OK another thread on the forum as brought up tolerances in wood when milled and several threads have talked about what the tool is off perfect by. The question I have as a hobbyist is what kind of tolerance I can live with. I really do not know. I mean if a table top is off buy .012 or some such does it really effect me? I understand that in some tools a greater error is ok, and I would expect that in a hand plan (say) it would be more important to be accurate but to what degree? I have a brother that does very fine millwork (tool and die stuff) and he works in tolerances that are way down there and he talks about how many items get scraped while being made to really fine tolerances (he does a lot of small orders that would be hard to set up a CNC for I guess) but do we has wood workers need to get that close?
You always see articles listing these pretty numbers but if everyone is like me they really don’t mean anything to them. Ok tool X is out of wack by .012 and tool Y is out by only .007 so tool Y is better. But if a tolerance of say .4 is acceptable then the difference between the two is pointless.
So how do you know what is ok and what is not? The reviews don’t seam to ever tell you. That leads me at least to ignore them for the most part as the date means nothing to me, with out some to judge it against.
Doug Meyer
Replies
The end result is what you care about. If you can joint an acceptable edge on two pieces of wood on your jointer and edge glue them up then, it's good. It's really difficult to quantify the flatness of a table comparing a jointer to a table saw and there's lots of variations around that. For instance PM allows .015" on the PM 66. If it's .030" dipped and under warranty they will take care of it. For most folks that wouldn't "affect" their work. You can have a table well within tolerance and some guys won't get good results and will start to blame the tool.
I don't think it's worth to print up spec sheets on tolerances with machines as some guys will start to go crazy measuring stuff and will complain about something being .002" off the specs. There's no telling what he used to get that reading. I could see that as being a real problem that doesn't have to be. I find old metal working machines have tighter tolerances than new woodworking machines. Wood doesn't need to be worked in thousandths. Just a brezze blowing over a peice of wood will cause the wood to move several thousandths. If you can work the wood to 1/64" you are doing better than most aspire to and probably better than you need to.
Seams logical to me. But if a magazine is going to do a review and give dimensions of how flat something it (or what ever) then should they not tell you what those mean? Otherwise they are just printing things to impress you and not to inform you.
If I went to the Autoshow here and reviewed a new car and said it was nice but that the engine only produced 1.5 DRPU's and that another produced 1.71 DRPU's but never said anything about what is acceptable then it would be of no use. And while we all "know" what .012" means I don't think many of us (and probably most of us hobby types) have any idea if that will effect us.
I am like you I want to know if the unit will produce the results I want. I really don't car if the saw has a 2" dip in one side. I just want to get it to work. (Yes 2" is obviously to much) So unless they are telling us what is a number we can live with then why tell us the numbers? In an article that just reviews on unit you don't even have other units to compare the numbers to. So in that case you cant even say that this one is better then that one as you only have the one number. If you want to tell me that the table is out by .012 should you not say that on average for the last 20 saws we reviewed the average was XYZ and thus this is either better or worse then average? I mean give me something to work with here please?
Doug Meyer
Part of the effect of a tolerance indication has to do with the size of what is being tested. having a dip of .012" over a surface of 25"x32" isn't bad but if the surface is 2-1/2"x8", it can make more of a difference, especially if the dip is at an edge in front of the mouth in the case of a hand plane. If the material being worked will move in relation to the cutter or blade because of the +/- area, it matters. If it never changes that relationship, it doesn't matter. It would be great to be able to make machines that are totally without fault but it's not possible or practical. Also, if the tools used to check this aren't made to much tighter tolerances and are undamaged, the test is invalid because they could be the source of the inaccuracy.What they should do is guarantee that the top will be no more than .xxx" from flat over a specific distance when ABC model from XYZ brand of measuring equipment is used. If I use a dinged up piece of pine that I gnawed off of an old door with a circular saw that has a bent blade and a set of rusty, bent feeler gauges to check my tablesaw top for flatness and then complain about it, I should A) be shot for being an idiot and B) charged for any time the company has to waste in trying to rectify a situation that doesn't even exist. OTOH, if I find a problem when I use a straightedge or rails and a dial indicator that were tested by the National Institute of Weights and Measures, ASTM or someone like that and they're found to be accurate to within .00000001", I can reasonably say that the measuring devices are not going to be the source of the reading. "I cut this piece four times and it's still too short."
