Slate flatness...
- Thread starter EL Picos
- Start date
Tablemaker
Registered
As per BCA specifications:
Playing Bed: The playing surface must be capable, either by its own strength or a combination of its strength and that of the table baseframe, of maintaining an overall flatness within a tolerance of .020" lengthwise and .010" across the width. Further, this surface should have an additional deflection not to exceed .030" when loaded with a concentrated static force of 200 pounds at its center. If more than one slab is employed, the slab joints must be in the same plane within .005" after leveling and shimming. The bed must be covered with a fabric, the major portion of which is made of wool, with proper tension to avoid unwanted ball roll-off. Commercial tables must have a 1" - 3 piece set of slate with a wooden frame minimum 3/4" attached to slate. All playing surfaces must be secured to base frame with screws or bolts.
Playing Bed: The playing surface must be capable, either by its own strength or a combination of its strength and that of the table baseframe, of maintaining an overall flatness within a tolerance of .020" lengthwise and .010" across the width. Further, this surface should have an additional deflection not to exceed .030" when loaded with a concentrated static force of 200 pounds at its center. If more than one slab is employed, the slab joints must be in the same plane within .005" after leveling and shimming. The bed must be covered with a fabric, the major portion of which is made of wool, with proper tension to avoid unwanted ball roll-off. Commercial tables must have a 1" - 3 piece set of slate with a wooden frame minimum 3/4" attached to slate. All playing surfaces must be secured to base frame with screws or bolts.
Tablemaker
Registered
You would probably need to contact the companies machining the slates in Italy, Brazil, and China, (or whatever country) to get that answer.
This is an intriguing concept. I'm curious to see the design.
.001" straightness tolerance along 9' of length??? I'm curious what you'll be using for materials to accomplish this, and what the weight will be.
Based on assumptions, I have a few ideas that you may wish to consider..
(9 measuring points per slate) X (3 slates) = 27 points of measurement
27 points of measurement is what many of us would consider the minimum standard for checking level. However, this is through the use of a machinist level. A machinist level is a point-to-point measuring instrument. It checks the level from Point A to Point B. Using a dial indicator is a singular point of measurement, which does not span the same area of a 12" machinist level. Therefore, you would need to reevaluate your measurement points, to ensure that you are spanning the same area. In this case, to match the precision of using a machinist level, you would need to touch at least 54 points with an indicator. Or, 18 points per slate.
According to this video: Brunswick slates are .010" across the entire surface
Keep in mind that the world is made of rubber.
Slate bends/sags, that is why it requires practiced table techs to set up & shim.
Geoff's analyses of level vs SE is a very good reference on the subject.
That is, either method can work, which is more practical, which is more cost effective, etc.
Until the foundry sold out and my patterns were stolen or destroyed, i used to have cast and provide blanks for true, "camelback" straight edges. I have planed and scraped them up to 5' long (other peoples castings at that length) & owned an 8 footer out of Xerox corp until i got too old to lift it, and sold it on.
Your spec of (honest) .001 in 9 ft is attainable, but it won't likely be something you move around by yourself.
If you use a lighter weight tube or truss frame, heat issues can become almost insurmountable. The gage (the SE) will bow up and down depending on local heat sources, and where or how long you hold it in your hands.
That said, you don't actually need the tube or frame to be perfectly straight, so long as you have a reference to set the indicators.
& good insulating handles to scrupulously use for moving it. But it begs the question, what do you have to set the gage to in your shop, and how repeatable is the indication? Do you have an 8' surface plate? (tinch shy of 9' diagonal.) My mind doesn't hold unused specs well anymore, does an 8' grade B surface plate even spec .001" unilateral tolerance corner to corner?
There are machinist (not builder grade) lasers that are used for erecting and scraping machine systems that could work.
Taut wire technology is very old school and sag tables are known. In fact at least until recently they were more accurate for setting up things like (atomic) linear accelerators because dust in the air at distance interfered with the lasers.
It has been reputed that 4' florescent tubes (the old large dia type) are quite straight in machinist terms. I never blued one up to check.
It all sounds tricky, unreliable, and difficult to maintain repeatability compared to a set of qualified levels.
OTOH that is how progress happens. Good luck!
