Making granite parallels and squares - is any granite sufficient?

[jccaclimber]

I think it does, I noticed differences but I know not the mechanism. A geologist could be of help here. I know one and I will ask in a couple of days.

What magnitude of differences and under what conditions?

 

[woodchuckNJ]

What magnitude of differences and under what conditions?

from the little I know about it, whatever temp a surface plate is trued and calibrated at, is the temp that it's true.

after that, it would depend on the makeup of the granite, which varies.

so I don't think you are going to get a hard / sure answer on that.

 

[jccaclimber]

What magnitude of differences and under what conditions?

from the little I know about it, whatever temp a surface plate is trued and calibrated at, is the temp that it's true.

after that, it would depend on the makeup of the granite, which varies.

so I don't think you are going to get a hard / sure answer on that.

Temperature variation I understand. The question is humidity variation.

 

[Rob F.]

Some of the polishing tables we did were water cooled. Water channels inside the table to keep an even temp. These were two pieces of granite that had the channels milled into the thicker lower half before the top was epoxied on. Aprox 30" dia, 5" lower and 2" upper.

Are you are ever down near Paso Robles or San Luis Obispo?

 

[hanermo]

The surface plate manufacturers state that granite will not absorb any moisture nor distort from it.
I believe them.

Granite is quite bendy and stretchy, but does not want to compress at all.
Thats why granite surface plates need to be thick.
Granite bends real easy, like alu.
It´s easy to modify with abrasives, like sanding, which is not lapping.

A 440 x 600 mm x 100 mm thick surface plate is under 200€ here in EU land (my price).
Engineering companies want 400€+ over here.
It´s accurate to about 2 microns.
I have 3.
All come with computerised error maps - that I believe.

Most-Any marble/granite shop will have a large diamond water cooled cnc slab saw.
They can cut pieces off your piece, for about 10$ each, off-peak.

Easiest and cheapest way to do it.

You could actually net money doing this.
Buy a large granite surface plate from an auction.
Cut it into strips.
The perhaps 200$ 1x2 m slab would probably net 1000 $ when sold on ebay as 20 x 50$ precision granite parallels of 1 m each.
And would be 1/3 the typical cost of same.

Here in Spain, such slabs are not typically available at cheap prices.
They go for junk from closing machine shops, but ordinary people don´t get to bid on them.

 

[jccaclimber]

Some of the polishing tables we did were water cooled. Water channels inside the table to keep an even temp. These were two pieces of granite that had the channels milled into the thicker lower half before the top was epoxied on. Aprox 30" dia, 5" lower and 2" upper.

Are you are ever down near Paso Robles or San Luis Obispo?

No near term plans, but sent you a PM.

 

[EmGo]

No near term plans, but sent you a PM.

Don't go to San Luis Obispo. You won't want to come back. It's what northern California used to be like.

What the heck, jc - grab one of these while you're at it

CNC GRANITE MARBLE QUARTZ STONE CUTTING MACHINE BRIDGE SAW BRAND NEW...

 

[dian]

Unless I'm misreading the standard a grade A parallel is closer to 8 microns/meter than it is 1 to 2, and that will likely exceed my needs. I'm planning on making these in the 6" to 9" range, although I could see making a 12" set if it's convenient at the time.

I have realized that while measuring parallelism in the long direction will be straight forward, measuring it in the short direction is a bit less straightforward. Also likely less important, but I'm open to ideas here.

many parallels have a larger deviation on the narrow side (thickness) than the long side. it depends on what you use them for if this matters.

 

[TGTool]

many parallels have a larger deviation on the narrow side (thickness) than the long side. it depends on what you use them for if this matters.

Can you explain a little more about what you mean there? Narrow edges have not been lapped as flat and parallel as the flat sides? That sounds surprising. That the parallels sag more when supported on their ends and lying flat than they do standing on edge? That should be pretty obvious unless there's some extreme non-linearity in the stone.

 

[dian]

so lets say you put a parallel in a vice for milling. looking down there is the LONG side (parallel to the jaws) and the SHORT side (width). when checking the parallel you usualy pull it under the indicator lenght wise. you get some error. the error can be due to deviations across the SHORT side. sometimes if you manage to keep the indicator exactly in the middle you get a smaller error lenghwise than the max. error across the SHORT side.

i think this has to do with the specifications and what side (lenght) the flatness referrs to, but im not sure. last time i looked the SHORT side flatness seemed to be refered to the lenght of the parallel while the LONG side was referred to the height, which would explain it nicely.

