Mechanic looking clarity.

[EinarsZ]

Hey everyone, I’m looking for some clarity on this topic.


Let’s start with slate precision. What’s the maximum acceptable error? For me, anything more than 0.02 mm/m is already too much. Normally I aim for less than 0.01 mm/m, though I sometimes feel like its not necessary. I’m the only one going this far in 700km radius.


Is this level of precision really needed? Hard to say. If you level with a ball, the margin of error is about 0.06 mm/m (without cloth). And once cloth is on, part of that precision is “canceled out.” In reality, I wonder: could even the best players tell the difference between 0.01 / 0.02 / 0.03 mm/m? My guess — probably not, they’d all feel the same.


Things get more complicated with cheap tables vs. normal tables. For me, slate should always be as close to 0.01 mm/m as possible. But with cheaper tables, a bent frame isn’t even considered a warranty issue — they’re just not built to be flat. Achieving high precision is only possible if installation time is 2–3× longer than on a standard table.


The problem isn’t with the clients — they expect a flat table, and that’s reasonable. The real issue is with sales personnel: they don’t understand the technical requirements, and they don’t communicate them properly. I can’t just leave a table “not flat,” and I can’t call anyone an idiot. The only solution is to charge extra and upgrade the frame to meet the client’s expectations.


For me, the way out is to stop selling cheap tables altogether. If I sell tables myself, I can control expectations. But even then, expensive tables (5–7000€) don’t always mean the parts fit together properly. Becoming a manufacturer is possible, but I can’t make them cheaper than China or with better design than Italy. What I can do is build a much better sport table, because most so-called “sport tables” are really just standard tables with correct dimensions.


Another option is moving to a location with better opportunities — but that’s a different story. If have anything to offer im open.

 

[bradsh98]

Hey everyone, I’m looking for some clarity on this topic.


Let’s start with slate precision. What’s the maximum acceptable error? For me, anything more than 0.02 mm/m is already too much. Normally I aim for less than 0.01 mm/m, though I sometimes feel like its not necessary. I’m the only one going this far in 700km radius.


Is this level of precision really needed? Hard to say. If you level with a ball, the margin of error is about 0.06 mm/m (without cloth). And once cloth is on, part of that precision is “canceled out.” In reality, I wonder: could even the best players tell the difference between 0.01 / 0.02 / 0.03 mm/m? My guess — probably not, they’d all feel the same.


Things get more complicated with cheap tables vs. normal tables. For me, slate should always be as close to 0.01 mm/m as possible. But with cheaper tables, a bent frame isn’t even considered a warranty issue — they’re just not built to be flat. Achieving high precision is only possible if installation time is 2–3× longer than on a standard table.


The problem isn’t with the clients — they expect a flat table, and that’s reasonable. The real issue is with sales personnel: they don’t understand the technical requirements, and they don’t communicate them properly. I can’t just leave a table “not flat,” and I can’t call anyone an idiot. The only solution is to charge extra and upgrade the frame to meet the client’s expectations.


For me, the way out is to stop selling cheap tables altogether. If I sell tables myself, I can control expectations. But even then, expensive tables (5–7000€) don’t always mean the parts fit together properly. Becoming a manufacturer is possible, but I can’t make them cheaper than China or with better design than Italy. What I can do is build a much better sport table, because most so-called “sport tables” are really just standard tables with correct dimensions.


Another option is moving to a location with better opportunities — but that’s a different story. If have anything to offer im open.

I suppose the better question is: How are you measuring this level of precision? It is not uncommon for slate to not be perfectly flat, despite being precision ground. This is due to many factors. That said, you can span a straight edge across the entire surface, and take some readings from your levels. However, that does not address the flatness. You can measure flatness, but that requires some precision measuring equipment.

While some of the flatness issues may be corrected, we are often limited by the design of the table. Sure, we could always modify the design of the table. However, that is a manufacturer issue. If you are going to modify the design, you should most certainly charge appropriately, and be certain that the customer is aware of your plan, as well as the anticipated effect on the overall quality improvement. Though, this may still be difficult to measure.

I have often struggled with the same things that you have mentioned... Where do you draw the line? While we may have the ability to greatly improve every table that we work on, with a few simple design changes, is it worth it?? In some cases, maybe. In most cases, the customer should have simply sourced a higher quality table....

For me, I have decided that I will install each table to the highest degree of precision that the design allows for. If the customer is requesting more than that, I am happy to sell them a higher quality table.


In regards to your question of whether or not the best players in the world would notice a difference: watch some of the coverage of the US Open, which just concluded yesterday... I watched lots of bad rolls. While unfortunate to see, at the highest level of competition, it is still part of the game.

Just remember, not all pool table owners will require the highest level of precision. Most are just casual players, and wouldn't even notice if the cloth were upside-down. That doesn't mean that they don't deserve a quality installation, but they definitely don't need anything more than what the table design allows for.

