Aluminum for straight edge?
- Thread starter kooshball
- Start date
- Status
FWIW, a box level with a ground face (to within 0.0005") is under $50 for a 4ft level and under $100 for an up to 8ft level, so they are not impractical investments for many -- the longer levels can be very useful if one needs to adjust bandsaw wheels for coplanarity, otherwise a 6ft level is perfect for most jointers (which typically have 5-7ft wide beds).
The Master Bar approach, and I am aware of it, is fine for those who can not (or will not) make that investment and would rather invest the added time to create one -- that said, it is a somewhat awkward way of characterizing the overall flatness of a table because it is only testing at specific points, which can create some confusion that would be readily alleviated with a good straightedge. However, while MDF is much more stable than wood, the Master a Bar, unlike a well cared for level or straightedge, will need to be checked for calibration before each use because MDF can still distort if it is unevenly exposed to moisture while in storage (such as if one end is closer to a damp floor or a colder/warmer exterior wall, etc.). Nonetheless, I am grateful that you thought to mention it in this thread as it may well be a way some members would rather go.
The poor man's way to a straght edge is to take three flat long objects (can be aluminum, steel, wood, etc basically any stable dense material), mark one edge on each (I marked mine A, B, & C), and then match the edges and file, sand, rasp, plane, etc. until each edge gives a light-tight fit to both of the others. In the end, you end up with three. Use two and save one as a check standard. If wood, best to seal them after you are done to ensure they don't warp with humidity changes.
I did this using 3/4" hardwod ply to check the edges on the planks used for my work bench, and then again for my dining room table.
However, this can be a tedious process, and so i did end up buying a good 4' and 6' box level. They are not truly flat. Every box level I have checked is slighly bowed on the machined edge, with most bowed up in the middle with the edge down. Of those i have checked, most in the 4 and 6' range are bowed around .003" minimum. Most levels are made this way on purpose, because the accuracy stated is the accuracy of the level bubble in relation to the two ends, not straighness of the edge. It easier to slightly adjust an end than the entire surface to get it in tolerance. Aluminum ones will flex enough under moderate pressure to fit tight, tho. When I bought mine, there were several levels of each size in the rack, so I pulled them out and checked them against each other until I found the three closest, and bought one of them. (Yeah, i got some strange looks at Lowes, but most of the people there know me well enough that they kind of accept it from me, LOL))
In practice, box levels work fine once you get the eye-ball used to what amount of light needs to show in the center. I have done this by planing board edges, and have found how much I need so that the board edges are light-tight to each other. If you plane the boards to exactly match the level, you will find you have light showing on the ends when you put the boards together.
If you use hand planes to true up your glue edge, the edge does not have to be prefectly straight. As long as they match, you will have a good tight glue joint. Some also want a little light in the middle for tighter end glue as stated above. My process now is to use the level to get the first board as straight as I can. I then adjust the mating piece to get a tight fit.
If you are using the straight edge for set-up, the slight bow of a box level may not be acceptable, but it does work fine for matching matng surfaces.
JMTCW
Go
I did this using 3/4" hardwod ply to check the edges on the planks used for my work bench, and then again for my dining room table.
However, this can be a tedious process, and so i did end up buying a good 4' and 6' box level. They are not truly flat. Every box level I have checked is slighly bowed on the machined edge, with most bowed up in the middle with the edge down. Of those i have checked, most in the 4 and 6' range are bowed around .003" minimum. Most levels are made this way on purpose, because the accuracy stated is the accuracy of the level bubble in relation to the two ends, not straighness of the edge. It easier to slightly adjust an end than the entire surface to get it in tolerance. Aluminum ones will flex enough under moderate pressure to fit tight, tho. When I bought mine, there were several levels of each size in the rack, so I pulled them out and checked them against each other until I found the three closest, and bought one of them. (Yeah, i got some strange looks at Lowes, but most of the people there know me well enough that they kind of accept it from me, LOL))
In practice, box levels work fine once you get the eye-ball used to what amount of light needs to show in the center. I have done this by planing board edges, and have found how much I need so that the board edges are light-tight to each other. If you plane the boards to exactly match the level, you will find you have light showing on the ends when you put the boards together.
If you use hand planes to true up your glue edge, the edge does not have to be prefectly straight. As long as they match, you will have a good tight glue joint. Some also want a little light in the middle for tighter end glue as stated above. My process now is to use the level to get the first board as straight as I can. I then adjust the mating piece to get a tight fit.
If you are using the straight edge for set-up, the slight bow of a box level may not be acceptable, but it does work fine for matching matng surfaces.
JMTCW
Go
Mark,
I can say that none of my ground box levels are arched as described as if two are held against one another there is no light between them (and even a 0.001" gap allows plenty of light if you are accustomed to looking for it). The same was true for a number of other box levels that I also checked in the stores at the time several years ago. I'm sure there are exceptions, but it was something that I was specifically looking out for at the time I purchased them.