 

[TGTool]

Okay, there might not be a need to hold a tight tolerance on squareness of edge to side. But if the indicator reading changes in the short direction on one edge (across the thickness) it says that the two faces are not parallel planes. Or that there's swarf or a burr underneath somewhere.

In an article several years ago Mcgyver (Michael Ward) pointed out how two parallels could be evaluated by setting them side by side, then using engineer's blue and a flat to mark them up. Reverse position and mark again. If these faces aren't completely parallel to the opposite edges it will show up in an edge-to-edge difference in the blue.

Squareness with the widest face is nice, but not critical. Having each set of opposite edges being absolute parallel planes really is important.

so lets say you put a parallel in a vice for milling. looking down there is the LONG side (parallel to the jaws) and the SHORT side (width). when checking the parallel you usualy pull it under the indicator lenght wise. you get some error. the error can be due to deviations across the SHORT side. sometimes if you manage to keep the indicator exactly in the middle you get a smaller error lenghwise than the max. error across the SHORT side.

i think this has to do with the specifications and what side (lenght) the flatness referrs to, but im not sure. last time i looked the SHORT side flatness seemed to be refered to the lenght of the parallel while the LONG side was referred to the height, which would explain it nicely.

 

[Richard King]

Years ago I had a student who took the class we did in Maryland at the Tuckahoe Steam and Gass Association Machine Shop Museum and one of the students bought a large granite plate at an auction and look it to a Gravestone maker and they cut it to make granite straight-edges. Paolo MD one of our members should be able to tell us who it was. He was tickled at how cheap he got it done for. You may check with Tru-Stone Starrett and see if they have seconds or this CA based Mfg. Sales Items

 

[johansen]

Does granite shift significantly with humidity or water absorption? I was planning to grind with a water based coolant to keep the wheel cool and dust down, but have not considered drying the parts before measuring? I always assumed moisture wasn’t an issue as many surface plate cleaners are water based, but I’ve never given it much thought.

i have played around with water and granite because i built a milling machine from granite, and one problem i ran into was skin oils on the granite surface preventing epoxy from adhering to it. when the granite is wet, you can heat the surface up with a torch and the granite will condense water from the propane torch quickly, but it doesn't take much heat to dry it off completely, which again points towards very thin water absorption layer.

water and oil penetrate into the granite but not very far. you can wipe the surface off, then sand it with 240 grit diamond lapping plate by hand and within a few seconds you have dry granite. but obviously it varies a lot from one type of granite to another.

orbital77's claim that he thinks it does.. is most likely going to be due to the thermal effect of water cooling off the surface. i have read some papers on the creep rates of granite and other papers as well claiming as much as 1% water absorption by weight.. but they were small samples and we're talking soaked in water for a long time, or soaked in water with pressure to force water in.. vs drying out the samples.

I have read posts on this forum about 1-2F temp difference from one side of surface plate vs the other (due to ambient light or other heat sources) being a significant source of error when measuring plates for global flatness.

grinding a granite parallel on a surface grinder will easily cause far greater temp differences.

heating up a thin granite parallel to bake all the water out.. then soak it in water for a day and wipe it dry, my guess is you could easily measure .5% water absorption. if its on only one side of the granite then yeah it might move just a little bit. but if its a thin granite parallel flushed with water in a grinder, its not going to matter since you're going to flip it over repeatedly to grind both sides flat.

this site claims .01 to .8% water absorption.
Quality Control 101: What Is Absorption? | Kafka Granite, LLC

 

[dian]

i have played around with water and granite because i built a milling machine from granite, and one problem i ran into was skin oils on the granite surface preventing epoxy from adhering to it. when the granite is wet, you can heat the surface up with a torch and the granite will condense water from the propane torch quickly, but it doesn't take much heat to dry it off completely, which again points towards very thin water absorption layer.

water and oil penetrate into the granite but not very far. you can wipe the surface off, then sand it with 240 grit diamond lapping plate by hand and within a few seconds you have dry granite. but obviously it varies a lot from one type of granite to another.

orbital77's claim that he thinks it does.. is most likely going to be due to the thermal effect of water cooling off the surface. i have read some papers on the creep rates of granite and other papers as well claiming as much as 1% water absorption by weight.. but they were small samples and we're talking soaked in water for a long time, or soaked in water with pressure to force water in.. vs drying out the samples.