 

[EinarsZ]

his is due to many factors. That said, you can span a straight edge across the entire surface, and take some readings from your levels. However, that does not address the flatness. You can measure flatness, but that requires some precision measuring equipment.

While some of the flatness issues may be corrected, we are often limited by the design of the table. Sure, we could always modify the design of the table. However, that is a manufacturer issue. If you are going to modify the design, you should most certainly charge appropriately, and be certain that the customer is aware of your plan, as well as the anticipated effect on the overall quality improvement. Though, this may still be difficult to measure.

I have often struggled with the same things that you have mentioned... Where do you draw the line? While we may have the ability to greatly improve every table that we work on, with a few simple design changes, is it worth it?? In some cases, maybe. In most cases, the customer should have simply sourced a higher quality table....

For me, I have decided that I will install each table to the highest degree of precision that the design allows for. If the customer is requesting more than that, I am happy to sell them a higher quality table.


In regards to your question of whether or not the best players in the world would notice a difference: watch some of the coverage of the US Open, which just concluded yesterday... I watched lots of bad rolls. While unfortunate to see, at the highest level of competition, it is still part of the game.

Just remember, not all pool table owners will require the highest level of precision. Most are just casual players, and wouldn't even notice if the cloth were upside-down. That doesn't mean that they don't deserve a quality installation, but they definitely don't need anything more than what the table design allows for.

I got bubble level with precision 0.01mm/m and it's also calibrated once a year or when it starts to show wrong.

Asked chatgpt to read this forum and found measurements. He's a answer.

Max allowable error (slate levelness)

Professional / Tournament spec (Diamond, Brunswick GC, 3-piece slate, 1" thick):

Tolerance is typically ±0.005" (≈0.13 mm) over the entire length of the playing surface.

This is about the thickness of a sheet of printer paper. Anything beyond that starts to show in slow-rolling balls.

Many top installers aim for better than spec (±0.002–0.003", ≈0.05–0.08 mm).


Commercial poolroom installs (good but not tournament-prep):

Acceptable up to ±0.010" (≈0.25 mm) across seams or end-to-end.

Still invisible to casual players, but a seasoned player might see rolls near seams.


Cheap tables / home-grade furniture tables (light frames, thin slate, or MDF “slate”):

Realistically you may see ±0.020–0.030" (≈0.5–0.75 mm) even after best effort, because the frame doesn’t stay flat and the slate isn’t precision-ground.

On MDF or 3/4" import slates, seam shifts and sagging can create visible rolls within months.


Looks like my line is slate itself, im not ready and have no tools to change sanding of the slate. I modify table design also if customers agree to pay extra to get best possible out of things they bought. I dont have option to sell better at position i am. Only modify that client has bought to achieve my standards.

Maby that's because have made lot of mistakes myself learning this skill and have reached ~70% knowledge to start compete at manufacturing and probably going to do it as for me its easier they constantly fixing manufacturing errors or figuring out how to make parts fit in clients living rooms. Have few innovative ideas that becoming a blueprints. Will see where it goes. Cant tell nothing more as i need first understand if those innovations somehow fits for pattern. Yes i can not pattern table but new mechanisms inside, maby.

My point of view is that client doesn't deserve table that is not flat or in wrong parameters only because it's cheap(not talking about mdf surfaces). Or because its cheap it's okay that parts won't fit together. In my mind cheaper means made right with not as good materials, less life time. Rubber lifetime reduced.
If we allowed those cheap tables to be unplayable then how we can get new clients for good ones as bad table destroy chance that this person ever going to play it. Who are going to keep doing anything if nothing works out, but if things works out you probably will continue doing it. Sorry but if private person buys table, someone in that building likes billiard and it depends of first table he owns if he would love it or hate it. From this actually beginner also have same needs as pro.

 

[boogieman]

I got bubble level with precision 0.01mm/m and it's also calibrated once a year or when it starts to show wrong.

Asked chatgpt to read this forum and found measurements. He's a answer.

Max allowable error (slate levelness)

Professional / Tournament spec (Diamond, Brunswick GC, 3-piece slate, 1" thick):

Tolerance is typically ±0.005" (≈0.13 mm) over the entire length of the playing surface.

This is about the thickness of a sheet of printer paper. Anything beyond that starts to show in slow-rolling balls.

Many top installers aim for better than spec (±0.002–0.003", ≈0.05–0.08 mm).


Commercial poolroom installs (good but not tournament-prep):

Acceptable up to ±0.010" (≈0.25 mm) across seams or end-to-end.

Still invisible to casual players, but a seasoned player might see rolls near seams.


Cheap tables / home-grade furniture tables (light frames, thin slate, or MDF “slate”):

Realistically you may see ±0.020–0.030" (≈0.5–0.75 mm) even after best effort, because the frame doesn’t stay flat and the slate isn’t precision-ground.