Edited 1/11/2007 11:32 am by highfigh
Good morning highfish...
I just posted my reply to this thread on Steel City Smear, post #38. I didn't know you had started this thread as I just caught it. My reply to you is there concerning a BS. And I would find run-out on say a TS or table flatness on a surface jointer a bit more critical if it's to the point that causes extreme out of square.
Even then when dealing with wood, it would not present problems that could not be easily over-come as compared to metal where friction plays a major role!
Regards...
SARGE.. jt
Coincidentally, I just started a Knots thread on this. It's: Runout or Red Herring?
Pete
Pete,
I think we where putting our threads together at the same time. I did not see yours until this was posted.
To the rest. I think what I am trying to say is that if you are doing a review and you want to give me some numbers then you should tell me what those numbers mean. So if you got a result .012 out then you should tell me what is good and what is bad and what is average. The number in and of itself means nothing. It is like saying the glass is half full. If you don't know the size of the glass you still have no idea how much water it has in it.
I am not sure that the companies should issue their tolerances as in doing so they will open themselves up to a lot of comments by people that will look at those even when they really don't need the info.
So I guess my point in starting this was more to do with reviews then companies.
Doug Meyer
Doug,
<<I am not sure that the companies should issue their tolerances as in doing so they will open themselves up to a lot of comments by people that will look at those even when they really don't need the info.>>
I don't know; it might be useful, or at least entertaining.... Back in the 70s and 80s, audio equipment companies went to great lengths to publish and advertise their specifications: x% wow & flutter....y%IMD....z% THD, etc. The audio reviewers would take the equipment into their labs, test them, and run it through their testing procedures to evaluate performance. Then they'd write their review. Most of the time, the reviews were a very good starting place, to help narrow down the amount of equipment to be auditioned before making a buying decision.
Most of the reputable companies published fairly conservative, although rather tight, specs, which their equipment normally easily exceeded. In fact, I remember that one company -- McIntosh -- was very well-known back then for the very conservative nature -- almost to the point of being ridiculously conservative -- of their published specs, and for the meticulous and careful construction of their audio gear.
I suspect, that were woodworking machinery companies to do the same thing, it would benefit, at least to some degree, both the pro and hobbyist woodworker, and would generate all sorts of fun and interesting advertising....Beste Wünschen auf ein glückliches und wohlbehaltenes Neues Jahr!
Tschüß!
Mit freundlichen holzbearbeitungischen Grüßen aus dem Land der Rio Grande!!
James
"Back in the 70s and 80s, audio equipment companies went to great lengths to publish and advertise their specifications:"Yes, they did and they showed specs that 99.9% of the people buying the stuff didn't have a clue as to what they meant. Then, it became a specs race, with more of some and less of others being the main basis for the buying decision. I don't remember how many thousands of people came in asking for brands we didn't sell because they were unreliable junk, but the specs were great. I also don't remember how many times I asked, "How does it sound?". Being conservative in showing tolerances when they're already in the .001" range is tricky- going from .001" to .002" is twice as bad! Oh, no! Think of the horrific life we would lead if a TS top was .005" from flat! Now, a sloppy hole, race, arbor, arm or something that will introduce wobble or oscillation is another story.
"I cut this piece four times and it's still too short."
Hi All, All the numbers in the world ( advertising specs ) don't mean a thing if your particular machine is out of wack. Mill some wood. The proof is in the pudding. If you're happy with the results, then your machine is within specs. Yes?
Have fun woodworking, Paul
Good point- if it does what is needed, it's fine. "If you're happy with the results, then your machine is within specs. Yes?"Not necessarily, but it's probably close enough for Rock & Roll.