I have read posts on this forum about 1-2F temp difference from one side of surface plate vs the other (due to ambient light or other heat sources) being a significant source of error when measuring plates for global flatness.

grinding a granite parallel on a surface grinder will easily cause far greater temp differences.

heating up a thin granite parallel to bake all the water out.. then soak it in water for a day and wipe it dry, my guess is you could easily measure .5% water absorption. if its on only one side of the granite then yeah it might move just a little bit. but if its a thin granite parallel flushed with water in a grinder, its not going to matter since you're going to flip it over repeatedly to grind both sides flat.

this site claims .01 to .8% water absorption.
Quality Control 101: What Is Absorption? | Kafka Granite, LLC

i would be extremly interested to see some pictures of the granite mill. gantry type? did you drill it?

 

[Toolmaker51]

Additional question, I was planning on slicing this up on a tile saw as I think that would make straighter cuts than on my (very small) diamond blade bandsaw. However, there is a waterjet sitting 20' from the grinder. Any reason not to just slice them up on that? I'll need to clean and square the surfaces regardless, so I don't know that the kerf will be a bigger issue with one method vs. the other.

A tile saw with sufficient blade diameter isn't as efficient compared to band saw, but unless intentional (thickness unknown) the water jet kerf will require a lot of unproductive dressing to square up.
That said, if I was bandsawing, the "back side" surfaces would be inclined; clearly indicating those actually square and parallel.

 

[Toolmaker51]

Can you explain a little more about what you mean there? Narrow edges have not been lapped as flat and parallel as the flat sides? That sounds surprising. That the parallels sag more when supported on their ends and lying flat than they do standing on edge? That should be pretty obvious unless there's some extreme non-linearity in the stone.

Probably why parallels in general have rectangular cross sections; and when we make substantial steel bar parallels, they are drilled and I-beam profile.

 

[michiganbuck]

Sounds like a good/interesting project, and well worth the time because buying they are expensive..

1.5X4X30AA2PL Precision Granite (Laboratory) Grade AA, 2 Face Parallels 1 ½" W x 4" H x 30" L

Granite Tools - 1.5X4X30AA2PL Precision Granite (Laboratory) Grade AA, 2 Face Parallels 1 ½" W x 4" H x 30" L (1.5X4X30AA2PL)

store.gaging.com

 

[jccaclimber]

Seeing as this got a bit more life, a few things we learned in this once more back burner project:
1. Apparently I’m still bad at hitting “send” on DMs after I write them. That or the forum software doesn’t like me.
2. While our gage head is capable, our surface gage and other fixturing, granite table, and other steps in the metrology process are not.
3. Granite cuts just fine on a water jet, though it does leave the expected kerf.
4. I don’t remember how fast we were cutting granite with the diamond wheel, but when making flat cutting stones we could take several thou per pass and it was as if we weren’t even cutting anything. A very different experience than grinding anything conventional. We didn’t find the upper limit because we exceeded the automatic downfeed capability of the surface grinder. At that point it was easier to just use that. As a result, for low quantities the waterjet kerf really wasn’t an issue. If we were making these in real volume we’d probably care.

 

[michiganbuck]

with the diamond wheel, but when making flat cutting stones we could take several thou per pass and it was as if we weren’t even cutting anything.

A true diamond wheel running wet on granite and at a creep feed acts like cutting air.

Best to not take stock on a climb grind because the part might suck the part forward if there is any slack/free play in the feed. Taking a larger amount and creeping through is much better than going back and forth and down feed on both sides of long travel...Must be wet to get the best results.

 

[jccaclimber]

In this case the stone was secured in a vise. Jaw on one side and a piece of aluminum TIG wire on the other, so not much risk of movement. We were feeding at full table speed.
Keep meaning to make a set of clamps like Robin shows in his video, but the vise has done the trick so far. It’s much easier once you have one or two known flat sides.