I used to (1980's) go with a table builder to the slate quarries for table slate and saw the processes the whole way through. They honed the matched set on a purpose built machine with a diamond hone in which the saucer shape wheel was slightly wider (larger diameter) than a 9' slate set. The slate was not clamped, and the bed they laid on was occasionally dressed. As one would a surface grinder chuck. Regrettably, i though John would always be around to ask whatever was necessary, and didn't pay attention or ask specs myself at the time.
Of course once you pick your parts off a machine, especially if they were clamped for the process, all bets are off.
smt
Slate bends/sags, that is why it requires practiced table techs to set up & shim.
Geoff's analyses of level vs SE is a very good reference on the subject.
That is, either method can work, which is more practical, which is more cost effective, etc.
Until the foundry sold out and my patterns were stolen or destroyed, i used to have cast and provide blanks for true, "camelback" straight edges. I have planed and scraped them up to 5' long (other peoples castings at that length) & owned an 8 footer out of Xerox corp until i got too old to lift it, and sold it on.
Your spec of (honest) .001 in 9 ft is attainable, but it won't likely be something you move around by yourself.
If you use a lighter weight tube or truss frame, heat issues can become almost insurmountable. The gage (the SE) will bow up and down depending on local heat sources, and where or how long you hold it in your hands.
That said, you don't actually need the tube or frame to be perfectly straight, so long as you have a reference to set the indicators.
& good insulating handles to scrupulously use for moving it. But it begs the question, what do you have to set the gage to in your shop, and how repeatable is the indication? Do you have an 8' surface plate? (tinch shy of 9' diagonal.) My mind doesn't hold unused specs well anymore, does an 8' grade B surface plate even spec .001" unilateral tolerance corner to corner?
There are machinist (not builder grade) lasers that are used for erecting and scraping machine systems that could work.
Taut wire technology is very old school and sag tables are known. In fact at least until recently they were more accurate for setting up things like (atomic) linear accelerators because dust in the air at distance interfered with the lasers.
It has been reputed that 4' florescent tubes (the old large dia type) are quite straight in machinist terms. I never blued one up to check.
It all sounds tricky, unreliable, and difficult to maintain repeatability compared to a set of qualified levels.
OTOH that is how progress happens. Good luck!
I used to (1980's) go with a table builder to the slate quarries for table slate and saw the processes the whole way through. They honed the matched set on a purpose built machine with a diamond hone in which the saucer shape wheel was slightly wider (larger diameter) than a 9' slate set. The slate was not clamped, and the bed they laid on was occasionally dressed. As one would a surface grinder chuck. Regrettably, i though John would always be around to ask whatever was necessary, and didn't pay attention or ask specs myself at the time.
Of course once you pick your parts off a machine, especially if they were clamped for the process, all bets are off.
smt
Last edited:
I have to ask, what were you involved with that you used those long straight edges??
When i started buying rudimentary machine tools in the early 80's, what was available (before ebay and C-list) was either near-new price good machines or high-priced junk, pretty much only from dealers. Or lower priced machines at auction that were worn enough even the dealers did not want them. or the early imports. I learned to scrape starting with small stuff - cross slides and ways on small machines that could be done with parallel & "camelback" straight edges up to about 24". Being young, dumb, and still healthy, i moved on up the curve over time.
The gages (straight edges and scraped or granite references) were always important to me, if they were seen anywhere cheap - auctions, early ebay, etc. If you can scrape ways and have the simple references to test and prove them, you can make any machine as new & precise as the rigidity of the castings allows. Simple digging in the dirt muscle power (or biax), intelligent approach, and insight into current conditions and where you want to lead them efficiently.
After the internet came along, a lot of people on machinist sites got into scraping for the same reason i had. Make cheap worn machines better than new. But there were not a lot of gages and such out there. (at the time, 20+ years ago.) So i made patterns for a 30" camelback, which is just the size that can be machined on a 42" Bridgeport table, and yet it just about long enough to efficiently re-scrape the ways on all the axes of the same machine. originally i sold raw castings; then later, machined, thermal stress relieved castings. The purchaser did any final machining, and scraped them to a surface plate. A cored out light-weight version was developed, and only had the first run of 10 or so cast when the foundry went under and both sets of my patterns disappeared.
