Squaring up raw stock

Discussion on all milling machines vertical & horizontal, including but not limited to Bridgeports, Hardinge, South Bend, Clausing, Van Norman, including imports.

Moderators: GlennW, Harold_V

User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Squaring up raw stock

Post by rmac »

What do you guys think of the method shown here:

https://www.youtube.com/watch?v=ZghLslZZ6sg

You still need a good vise, but it looks to me like it's more efficient than the method (discussed earlier in this thread) that you see in all the textbooks.

-- Russ
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: Squaring up raw stock

Post by pete »

Squaring up raw stock? I use the traditional method shown in the text book on page 1. Not to be pedantic, but it's also just how square is good enough? All this is well outside the question, but it is highly relevant to what was being asked because it's also got a lot to do with the size and rigidity of the machine being used and how true it's movements and head alignments are. And as already mentioned, just how accurately made and rigid is the vise. I'd never argue Joe Pie's methods as he's probably forgotten more than I'll ever know. But his procedure in that video does assume quite a bit. Two of them being that the Y axis is in fact square and exactly 90 degrees in it's travel to the X axis and there's zero deflection in the vise. I can guarantee the Y axis isn't 100% square on any of our machines if your measuring equipment is accurate enough since nothing mechanical is ever perfect. Whatever imperfections are there may or may not be enough to matter. But unless you check there's no way to know just how much is there.And since smaller or off shore machines are so common in home shops, the smaller and/or less you paid for the machine and it's vise the higher the probability is of them being further from that perfection. Even the parallels used are suspect unless you've personally verified and checked there good enough. I've bought one set that sure isn't. The Moore Tools book Foundations of Mechanical Accuracy graphically shows there's 6 axes of possible inaccuracy on each single axis of the machine. It can be out in one or more likely various amounts and in combinations of all 6. That also holds true for the vise and the surface it's fixed to, as well as normal and maybe unavoidable vise deflection due to tightening forces. Add to that there's possible cutting tool, head and spindle deflection under those cutting forces.

I started out on my BP clone with a couple of 4" off shore Kurt type vises. There fixed jaw deflection under even moderate closing pressures made them imo just about useless. But even a real 80+ lb Kurt is going to deflect a measurable amount if the closing pressure is high enough. This video is the best example I know of that clearly demonstrates jaw deflection. https://www.youtube.com/watch?v=ckknLkwx9Sw and what can increase or decrease it. I bought both the Moore books Stefan mentions long before he posted that video, but I'd very much agree with his closing comments. Unfortunately neither of those books are exactly cheap. The method shown in that Joe Pie video would be quicker, I'd have to say the textbook method is probably more accurate since your not using the X,Y axis combination. With the way Joe's doing it if vise deflection happens or there's inaccuracy between the X,Y axis you could end up with a slightly parallelogram shaped part instead of the exact square you were expecting. To detect it might take a surface plate and 10ths or better indicator, but it will be there. We tend to assume and I used to as well that our machines are all built with perfect geometry and they don't flex. It just ain't so.
Last edited by pete on Fri Dec 04, 2020 11:55 pm, edited 1 time in total.
atunguyd
Posts: 199
Joined: Tue Oct 20, 2009 9:39 pm
Location: Durban South Africa

Re: Squaring up raw stock

Post by atunguyd »

Here's a great video on squaring up stock

https://youtu.be/tW8HNAlUXxU

Sent from my SM-N975F using Tapatalk

User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Squaring up raw stock

Post by rmac »

Thanks, Pete, for shedding some light on all that. I polluted this forum with a bunch of rookie questions eight or ten years ago and got lots of really good information from all you guys (Harold V in particular). I wound up spending more time on my day job than in the shop for much of the time since then, but now I'm retired with a lot more time to play with my tools.

For whatever reason, I have been pretty successful with the lathe, but I'm still having trouble doing anything halfway accurate on the mill, especially in steel. It's probably a combination of still not knowing what I'm doing, maybe having unrealistic expectations, and having to deal with relatively lightweight equipment.

As an example, I discovered just the other day that the "fixed" jaw on my homemade (by my dad) screwless vise will deflect a HUGE amount if you just look at it sideways. That explains a lot of problems right there, and it's something I never thought to check. Same with the squareness of the mill axes. I've never checked that, or even thought to do so. I'm looking at getting a better vise, and in the meantime have reverted to clamping work the directly to the table or to a nice beefy angle plate.

Anyway, your "you can't assume nothin" advise is spot on, and it's good to be reminded of that every once in a while.

-- Russ
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: Squaring up raw stock

Post by pete »

Thanks and happy someone got something from that Russ. Fwiw I've been there done that with the light weight equipment and unrealistic expectations. The expectation part is a good thing imo, you can't advance to doing better if you don't have that. :-)
User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Squaring up raw stock

Post by rmac »

I checked the squareness of my mill table today. I started by indicating the head of a good combination square so it was parallel to the Y axis motion, and then sweeping the indicator in the X direction along the blade of the square. Turns out the table is out of whack by a little more than .001" per inch. I'm not exactly thrilled by this result, but it's definitely a good thing to be aware of. (See "unrealistic expectations", above.)