I can say that none of my ground box levels are arched as described as if two are held against one another there is no light between them (and even a 0.001" gap allows plenty of light if you are accustomed to looking for it). The same was true for a number of other box levels that I also checked in the stores at the time several years ago. I'm sure there are exceptions, but it was something that I was specifically looking out for at the time I purchased them.
Mark
Both my aluminum levels are ground I beam levels and as such do not have a bow in the middle as you referenced. They are as straight as straight can be for the length. I would classify both far superior to HF or any other el-cheapo brand level at your big box stores. I also have noticed the empire levels sold at home depot and lowes today are nowhere near the quality they were 10-12 years ago. If I were to buy a aluminum level today for a straight edge I think it would be a Stabilla (sp) brand.
Just googled stabila and apparently they don't make an I beam level anymore they replaced it with what they call a R-beam design
Both my aluminum levels are ground I beam levels and as such do not have a bow in the middle as you referenced. They are as straight as straight can be for the length. I would classify both far superior to HF or any other el-cheapo brand level at your big box stores. I also have noticed the empire levels sold at home depot and lowes today are nowhere near the quality they were 10-12 years ago. If I were to buy a aluminum level today for a straight edge I think it would be a Stabilla (sp) brand.
Just googled stabila and apparently they don't make an I beam level anymore they replaced it with what they call a R-beam design
Last edited:
Unfortunately no -it is because we will not buy an american designed and manufacture machine tool any more... (I will put in a plug for poor management) and UAW labor rates as well - but on the whole our group only went on strike once (before I worked there) and I found out that was because their "Brothers" required them to!
If you walk into most tool shops today (especially mold makers) they will have a Devlieg Jig mill. The last one I worked on had two machines facing each other with a 8 foot high table in between for Allison Transmission the boring bars extended 36 inches from the machine and line-bored the transmission case. The set-up of the machines were (obviously) critical. Allison wanted .0001 true position between the two bored holes that were 22 inches apart in a casting!
In machining capability we use to use a factor of ten - if you need .001 Position - you have to be capable of maintaining .0001 True position statistically. So to maintain .0001 we needed to show capability over 6 feet of holding .00001 tru position, while cutting a ~6- 8" 150mm - 200mm hole!
Let's see somebody in China or India do that!
Just a note on Empire levels and squares; they have 2 levels of quality and both wind up in box stores, TRU BLU are the only ones to buy, the vial liquid is blue on their adjustable squares which are normally NOT, read the print on the tru blu; within .001 to .002 full length, levels are way better than the yellow vials. NO they are not Starrett,nor Mitutoyo,B&S, etc, however we are working with wood, a material that will not be what is used to be 5 mins after you machine it.
Thats because everyone wants something for nothing anymore hank.
Half the post on this site starts with " where can I find the cheapest" instead of where is the best quality.
Maybe I just notice it more then the rest of you after being self-employed 95% of my adult life.
I've always said if everyone would pay for products and services like they want to be paid we wouldn't have people working for poverty wages in this country.
The "10x rule" is pretty much universal amongst standards.
Even in electronics calibration one ideally wants the calibration source to be at least ten times more accurate than the device under test. It creates a bit of the chicken and egg problem when it then comes to calibrating the most accurate standards as they won't have a calibrator ten times more accurate to work against. It is absolutely amazing how well good test equipment maintain their calibration even 20, 30, or more years after their initial calibration.
All this talk about .0000x inches (do I have enough zeroes there?)......all this time I thought 1/64" was close enough for woodworking. As far as that goes, 1/32. As far as machinery, I've never checked the flatness of any of my machines -- and if it wasn't perfectly flat, what could I do about it? I question whether there are even any cast iron tops specs of .003" in flatness? I'll believe it when I see it!.
not everyone can afford the best and sometimes the cheapest is better than nothing at all.
Measure it with a micrometer, mark it with a piece of chalk, and cut it with an axe. Precision work, because we measured to the thousandth! That's what this thread reminds me of. Wood is wood, it's going to move. I know my jointer has a low spot in one of the tables, despite being a ground surface, so even cast iron moves.
You've been given some good advice. Buy a quality ground aluminum level you've got a straight edge sufficient for anything in a woodworking environment and a multi-purpose tool that can be used for more then one task, instead of buying a dedicated tool. Its called maximizing your dollars that you can't afford to spend.
I don't have any issue with using a ground level for a setup tool. I admit it would be better than my HF level. Most recently I was using my HF level for cutting up drywall. It was plenty straight enough for that.
Which is an introduction into my main reason for this reply. We need to think in terms of necessary accuracy, not some uniform standard we apply to everything. For drywall, most of my cuts are within about 1/4 inch of what they need to be except around electrical boxes where I strive for less than 1/8. That means, for me, hanging it with an undersized hole and trimming in place. But when I do woodworking, I can't stand that sort of crude fit-up and have to use different techniques - that would be too time consuming for hanging drywall.