The eight footer was honestly because it was cheap, but i did use it to set up machines that benefitted from the length and accuracy.
The last biggish machine i bought, about a dozen years ago, is a 6' planer. Fully functional machine used here quite a bit for metal and for wood projects (can be seen in some of my cue building posts - it is my taper/shaper.) In the "olden daze" planers were kept at least as accurate as surface grinders. Big planers, moving up past, say 40 footers, were accurate over acres. Or a large percentage of a tennis court, anyway.
My little machine from 1920's has some wear. It is within 1/2 a thousandth across the rail, but on the bed, goes from about .001" in 4'; to rapidly increase every further 6" to about .006 - .008" in 80" total travel. The bed is about 90" long; the bed ways are about 11'. So it seemed the 8 foot SE would be useful to rescrape that machine as well.
A few years ago it was forced on me that i could no longer effectively manage the 8' SE by myself. I never had a surface plate over 24 x 36 to easily prove & maintain it. (can be done with levels and shorter SE's, but painfully) The 8 footer had also gotten somewhat more valuable, so was sold to a friend. If the planer is ever to be rescraped (dubious at this point) it will require some purpose built and serious rigging. To remove, support, and flip the table between scraping passes, e.g.
However......
Getting directly back on the subject of level vs straight edge.....
Most really big machines are re-scraped during rebuilding by using a combination of relatively short (4' to 6' "camelback" straight edges), and precision levels like the Starrett 199's. So much less wear and tear physically, and easier to maintain the gages.
probably more than you wanted to know, sorry.
smt
Last edited:
When i started buying rudimentary machine tools in the early 80's, what was available (before ebay and C-list) was either near-new price good machines or high-priced junk, pretty much only from dealers. Or lower priced machines at auction that were worn enough even the dealers did not want them. or the early imports. I learned to scrape starting with small stuff - cross slides and ways on small machines that could be done with parallel & "camelback" straight edges up to about 24". Being young, dumb, and still healthy, i moved on up the curve over time.
The gages (straight edges and scraped or granite references) were always important to me, if they were seen anywhere cheap - auctions, early ebay, etc. If you can scrape ways and have the simple references to test and prove them, you can make any machine as new & precise as the rigidity of the castings allows. Simple digging in the dirt muscle power (or biax), intelligent approach, and insight into current conditions and where you want to lead them efficiently.
After the internet came along, a lot of people on machinist sites got into scraping for the same reason i had. Make cheap worn machines better than new. But there were not a lot of gages and such out there. (at the time, 20+ years ago.) So i made patterns for a 30" camelback, which is just the size that can be machined on a 42" Bridgeport table, and yet it just about long enough to efficiently re-scrape the ways on all the axes of the same machine. originally i sold raw castings; then later, machined, thermal stress relieved castings. The purchaser did any final machining, and scraped them to a surface plate. A cored out light-weight version was developed, and only had the first run of 10 or so cast when the foundry went under and both sets of my patterns disappeared.
The eight footer was honestly because it was cheap, but i did use it to set up machines that benefitted from the length and accuracy.
The last biggish machine i bought, about a dozen years ago, is a 6' planer. Fully functional machine used here quite a bit for metal and for wood projects (can be seen in some of my cue building posts - it is my taper/shaper.) In the "olden daze" planers were kept at least as accurate as surface grinders. Big planers, moving up past, say 40 footers, were accurate over acres. Or a large percentage of a tennis court, anyway.
My little machine from 1920's has some wear. It is within 1/2 a thousandth across the rail, but on the bed, goes from about .001" in 4'; to rapidly increase every further 6" to about .006 - .008" in 80" total travel. The bed is about 90" long; the bed ways are about 11'. So it seemed the 8 foot SE would be useful to rescrape that machine as well.
A few years ago it was forced on me that i could no longer effectively manage the 8' SE by myself. I never had a surface plate over 24 x 36 to easily prove & maintain it. (can be done with levels and shorter SE's, but painfully) The 8 footer had also gotten somewhat more valuable, so was sold to a friend. If the planer is ever to be rescraped (dubious at this point) it will require some purpose built and serious rigging. To remove, support, and flip the table between scraping passes, e.g.
However......