-- Russ
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: Squaring up raw stock

Post by pete »

Yeah that's usually the unwelcome result when you start checking things we assume are correct. There's also been the odd time I've been pleasantly surprised, so I guess it evens out. But knowing for sure it's out gives you a base line to work from. If the part accuracy demands are high enough, it's now possible to shim the part or even retram the vise to account for that inaccuracy when using the Y axis and do work to a higher accuracy level than the machine is capable of producing on it's own. Congratulations, there's not many who have made that check and the knowledge will stick with you. So that's an advancement right there. :-)
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: Squaring up raw stock

Post by Harold_V »

For starters, it might be a good idea to consider that the square you used isn't. A combination square is even more suspect, even if it's made by a well respected tool maker. You would be well served to determine if the square is, indeed, square----- before making judgments about the machine.

There's a little one needs to know when it comes to squaring the head of a mill. That's especially true if the mill in question has a column, be it round or rectangular, that bolts to the base (like a mill/drill). Might not hurt to make mention of the mill you use, and I apologize if you have done so and I've missed it.

A question. Is the saddle gib tight? Is the machine worn, or is it reasonably snug full travel? If your machine is worn, loose mid-travel, the saddle may be swiveling on the knee. That condition tends to be reversing, shifting one way, then the other when the feed direction is reversed.

Mean time, assuming your machine does not meet the squareness you prefer, take note that the dodges mentioned will serve no useful purpose. You can compromise your setup in any way imaginable, but if there's error in squareness, it will still be there. Altering how the vise is oriented, as an example, won't change that in any way. It's like having a rotab with the table surface not parallel with the base, with the error at the juncture of the table with the base. Anything you do to correct the error by way of altering the table surface will just move it elsewhere. The error can be corrected ONLY by removing the error where it exists.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Squaring up raw stock

Post by rmac »

For starters, it might be a good idea to consider that the square you used isn't. A combination square is even more suspect, even if it's made by a well respected tool maker. You would be well served to determine if the square is, indeed, square----- before making judgments about the machine.
I'm using a Starrett combination square in pristine condition. It matches a second Starrett square as close as I can tell, and passes the standard "scribe a line, flip the square over, check against the line" test, again as close as I can tell. It may not be absolutely perfect (as Pete noted earlier, nothing ever is), but I'm pretty sure it isn't out nearly as far as the mill table appears to be.
There's a little one needs to know when it comes to squaring the head of a mill. That's especially true if the mill in question has a column, be it round or rectangular, that bolts to the base (like a mill/drill). Might not hurt to make mention of the mill you use, and I apologize if you have done so and I've missed it.
It's an Enco Model 91000 mill/drill, with a round column (yuck) that bolts to the base. To be clear, what I'm worried about at the moment is just the X-Y travel on the table, nothing to do with the head. Having said that, the previous owner trammed it with shims between the column and the base so that the spindle centerline is perpendicular to the table. I've checked that and it's okay, if not great. Whether the spindle centerline is parallel to the column is anybody's guess, but I don't think there's much that could be done about that anyway.
A question. Is the saddle gib tight? Is the machine worn, or is it reasonably snug full travel? If your machine is worn, loose mid-travel, the saddle may be swiveling on the knee. That condition tends to be reversing, shifting one way, then the other when the feed direction is reversed.
The knee. How I wish it had a knee! Anyway, the saddle is not loose, and the error does not change with the Y axis feed direction.
Mean time, assuming your machine does not meet the squareness you prefer, take note that the dodges mentioned will serve no useful purpose. You can compromise your setup in any way imaginable, but if there's error in squareness, it will still be there. Altering how the vise is oriented, as an example, won't change that in any way.
I understand and agree. In theory I suppose you could sort of work around the problem by retramming the vise (or the work itself if not in a vise) every time you wanted to switch between X motion and Y motion on the table. But that would A) be prohibitively tedious, B) introduce all kinds of new sources for error that would be worse than the out-of-square table was in the first place.

Bottom line here is now that I know what's going on, at least it won't be so frustrating when parts don't turn out perfect.

-- Russ
User avatar
Harold_V
Posts: 20231
Joined: Fri Dec 20, 2002 11:02 pm
Location: Onalaska, WA USA

Re: Squaring up raw stock

Post by Harold_V »

Sounds like you've done your homework well, and understand the problem perfectly.

I made mention of the bolted column of mill/drill machines for good reason, and you have addressed it here. I understand that it has nothing to do with the problem you reported. It is not uncommon for owners of such machines to eliminate irregularities in the relationship between the spindle and the table by shimming the base of the column. For a round column machine, where one loses registration when the head is moved, that's not an objectionable way to dial in the head (I don't like the word *tram* for this purpose. Trammels have nothing to do with this operation). However, on machines with ways (instead of a round column), that isn't wise, as any lean in the column translates in to lost orientation of the spindle with the work if the head is moved, which defeats one of the benefits of not having a round column.