"I cut this piece four times and it's still too short."
Ok, so now that pretty much everyone on this thread agrees that the specs are not of big use, and that the dimensioning craze in reviews does not tell you all that much, the question becomes what is the point of this in the reviews and how do we get reviews to quit doing the measure it up bit and start doing the how it works (or does not) type of review instead?
I mean if the numbers don't really mean all that much and that most of us don't understand the numbers then why give me the numbers? The space and time could be better used doing something else.
Doug Meyer
I think they should still show them because if they stop, the TS tops may start looking like a washboard. If I'm cutting a lot of short pieces (less than 18", I want the top flat and the miter slot nice and snug, with no wobble from the arbor. If I'm ripping, I want the top square to the blade, the fence parallel to both and the rails straight, not bowed. For sheet goods, The blade/top squareness, no wobble and fence accuracy are the ones I really care about. For sheets, waviness in the top doesn't matter unless it's a really large dimension.We still need to hold the manufacturers to high standards, though. Otherwise, we'll end up with the equivalent of 1970s American made cars- big, inefficient, poorly made and useless in a short time. Sure, they'll still be working but it'll be like what MOPAR stood for (not the good meaning, either).
"I cut this piece four times and it's still too short."
Maybe a factory quality control number or rating might be of use in predicting(sp) what will end up in your shop.
Paul
If the number is explained, I wouldn't have a problem with that. If it just seems arbitrary, I wouldn't be satisfied with it. Even if they showed the allowable specs and showed that it was within those, I'm OK as long as the tolerances are reasonably tight.
"I cut this piece four times and it's still too short."
Well I guess in that case either dump the number or tell me what they mean and what is a good number and what is not.
Right now I look at something that is .012 out and say that being as I am lucky if I can keep my wood to .0625 or on a very good day .0313 I don't have any idea why being off by about 1/3 to 1/5 of that would matter to anyone much less be a reason to criticize a tool.
So either explain this in a review, or quit giving numbers that no one understands as it is not helping any thing, when I look at a number and think that it is pointless i question the whole review. And no I am not saying that .012 is acceptable I am saying that based on all that I know it seams to be to be well with in acceptable but that I really have no way of knowing. And unless I am mistaken the point of the review is to give me a way of deciding if I want to buy the tool. So telling me something that means nothing to me is a waste of time.
Doug Meyer
One thing I would like to see is the results of a series of cuts and add up the total deviation from whatever they're looking for. As an example, if 10 pieces of 1" stock (actual or 3/4", I really don't care) are cut on a table saw, bandsaw, jointer or whatever is supposed to be at a specified angle, stacking all of the pieces and lining the cut faces up vertically should prove to be at that angle. If it's off, how much is it off? That would be a useful test. If they're supposed to be at 90 degrees and they end up leaning 3 degrees one way or the other, that's a big difference. Going 4 places to the right of the decimal is ridiculous unless it's a tool and die maker, machinist or someone who needs to see that kind of accuracy to keep breathing. Keeping it to a few thousandths is fine unless a lot of pieces will be assembled in a way that "tolerance stacking" will cause a problem. That way, cutting 90 strips of wood that were cut to -.5 degrees and gluing the edges together won't give you a gentle curve that causes the last piece to be 45 degrees from the first.
"I cut this piece four times and it's still too short."
At some point wood movement comes into play before that last thousants that your table isn't flat. but I do agree that there needs to be some kind of referance standard that manufactures have to abide by.
Paul
If the weather changes quickly, it can happen before you even get a chance to do the assembly and glue-up (ask me how I know this). The only time I made cabinets from melamine covered hot dog wood (particle board), I cut me dados on a Monday and wanted to do the glue-up ASAP and Wednesday was the day. Nothing fit. Never again with PB. Sharp stick in the eye, anything else but not that garbage. Angles shouldn't change unless the board cups but widths and thicknesses can definitely do some dancing.
"I cut this piece four times and it's still too short."
This forum post is now archived. Commenting has been disabled