Getting directly back on the subject of level vs straight edge.....
Most really big machines are re-scraped during rebuilding by using a combination of relatively short (4' to 6' "camelback" straight edges), and precision levels like the Starrett 199's. So much less wear and tear physically, and easier to maintain the gages.
probably more than you wanted to know, sorry.
smt
You remind me of my old man, may he rest in piece. I’ve got a shaper attachment for Cincinnati he scraped. Been sitting for years. I need to get rid of it and a lot of other stuff. Thanks for the memory.
Oh, I'm all about jigs, fixtures, gages, etc - seldom saw a gadget that didn't intrigue me at some level.
I believe you can make that work, probably even work "well".
However, when it's all done, how do you prove it?
roll some balls around?
Check a few places with levels?
Did it save time?
Not arguing any of above, just curious how it plays out.
Couple notes, minor book-keeping:
it's differentials. Try it on your 20' surface plate. let the unit thermal stabilize for a few hours and check.
next, hold you hands on top of a long tube for a few minutes, and check again.
Or, classic "gotcha" put one side in the sun (near window) and other shaded by, say some shop equipment.
I realize this is not a factor in a poolroom with the blinds drawn, but other subtle sources of differential heat may be.
Is it a factor for .005" in the area you are working?
Probably not.
Is it a factor if .001" unilateral tolerance is advertised?
Probably.
Does it matter for a pool table?
Actually, that is outside my experience. Truly don't know.
I'm just working to keep you honest on gaging.
Also shorter Moore & Wright that is .00025" (quarter thousandth) per ft.
Both will probably be ambiguous on cloth, on a wood framed table, though.
Have someone walk around the table on even thin concrete and it might be possible to see the bubble move.
Wooden floors, forget about it. So I get where your gage would overcome that limitation.
However, all said and done, it is level that matters for pool ball roll.
Once was co-presenter at a scraping seminar in a plant that had Brunson equipment for erecting large machines.
They gave me a (very) short intro to optic systems. It has occurred to me that a Brunson telescope with optical micrometer would be cool to play with when/after setting up a pool table. It's basically an exceptionally rigid tripod and telescope "transit", the telescope cross hairs can be dialed by .001" & are clear enough to split that difference.
I'm truly interested in learning how your project works out.
Again, best of luck.
smt
I believe you can make that work, probably even work "well".
However, when it's all done, how do you prove it?
roll some balls around?
Check a few places with levels?
Did it save time?
Not arguing any of above, just curious how it plays out.
Couple notes, minor book-keeping:
It's not absolute temperature that wreaks havoc at the tolerances you advertise;
it's differentials. Try it on your 20' surface plate. let the unit thermal stabilize for a few hours and check.
next, hold you hands on top of a long tube for a few minutes, and check again.
Or, classic "gotcha" put one side in the sun (near window) and other shaded by, say some shop equipment.
I realize this is not a factor in a poolroom with the blinds drawn, but other subtle sources of differential heat may be.
Is it a factor for .005" in the area you are working?
Probably not.
Is it a factor if .001" unilateral tolerance is advertised?
Probably.
Does it matter for a pool table?
Actually, that is outside my experience. Truly don't know.
I'm just working to keep you honest on gaging.
One graduation on my 15" long Starrett 199's is .0005"/ft (1/2 thousandth per ft)
Also shorter Moore & Wright that is .00025" (quarter thousandth) per ft.
Both will probably be ambiguous on cloth, on a wood framed table, though.
Have someone walk around the table on even thin concrete and it might be possible to see the bubble move.
Wooden floors, forget about it. So I get where your gage would overcome that limitation.
However, all said and done, it is level that matters for pool ball roll.
Once was co-presenter at a scraping seminar in a plant that had Brunson equipment for erecting large machines.
They gave me a (very) short intro to optic systems. It has occurred to me that a Brunson telescope with optical micrometer would be cool to play with when/after setting up a pool table. It's basically an exceptionally rigid tripod and telescope "transit", the telescope cross hairs can be dialed by .001" & are clear enough to split that difference.
I'm truly interested in learning how your project works out.