Pete has it right. There is little in life, if anything, that is perfect. We learn to live with imperfections when they can not be eliminated. And, it's all a matter of degree. I agree that it will be helpful for you to understand that when you are struggling, that the problem may not be you. It's much the same as using calipers for precision measurements, and struggling to achieve proper fits. Pretty hard to do when you start with erroneous measurements, and that's precisely what one gets with calipers, with rare exception.

H
Wise people talk because they have something to say. Fools talk because they have to say something.
User avatar
rmac
Posts: 787
Joined: Tue Jun 26, 2012 12:48 am
Location: Phoenix, Arizona

Re: Squaring up raw stock

Post by rmac »

As long as we're on the subject of squaring up stock, let's jump from the mill to the lathe. Among the stuff I inherited from my dad's workshop is this block of steel that he had obviously machined on his lathe. Other items that he made exhibit evidence of the same thing. He could have used this technique out of necessity before he had the mill/drill, or afterwards for some other reason.

Anyway, I'm wondering if this is something that's normally done, and if so, how and when and why. I somehow doubt that you could reliably get everything square using a 4 jaw chuck. Maybe clamping the part to an angle plate on a face plate would work better in that regard? And how about the nasty interrupted cuts? I hate making interrupted cuts, but maybe without any good reason.

-- Russ

piece_of_steel.jpg
pete
Posts: 2518
Joined: Tue Feb 10, 2009 6:04 am

Re: Squaring up raw stock

Post by pete »

As far as being sure of your square and the numbers you got, it's a simple geometry problem. If? the fixed jaw of the vise is indicated as square as you can get it (10ths) then check the Y axis from one side of the vise against the squares leg, record the numbers, then flip the square to the opposite side of the vise. Indicate from that side. You mentioned roughly .001" per inch Russ. So as an example let's say your Y axis is creating a slightly greater angle than 90 degrees on that side. Lets say the indicator shows an increase of .003" in 3" of Y axis travel, on the opposite side it would show a .003" decrease. Any numbers greater or lesser than that will demonstrate any inaccuracy in your square or in the set up. But it's fairly easy to prove what your doing to a high degree of accuracy. Simple calculations can allow that test to be done with even an inaccurate square if that inaccuracy is a known and measured amount.

Squaring stock in the lathes head stock? At one time fairly common in a home shop. My Model Engineer magazine collection goes back to 1898. Not that mills were ever common in any home shop prior to the war, but post WW II the UK as a whole was going through a long period of economic hardship that lasted up into the 1960's - maybe early 1970's. Very few with home shops had anything like a small or any mill simply because of there cost and/or unavailability of most smaller milling machines. They were being made, The Tom Senior company was at least one I know of in the UK. There were also more watch maker sized mills that were mostly referred to as jig boring mills. One of my M.E. magazines list a new price for one of those made by BCA in the 1950's and priced at just over 3 times what a then brand new Myford super 7 was selling for. The average home shop back then might have had a lathe, maybe a drill press, or if they were really lucky a shaper. Anyone that actually had a mill in good condition was also making very good money to even afford one. And mill tooling was even more expensive back then as well. Collet prices were listed as just about what we'd be paying today for good industrial quality collets without factoring in inflation changes. About $25-$35 back then for each collet.

So any type of milling was done like your example or bolted to the lathes cross slide, and / or milled using a lathe milling attachment. I've seen some more than complex projects in those magazines and all done without a mill. Some today will say a lathe milling attachment is completely useless. Compared to even the light poorly made off shore one's we have today, then maybe they are. But milling or part squaring like how yours was done still isn't impossible or even that rare. It's just not many do it that way today. Some parts milling depending on there shape are easier to align and machine using the lathe. One accessory called the Keats Angle Plate was designed and sold specifically for that. They also used a lathe face plate far more than most do today. Getting anything reliably square in a 4 jaw or on that face plate is just an alignment issue. One of my old American Machinist books shows jig boring and grinding master gauges used to check the shop floor gauges using nothing more than a sleeve bearing lathe, a face plate, shop made hardened and ground tool maker buttons, a shop made high amplification test indicator, and a very good micrometer. The parts were hand tapped into location and all this was done around 1908 before jig boring machines were even invented. If all you have is a lathe, then you figure out how to get the work done to the required accuracy level. Any probably 9" - 12" swing lathe has a much more rigid spindle to drive tools than even a Bridgeport has with it's over hung cantilevered head mount, ram and knuckle that's required to make them as versatile as they are. It's usually the part holding rigidity and short travels on a lathe that are the main issue and not how the tools are driven.

Myford and a few other lathe manufacturer's purposely designed there lathes with a tee slotted cross slide. It's generally known as a boring table since that's what it is and how it's meant to be used. No one who understands how they work would ever say a Horizontal Boring Machine (HBM) isn't one of the most versatile machine tools around. A lathe with that tee slotted cross slide can work much the same. A lot those HBM's have niceties most lathes don't that would definitely make things easier, faster and better, but between centers boring and quite a bit more are something that can be done very well with a lathe.
Post Reply