Results 1 to 14 of 14
12-25-2007, 09:27 PM #1
Checking Squareness with an Indicator
Does anyone know of a good setup or device to check for squareness on a surface plate with an indicator?
I know that they make magnetic cylindrical squares that give you a reference perpendicular to the surface that you can indicate with a surface gage, but these don't work in every application. For example, in some cases the part is too narrow to stick the magnet to or you want to indicate a slot or keyway.
Has anyone had experience with a Squar OL made by the company that makes the Indi Cal? It looks like a transfer gage but I can't tell exactly how it keeps the part that slides up and down with the indicator from rotating on the post and becoming out of square with the part looking from above.
Here are some links to the Squar OL and some other versions:
Last edited by J-Head; 12-25-2007 at 09:59 PM. Reason: More Info
12-25-2007, 10:50 PM #2
You have to have a reference square to set the indicator. The front edge of the base is rounded. You put the assembly up against the reference square and roll it side to side on the nose to set squareness. Then you can get a reference reading on your part.
There are some surface gages that also have this rounded nose feature for squareness checking.
12-25-2007, 11:06 PM #3
If you are trying to check to see if the bottom is square with the sides, simply check to see if the two sides are parallel if they are, roll the part side to side against the indicator, set the indicator at zero, then check the other side, whatever the reading is divide it by 2 and you have the out of squareness of the part.
I built a square-ol about 20 years ago, use it pretty much daily when squaring stuff on the grinder.
Its a duplicate of the 1st link.
12-26-2007, 12:36 AM #4
Built my own 15 years ago or so. But instead of the rounded surface I used two steel balls seated in holes that protruded from the front edge of the base. That way I had three points of contact with the part being checked and could easily tram down long parts on the surface plate.
12-26-2007, 01:25 AM #5
I made my own a couple of years ago from the plans in the back of Machine Shop Trade Secrets by James Harvey. Here is a link to the one I made. Click on the pictures to get a closer view. I don't use the Starrett Last Word in mine, I use a tenths reading Compac instead. The Last Word was too finnicky with the slightest side load on the probe.
The ones J-Head linked to above are very nice. Hermann Schmidt also makes a very nice (pricey) one.
Depending on how you value your time it may be better to buy one than to make one.
I use mine all the time when checking parts and setups for squareness. Very handy tool.
Some have mentioned using the little ball end of the rod on a surface gage as the curved surface at the base. I have found that method to be too fiddly to setup, inconsistent repeatability, and not very accurate. A method that seems to work well if done very carefully is to put a largis (1" dia) ball bearing in the "crotch" of a surface gage as the curved surface.
For checking the perpendicularity of a slot or keyway using the style of gage that I have and that J-Head gave links to I would simply clamp a qualified parallel in the slot or keyway and gage from that.
12-26-2007, 09:16 PM #6
Ok, so if I am understanding this correctly you set the indicator so the stylus is at about the same radius as the nose of the base and you sweep a reference square in an arc to find zero. This zero setting can than be used to check a part by sweeping the part in a similar arc.
The problem I am having is how do you check a part where the feature you are trying to indicate does not extend down to the surface plate where the base of the Squar OL can get at it. In other words if you had a tall part shaped like an angle plate with a keyway cut in its face how would I indicate the parrallel clamped in the keyway?
Once the indicator is set to zero on the reference, do you move the indicator up the column? I guess, assuming the column is square with the base the indicator would maintain a constant radius with the base.
I still don't understand how the Herman Schmidt style works with the point on the base instead of the radius. How do you set zero on the reference?
Thanks for posting the pictures of your nice looking gage Dave. It wasn't until I saw your pictures that I really began to understand how these work. I like to see home brew tools that others have made to suit their needs and I like making my own tools when I can.
12-26-2007, 09:41 PM #7
bolting or clamping a cylindrical square to the face to want to measure and running a dial over it works well
i think its suburban that makes a magnetic cylindrical square
12-26-2007, 09:59 PM #8
12-26-2007, 11:07 PM #9
How about a merkins surface gage?
12-26-2007, 11:40 PM #10
The Herman Schmidt version wouldn't work when checking a machinist's square, for instance, and in checking flat faces they would need to be as wide as the spacing between the contact points. Wouldn't a radiused face on the base be more versatile? Also I note they have four contact points underneath; using kinematic principles, three would be better - it is the old three-legged stool situation.
- Mike -
12-27-2007, 12:04 AM #11
There are several ways to go about making the measurement you want to make on the part as you descibe it. Since I haven't seen a picture or drawing of the part (hint hint ;-)) I will only suggest one of them but it should work for most anything:
Think horizontal: Lay your part over on to some V-blocks or 1-2-3 blocks or clamp it gently in a small precision vise. Whatever, it doesn't really matter, just get the feature (in this case a slot with a parallel in it) parallel to the surface plate. Check it with an indicator mounted on a surface gage until you are satisfied that it is as exactly parallel to the surface plate as you can make it.
Then with a gage setup like the one I made OR any of the ones linked to above check that the "bottom" surface of your part is at right angles to the surface plate.
The neat thing about the above method is that it will work in most cases even if there is a shoulder at the base or the vertical surface is somehow irregular. Most of the gages I have seen are basically for checking cylinders and rectangular blocks. The trick is to "present" a cylindrical or a flat surface to the gage. Theoretically one could take a piece of metal shaped like a potato, mill a flat surface on it and then somewhere else on the potato shaped lump mill a slot at right angles (normal to) to the milled surface... and the above method should work for measuring it's perpendicularity.
I hope the above desciption is clear... if not lemme know and I will make a diagram.
12-27-2007, 12:12 AM #12
i may be missing something in the question, but clamp (use a mini version of those woodworking clamps with the soft yellow pads) a high quality square that you only use for inspection to the work and indicate the two ends of the blade. they're so many tenths or thou's different over 6 inches, you can work out how much the work is out
12-27-2007, 12:28 AM #13
If the HS gage doesn't come with a convex curved surface as an accessory to replace the contact points that would be one of the first things I would make for it.
Also I note they have four contact points underneath; using kinematic principles, three would be better - it is the old three-legged stool situation.
I have a 12"high x 4"dia cylindrical master as well as various granite angle blocks and steel gage blocks. As long as the masters and the gage base are reasonably flat (easy to check) then the gage will work consistently.
12-27-2007, 12:32 AM #14
Excellent idea. I will remember that one.