Again, best of luck.
smt
Last edited:
When i started buying rudimentary machine tools in the early 80's, what was available (before ebay and C-list) was either near-new price good machines or high-priced junk, pretty much only from dealers. Or lower priced machines at auction that were worn enough even the dealers did not want them. or the early imports. I learned to scrape starting with small stuff - cross slides and ways on small machines that could be done with parallel & "camelback" straight edges up to about 24". Being young, dumb, and still healthy, i moved on up the curve over time.
The gages (straight edges and scraped or granite references) were always important to me, if they were seen anywhere cheap - auctions, early ebay, etc. If you can scrape ways and have the simple references to test and prove them, you can make any machine as new & precise as the rigidity of the castings allows. Simple digging in the dirt muscle power (or biax), intelligent approach, and insight into current conditions and where you want to lead them efficiently.
After the internet came along, a lot of people on machinist sites got into scraping for the same reason i had. Make cheap worn machines better than new. But there were not a lot of gages and such out there. (at the time, 20+ years ago.) So i made patterns for a 30" camelback, which is just the size that can be machined on a 42" Bridgeport table, and yet it just about long enough to efficiently re-scrape the ways on all the axes of the same machine. originally i sold raw castings; then later, machined, thermal stress relieved castings. The purchaser did any final machining, and scraped them to a surface plate. A cored out light-weight version was developed, and only had the first run of 10 or so cast when the foundry went under and both sets of my patterns disappeared.
The eight footer was honestly because it was cheap, but i did use it to set up machines that benefitted from the length and accuracy.
The last biggish machine i bought, about a dozen years ago, is a 6' planer. Fully functional machine used here quite a bit for metal and for wood projects (can be seen in some of my cue building posts - it is my taper/shaper.) In the "olden daze" planers were kept at least as accurate as surface grinders. Big planers, moving up past, say 40 footers, were accurate over acres. Or a large percentage of a tennis court, anyway.
My little machine from 1920's has some wear. It is within 1/2 a thousandth across the rail, but on the bed, goes from about .001" in 4'; to rapidly increase every further 6" to about .006 - .008" in 80" total travel. The bed is about 90" long; the bed ways are about 11'. So it seemed the 8 foot SE would be useful to rescrape that machine as well.
A few years ago it was forced on me that i could no longer effectively manage the 8' SE by myself. I never had a surface plate over 24 x 36 to easily prove & maintain it. (can be done with levels and shorter SE's, but painfully) The 8 footer had also gotten somewhat more valuable, so was sold to a friend. If the planer is ever to be rescraped (dubious at this point) it will require some purpose built and serious rigging. To remove, support, and flip the table between scraping passes, e.g.
However......
Getting directly back on the subject of level vs straight edge.....
Most really big machines are re-scraped during rebuilding by using a combination of relatively short (4' to 6' "camelback" straight edges), and precision levels like the Starrett 199's. So much less wear and tear physically, and easier to maintain the gages.
probably more than you wanted to know, sorry.
smt
Absolutely not more than I wanted to know, no apology needed, instead I will thank you for a great answer. I love machines and wish I had more of them. anytime you wanna discuss that stuff I'm all ears. Were you like a commercial scraper?? The first time (about 30 years ago) I heard someone tell me how ways and beds were leveled I thought the guy was on crack.
No. I have done commercial stuff with scraping, and in the past scraped some of the products made & sold over the years. Steel fixture sleds for small wood shapers was one product, some 35 yrs ago or so. Then i finally got a surface grinder, and scraped that back into shape to use instead. Still scrape some niche products, but not very often. (still scrape my own machines, as necessary) Been co-presenter for a couple seminars on scraping, but my background is self taught, so need someone else with gravitas when a company is paying.
Mostly it was buy woodwhacking machines for my regular work, and rebuild them. Tenoners, (automatic) lumber jointers, mortisers, even widebelt sanders have way & table systems that work best when they are "right" and some old machines are really not right. Banana shaped ways, and saggy worn out tables. Then started buying the machine tools to work on the wood machinery. Then machine tools to work on other machine tools, etc. So had to keep machining and scraping.
