if the images are at low resolution. Jaggies are the most frequently seen errors. (Reason - the processing program has to invent pixels for changed areas in lo-res pix, and they don't all make very good interpretations of what's needed.) One of the tricks is to keep resolution high, preferably in bitmap or tiff format, and do all the processing, rotating, cropping, and what have you at high res -- it results in virtually no induced distortion. (Reason: Hi-res images have so many pixels running around that the processor doesn't need to invent any.) Once processing is finished, THEN reduce resolution to reasonable filesize and post. Any help?[Incidentally, the monitor is a major factor, but the fact that you're seeing some of your pix as good and some as not-so-good suggests that your monitors' OK.]Scanner: Save max possible physical dimensions (but not larger than the photo you're scanning) in tif or bmp format - HUGE filesizes. Process those files for color rendering, rotation, etc., and save again - your corrected "masters." Open in processor, then crop, alter size, and reduce resolution, then save as different filename in jpg format for posting. Should see you go from, e.g., 300-1200kb hi-res to under-70kb lo-res.(By saving all those variations with different filenames - pic1, pic1a, pic1b, etc. - you have the option of later retaining or deleting those you no longer want.)As an alternate, if you know desired the "physical" (display) size, then tell the scanner to scan and save at that dimension, as long as it's not larger than the photograph being scanned. You can NOT successfully make (digitally) a small image larger with much success, but can easily make a large one smaller. Avoid gif format but for black and white line drawings.Digital Camera:
Resolution and "physical" size depend on camera's photosensor array - listed as "megapixels." A 5mp camera will save poster-sized images with 5 million pixels. All else being equal, the same camera at 3mp will save images of the same quality at smaller "physical" dimension.Consider your monitor, which is perhaps a 17" running at 1024x768. That means that it will display 1024x768 or 786,432 discrete image points - that is, pixels. A 5mp image reduced to 20% in physical area would more than fill the screen without loss of clarity.
However, if you reduce resolution before processing, to perhaps 70k filesize, then you only get a fraction of the screen's ability to display. Rotate that image, (or fiddle with it otherwise) and the program has to fill in makeup pixels, which it invents and which may be misaligned or poorly chosen - resulting in jaggies, smears, or fuzz. If you first process the image at its max resolution, then there will be no artificial filling in, and the program can more easily interpret what to do when you tell it only to reduce resolution.[No, "kb" filesize is not equivalent to pixels.
Each pixel requires several bytes of info for brightness and location for each of 3 gun colors, and each image must be defined for boundaries, filename, etc., all of which add bytes. Image kb is the sum of all of that info. There's more than that, but this is already 'way too long.](Note: Your dig. camera will most likely save only in hi-res jpg format - - perhaps 300-1200kb. That's OK - they'll process just fine if you follow the order suggested.)
-------
Upshot:
Make all adjustments at highest resolution - far too big to post. First rotate, then crop, then color/brightness, then dimensions, and finally save-as (newname) - still too big filesize, but it's the right dimensions, color, etc.
Once that's all done, then reduce res, save-as (newname2), and post.Depending upon what I'm illustrating, I frequently bite the bullet and post large files for pix, as lowering resolution is likely to also reduce clarity. I'll fiddle with one for a while at varying res levels, then choose the lowest that still shows what it needs to.[How does resolution change alter clarity?
Image processors affect res by altering number of colors, up to 16 million. If you lower res, you instruct the program to automatically reduce the number of colors. Now, let's say that at high res you have 4 pixels in a square at color values 1,2,3,and 4. When that's reduced, the program might look at those 4 points and make them all one color of average value 5. If the contrast among those 4 points and a zillion around it are critical to definition in the picture, then the substitution of a single color in place of 4 different ones will blur the original contrasts. Sometimes it makes a critical diff, sometimes not. Sooooo - ya fiddles and finds the least that looks usable.]All mud?... ?This whole business of digital image rendering is a quagmire - an impossible number of variables, and an equally impossible number of ways of coping with them. (The vagaries of trying to convert inherently analog functions to digital 1s and 0s.) That's part of why you see some processing programs at $30 and some (e.g., Photoshop) at $1000. The spendy ones do it better and do more things, but the common ones can do a credible job of what they do if you regulate the order in which you ask them to work. If graphics card and/or monitor are substandard, then the very best processing still looks substandard onscreen. Same kinda thing with digital sound. LOLKeep smilin' ...---John
John,
Briefly thanks for all that info-I have printed it off for on-going reference.
I have been "rotating" -so your information confirms that this is a source of the "jaggies".
Keep well.
You're most welcome, Keith. BTW - rotating ain't bad - it's just when and how ya do it. LOL---John
Mook:John has very good info. I didn't think about rotation on images, having been a professional photographer for 20 some odd years before going into radiology I tend to rotate the camera instead of the finished image.Sorry 'bout that.Leon Jester
Leon,
Now the laugh is that I have been rotating the bloody camera and then deciding to rotate it back when putting it into computer!
roflamo -- that's priceless.If you've got digital, or 35mm, shoot one each!Leon Jester
Dick,
I use my dial caliper. It is just wide enough to set the tail end (non the mounth) on the top of the saw and measure the height with the bit that slides out as you open the mouth. If you hold it just right you can crank up the blade and measure the change as you go.
Good luck.
Eric
I use a scrap Hunkof wood and cut it... Then measure...
This forum post is now archived. Commenting has been disabled