Scraping perpendicular surfaces questions

Hi, I'm building a little fixed gantry cnc mill and I struggled a little bit with scraping surfaces for rails installation. I understand how to scrape them flat but I can't find a way to scrape them perpendicular.

Here's a picture of what I want to achieve.



My tools are scraper, granite plate, dial indicator and try square.I don't have access to big machines like mill or lathe.

The problem is that it's not continuous surfaces but just stripes and my square is smaller than gap between them, so I can't figure out how to check them for squareness. Is it even possible in this case? If possible what would be the easiest way to do so in terms of required tools?

Another option to consider is a replicated joint.

Scrape your rail mounting surfaces flat (single plane).
For the column mounts you either have a series of set screws or a firm but compressible material (copper wire is one option) around the mounting perimeter. You use the mounting bolts and the set screws in opposition and precisely align the column / axes.
When all aligned you inject epoxy (making a 'replicated joint').
Specialist epoxies for just this purpose exist.
If you want to take it apart in future, wax one side of the joint well. Pin the joint for alignment.

I haven't given all the details. There are guides and videos on how to do this.

Advantage is that you don't have to scrape the parts for perpendicularity. You are handling alignment at the same time so takes care of a few issues.

He still needs a acurate enough 90°reference

Peter

You may be able to get the accuracy you need by using a right angle artifact (RAA, steel or granite) to go from the vertical to the horizontal plane. This could work for you: China High Precision Granite Right Angle Ruler - China Granite Right Angle Ruler, Granite Square L Shaped Tri Ruler but you might find better choices with additional searching.

Even if the RAA isn't tall enough on its own, you can scrape some spacing parallels to raise it up high enough to engage the upper rail mounting strips.

Your "try square" is almost certainly not accurate enough to make a precision right angle. The RAA I linked is doubtless low grade, but still should be closer than what you have now. If you buy two you can check your current square and other features by doing A:B:C: comparisons (angle 1 to angle 2, angle 2 to angle 3, angle 3 to angle 1 - if they all mate correctly, each one is accurate).

The OP may want to add a sensitive level to his list of tools. One with a 90deg face would be even better.

L7

A rounded nose surface gauge can be fabled-up to transfer the check of a good square block as reference to check your part.
Just a block with two parallel sides and a little out of square can be the master, you just have to figure in the out of a square condition of the master. Perhaps the master is .005 mm off so you add/subtract that...yes likely need a true surface to do your gauging on.

A surface gauge that has a V notch at one end can be used with putting a ball or good round thing in the v like the OD of a bearing.

Vintage Starrett Surface Gage #257-A, Original Box | eBay

This so from one point to another you can make a right angle. yes, likely still need a precision straight edge to go from one point/place to another.

If you're spindle carrier and it's rails were finished you could (tram) the upright to the finished bottom rails with your indicator

No new tools needed

Bravo and good luck on this very ambitious project

Ps have you read Moore or Connlley?
If not you NEED to

Thanks for replies

Milland said:

Your "try square" is almost certainly not accurate enough to make a precision right angle.

Click to expand...

I think I named it wrong. I have a din 875 square, exactly like this one:

Of course some granite or cast iron square with higher precision class will be better, but it quite expensive.

lucky7 said:

The OP may want to add a sensitive level to his list of tools. One with a 90deg face would be even better.

L7

Click to expand...

Like this one?

Maybe I can buy one, but I don't know how to use it in my case.

michiganbuck said:

A rounded nose surface gauge can be fabled-up to transfer the check of a good square block as reference to check your part.
Just a block with two parallel sides and a little out of square can be the master, you just have to figure in the out of a square condition of the master. Perhaps the master is .005 mm off so you add/subtract that...yes likely need a true surface to do your gauging on.

Click to expand...

It is a good way to go, but I still need a bigger square and straight edge. So far it seems like the only solution so maybe I need to make them myself.

Hbjj said:

If you're spindle carrier and it's rails were finished...

Ps have you read Moore or Connlley?
If not you NEED to

Click to expand...

They are not finished.
Thanks, I didn't read but definitely will.

I'm going to address the issue of the horizontal bearing surfaces only covering a small fraction of the "frontage" of the vertical bearing surfaces. If the original bearing surfaces protrude above the cast peanut butter, place a precision parallel across them. You can then rest the square or RAA on the parallel and access the width of the vertical face. Better would be a matched pair of precision parallels. (Parallels for vise work are not what I am talking about. You want nice fat ones.)

If the horizontal bearing surfaces do not protrude above the cast peanut butter, you will need a set of matched spacers, with surfaces either ground or scraped parallel and equal distances apart. Put those on the bearing surfaces, and rest your precision parallels on the spacers. I would probably surface grind spacers as a set, and call that good after a quick sanity check with a micrometer at three places around the perimeter of each spacer.

If you happen to have a surface plate that is parallel on the top/bottom, not just flat on the top surface, you can obviously use that instead of one or a matched pair of precision parallels.

In a pinch, you can make your own parallels, especially if you don't care about hardening them. If you buy used ones, you need to evaluate them and probably correct them (knock off burrs if nothing else).

vnuEndru said:

Thanks for replies



View attachment 291741



Like this one?
View attachment 291743

Click to expand...

Or like this





Denis

It would be very helpful to know the length and width of each way, the distance between them, and the height of each of the ways above the base. It would also be helpful to know the length and width of the surface plate, and the lengths of the square base and arm.

Were you able to use the surface plate to spot and scrape the two pairs of ways at the same time? Are the two pairs of ways in the same plane?

what precision do you want to achieve and how does that match with you instruments? what grade is your plate and what grade the square (2, 1, 0, 00)?

a cylinder square would be easy to make (youtube) and you could either tune your square to that, if its long enough, and check with shims/light or set up (make) a squareness comparator as buck mentioned.

The task has been done for a very long time if bar room ditties are to be believed.

In days of old when knights were bold
And mills were less particular,
They stood them up against the wall
And scraped them perpendicular.