On MDF or 3/4" import slates, seam shifts and sagging can create visible rolls within months.


Looks like my line is slate itself, im not ready and have no tools to change sanding of the slate. I modify table design also if customers agree to pay extra to get best possible out of things they bought. I dont have option to sell better at position i am. Only modify that client has bought to achieve my standards.

Maby that's because have made lot of mistakes myself learning this skill and have reached ~70% knowledge to start compete at manufacturing and probably going to do it as for me its easier they constantly fixing manufacturing errors or figuring out how to make parts fit in clients living rooms. Have few innovative ideas that becoming a blueprints. Will see where it goes. Cant tell nothing more as i need first understand if those innovations somehow fits for pattern. Yes i can not pattern table but new mechanisms inside, maby.

My point of view is that client doesn't deserve table that is not flat or in wrong parameters only because it's cheap(not talking about mdf surfaces). Or because its cheap it's okay that parts won't fit together. In my mind cheaper means made right with not as good materials, less life time. Rubber lifetime reduced.
If we allowed those cheap tables to be unplayable then how we can get new clients for good ones as bad table destroy chance that this person ever going to play it. Who are going to keep doing anything if nothing works out, but if things works out you probably will continue doing it. Sorry but if private person buys table, someone in that building likes billiard and it depends of first table he owns if he would love it or hate it. From this actually beginner also have same needs as pro.

No offense intended but I'm not reading that much AI slop. There is no way to know if it's even correct or just sounding correct. AI constantly gives wrong answers, there's no way I'd trust any of that without verifying.

 

[boogieman]

Hey everyone, I’m looking for some clarity on this topic.


Let’s start with slate precision. What’s the maximum acceptable error? For me, anything more than 0.02 mm/m is already too much. Normally I aim for less than 0.01 mm/m, though I sometimes feel like its not necessary. I’m the only one going this far in 700km radius.


Is this level of precision really needed? Hard to say. If you level with a ball, the margin of error is about 0.06 mm/m (without cloth). And once cloth is on, part of that precision is “canceled out.” In reality, I wonder: could even the best players tell the difference between 0.01 / 0.02 / 0.03 mm/m? My guess — probably not, they’d all feel the same.


Things get more complicated with cheap tables vs. normal tables. For me, slate should always be as close to 0.01 mm/m as possible. But with cheaper tables, a bent frame isn’t even considered a warranty issue — they’re just not built to be flat. Achieving high precision is only possible if installation time is 2–3× longer than on a standard table.


The problem isn’t with the clients — they expect a flat table, and that’s reasonable. The real issue is with sales personnel: they don’t understand the technical requirements, and they don’t communicate them properly. I can’t just leave a table “not flat,” and I can’t call anyone an idiot. The only solution is to charge extra and upgrade the frame to meet the client’s expectations.


For me, the way out is to stop selling cheap tables altogether. If I sell tables myself, I can control expectations. But even then, expensive tables (5–7000€) don’t always mean the parts fit together properly. Becoming a manufacturer is possible, but I can’t make them cheaper than China or with better design than Italy. What I can do is build a much better sport table, because most so-called “sport tables” are really just standard tables with correct dimensions.


Another option is moving to a location with better opportunities — but that’s a different story. If have anything to offer im open.

On my home table I use a Starrett machinist level. It's a 7' Diamond. I shoot for 1 mark off on the width and 2 marks or less on the length. This is probably overkill.

On the Valleys at the pool hall anything 3 marks and under is fair game. They are usually less than that, i.e. more level. The problem is, many of them have broken legs and Valleys can have crowned/swayed slate anyway. I set a 6' level on the cloth length ways and a 2' width ways and the machinist level on top of that. They don't roll off and that's as good as you can expect with tables with broken/stripped feet. A couple even had to have coasters under them since you can't adjust. Not ideal but the owner is cheap and doesn't want to buy a bunch of repair parts.

 

[garczar]

No offense intended but I'm not reading that much AI slop. There is no way to know if it's even correct or just sounding correct. AI constantly gives wrong answers, there's no way I'd trust any of that without verifying.

Exactly what i thought. sounded like 99% of the other AI sourced bs i've seen lately.

 

[boogieman]

Exactly what i thought. sounded like 99% of the other AI sourced bs i've seen lately.

To be fair he said he asked chatGTP. I'm not a Luddite by any means. I like most tech. ChatGTP and "AI" has proven quite often that they can be wrong. I mean, for what it is, it's a neat tech. It's fun to mess with. It can give you an answer almost instantly and the answer sounds good. But sounding good and trusting a model that just basically says whatever, be it right or wrong isn't the way to go on technical stuff or anything that could cost you if you just believed a lie.

The old saying is "Trust but verify." I don't even have the trust part with "AI."