The primary issue for a lot of my woodworking is fitup. Fitup with mating pieces of the project or fitup with where the project goes. For instance, I also made little insulated access doors for attic access in the room I'm working on. The necessary accuracy for the door jamb in the hole was supposed to be 1/2 inch (my choice) but ended up more like the typical 1 inch for house doors (due to an error I made). But the fitup of the door within the jamb is much tighter with a desired nickel width clearance all the way around. To avoid a lot of work, the door skins had to be less than 1/16 different from the softwood inner frame they mate to. But they could be cut to size after the softwood frame is assembled. For the skins, I needed to cut to within maybe 1/32 with less being preferred.
If we are setting up a jointer, we need the infeed table lower than the outfeed table and the two table parallel. How parallel is a reasonable question. My HF level might not be a good enough standard for making that judgement but I am confident a machined level willl be. My little old INCA is not very long so I don't need a six foot level to set it.
So regardless of what we are doing, we need to think about how the piece will fitup to other things and how precise we need to be. Arbitrary conclusions for setup or dimensioning can lead to an awful lot of extra work and/or wasted effort. One of the nice things about woodworking is that we do not need to be ultra precise all the time. And we need to recognize when higher accuracy is indicated.
Which is an introduction into my main reason for this reply. We need to think in terms of necessary accuracy, not some uniform standard we apply to everything. For drywall, most of my cuts are within about 1/4 inch of what they need to be except around electrical boxes where I strive for less than 1/8. That means, for me, hanging it with an undersized hole and trimming in place. But when I do woodworking, I can't stand that sort of crude fit-up and have to use different techniques - that would be too time consuming for hanging drywall.
The primary issue for a lot of my woodworking is fitup. Fitup with mating pieces of the project or fitup with where the project goes. For instance, I also made little insulated access doors for attic access in the room I'm working on. The necessary accuracy for the door jamb in the hole was supposed to be 1/2 inch (my choice) but ended up more like the typical 1 inch for house doors (due to an error I made). But the fitup of the door within the jamb is much tighter with a desired nickel width clearance all the way around. To avoid a lot of work, the door skins had to be less than 1/16 different from the softwood inner frame they mate to. But they could be cut to size after the softwood frame is assembled. For the skins, I needed to cut to within maybe 1/32 with less being preferred.
If we are setting up a jointer, we need the infeed table lower than the outfeed table and the two table parallel. How parallel is a reasonable question. My HF level might not be a good enough standard for making that judgement but I am confident a machined level willl be. My little old INCA is not very long so I don't need a six foot level to set it.
So regardless of what we are doing, we need to think about how the piece will fitup to other things and how precise we need to be. Arbitrary conclusions for setup or dimensioning can lead to an awful lot of extra work and/or wasted effort. One of the nice things about woodworking is that we do not need to be ultra precise all the time. And we need to recognize when higher accuracy is indicated.
Thanks for the feed back on my take on levels. I am not saying what I found was "as manufactured", but was "as found in the store". For reference, my 6' level is a Swanson, about 6 yrs old, bought locally at Lowes. My 4' is slightly less bowed but about the same considering the shorter length. It is a Empire Blue Max (yellow vial) about 12 years old, bought at HD in FL. To do it jsutice, it was probably 5 years old before I thought of trying using it for a precision straight edge and it has held up well under a lot of abuse while doing carpentry work. My 24" is a Johnson about 6 years old, and appears straight but also has fairly rough machining on the flat edge. It is the only one with a tolerance marked on the level itself, and that is "Accuracy in standard position 0.5mm per 1 meter = 0.029degrees).
So, if buying one, you may want to compare edges in the store to make sure the one you get has not been abused in transit/storage or while on display.
Go
So, if buying one, you may want to compare edges in the store to make sure the one you get has not been abused in transit/storage or while on display.
Go
David
Getting within .003 with wood is crazy. You can not even see light thru that gap. This will make a great straight edge. We're talking about things like a jointer that will vibrate like crazy. If you can check your tools to this tolerance consistently, you are way ahead of most people I know.
Just try not to drop it. Good luck and get back to making wood shorter.
dan
Getting within .003 with wood is crazy. You can not even see light thru that gap. This will make a great straight edge. We're talking about things like a jointer that will vibrate like crazy. If you can check your tools to this tolerance consistently, you are way ahead of most people I know.
Just try not to drop it. Good luck and get back to making wood shorter.
dan
Good points. We're woodworkers not machinists so we can tolerate more wiggle-wobble and human foibles. Are we comparing precision vs. accuracy and using the terms interchangeably?
http://www.mathsisfun.com/accuracy-precision.html
OMG, I'm gonna stop using my planes and ROSs because I've already got "the perfect boards" and they won't move even if I do!
http://www.mathsisfun.com/accuracy-precision.html
OMG, I'm gonna stop using my planes and ROSs because I've already got "the perfect boards" and they won't move even if I do!
Is that .003 accuracy with or without the light surface rust, subsequent removal, and follow-up paste wax deposits?
- Status
Premier Sponsor
Our Sponsors
LATEST FOR SALE LISTINGS
-
-
Graco 395 FinishPro air assisted airless sprayer for sale
- Started by RobH
- Replies: 1
-
-
-