It's not as impressive as it might sound, nothing here is huge & I'm not good about painting things after they work efficiently and "nicely", which is where my effort goes. Nonetheless it's been an enjoyable, if not overly remunerative life. per your note, it is funny, innit? Very few people think how basic machines were made and in most cases still are fit out. If they think at all, they imagine it was made on a bigger machine. To some extent. But then the mind boggle part is that up until linear ways (ball way systems, e.g., etc) were developed, all precision machines were hand carved to attain that precision. At some point there is no bigger machine, and it is back to hand carving. Linear way machines still have the pads (for the location of the linear ways, and for the trucks on the traveling part) hand scrape fitted. And big enough grinders to grind things like linear ways, and the ways of other machines, are handscraped themselves.
That's probably enough on a billiards forum, though.
smt
I'm quoting Tablemaker only to have the numbers restated.
Admittedly you guys are way more versed than I am but getting back to the original question, I have a simple comment.
In my experience with measuring stuff, particularly torque tool calibration equipment, the transducer used to Certify the tool needs to be 5X more accurate/repeatable than the tool being calibrated.
It seems to me (without too much thought) if you stayed with that simple formula, your device would do the job even if slates were machined to a tighter tolerances than stated above. That is if my assumption is correct, that the OP is building a device to level a table to BCA specs or better and not for measuring for flatness of the slates.
No offense to anyone, but, I work on the tables and there is nothing to "fix" The game is not perfect, nor the players and NONE of the tables will ever be..... The leveling systems in the 3 major players tables, guess what; THEY ALL WORK. Are they perfect, not a darn one of em, but, they all work their own way and require skill to make them work... Getting into any finer detail is just a waste of time. Just my $0.02..... Carry on!
TFT
The
TFT
The
For the majority of the 3-piece slate tables that I work on, I spend NO LESS than an hour to level the slates. In many cases, I will spend 2 hours+, on a beat up old Brunswick. This is while using (7) 12" machinist levels. If anyone can get a table more level than I can, I would love to see it... If they could do it faster, I'd pay to learn how.
I'm all for a new tool to level slates. Although, I am a bit skeptical of the proposed tolerances mentioned here.. I know from my own personal experience that the general tolerance for leveling a pool table is to be no further off than one set of graduations on a Starrett 98-12 level. Even at that, depending on how the surrounding area reads, you could see a slight ball drift.
In addition, the type/grade of cloth installed will either hide or magnify deficiencies in the level of the table.
I know you know your stuff. I would love to train with you and 7 starretts! We all have tricks of the trade to share.
TFT
ps. the point of my original post was made a bit more solid by your input Bradsh98
I am saying if it aint broke, dont go broke trying to fix it to all these people who think pool is so broken ! TFT
I am saying if it aint broke, dont go broke trying to fix it to all these people who think pool is so broken !
Practically speaking, & thinking as a woodworker i agree.
A wood frame is never going to be perfect over time, but it has some excellent perhaps overlooked benefits in the modern era, including decent stability with changing humidity within typical values in a human living space; and excellent thermal stability across the same.
Slate is quite stable, although it needs support and shimming in the thickness typically used for pool tables. Accepting those 2 materials, there does not seem to be much in terms of "adjuster" systems that would excel shims and levels for leveling.
Plus there are what, a billion
legacy tables out there? Playing fine?Thinking technically, though, Rasson is probably the way things will trend in the future. Commodity level tables with ease of set up, steel tube frames, built in leveling systems at similar cost to Diamond. It is even easy these days to start to imagine an automatic leveling system. Perhaps for only a few $K upcharge.
Slate is probably still the cheapest bed material. Especially as 3 pc slate sets.
OTOH, glass is more or less easy to make flat, weighs less than 2/3 wt of slate, and is somewhat stiffer & stronger (many x stronger for tempered) than most slate. Not thinking of "designer" tables which already include some. Thinking of using materials to build a commodity table, even "barboxes" the way Diamonds are now.
We are probably at a cusp where the old materials vs new are just about a toss up cost wise, but it could flip quickly. Wood sure has gotten expensive lately. And scarcer.
One issue with steel or alloy tube frame is that it is not thermally stable. That may or may not be a factor for tables, in "typical" living spaces.
My sense is that slate is more thermally stable than glass (can't find data" but at those levels probably irrelevent. Rail systems based on extrusions could be engineered with elastomers & wood to play as well as anything, and the esthetics easily tailored as well.
smt