Thread: From Tractor Weight to Inspection Square

Hi Josh,
Have you thought about clay? Thats the way they used to do it (hardening files is the thing that springs to mind straight away) and cheap too. Also it should contact 100% of the surface so there will be no oxygen to affect the iron.

Originally Posted by Ueee

Hi Josh,
Have you thought about clay? Thats the way they used to do it (hardening files is the thing that springs to mind straight away) and cheap too. Also it should contact 100% of the surface so there will be no oxygen to affect the iron.

Hi Ewan,

No I had not thought of using clay, I can see some advantages in cost. I do have some stuff to use on knife blades that melts on and and then shatters off when you quench PBC Special, better than SS foil for small tools.


I have made a few steps forward on the square. I have given it a coat of paint and ground all the measuring faces flat and parallel. flatness is now ~ 1µm on all faces and parallel is better than 2µm. I also put a chamfer on, I was aiming for a 0.3mm chamfer but I was not paying enough attention and my down feed stop had slipped 0.2mm so I ended up with a 0.5mm chamfer. The chamfer I think adds a nice finish detail but it was more of a practical item for grinding. One of the problems I have with grinding stuff down at this level is the grinding burr that forms, I stone it off after every operation, but some times it shows up again (because it is so small that you can not feel it, but its there) throwing a setup out by a few µm to my frustration The other reason for the chamfer is that it strengthens the corners from damage.
Squareness comes next, The method I'm using for testing square is the reversal method and it requires a proof of parallel for accurate measurement. Because it is a box I can also confirm the readings by flipping as well as reversal handy as it another test of parallel and squareness .I have squared one axis to 5µm TIR over 200mm and the other two ~11µm and ~17µm TIR over 200mm. so getting closer, I'm hoping that the chamfer takes out some of the variability I have been seeing during the flatness grinding.
Anyway here are a couple of pics.
DSCN2856.jpg
DSCN2857.jpg

-Josh

Hi Josh,
Two questions tonight.
1. why didnt you chamfer the short sides?
2. Have you done any maths on how errors on the plate could effect your readings? I'm wondering if you need a straight edge to make sure you are on the same spot on the table each time?

Stuart

I am just wondering since it's getting down to this sort of high precision, would a nice timber ring in the center to stop you hands from heating it, or a piece of timber in 4 spots help?
I do think a nice timber ring about 20-25mm wide would look nicer though.

Dave

Originally Posted by Stustoys

Hi Josh,
Two questions tonight.
1. why didnt you chamfer the short sides?
2. Have you done any maths on how errors on the plate could effect your readings? I'm wondering if you need a straight edge to make sure you are on the same spot on the table each time?

Stuart

1. It's not a lack of desire I just have not figured out a way to fit it under the grinder. I just put a new wheel on and now my clearance is only ~254mm. I do have an idea of how I might do it with the square off the edge of the chuck, but I will have to test if it will reach over far enough
2. I'm using the small 400x400 00 spec plate for inspection its error is about 2.2µm but most of that is near the edges the centre is quite good. But then again I did not do a full survey of the plate I just ran the interferometer across the centre lines to see it was going to be close to a 00 spec. So in the strict sense the error from the plate is unknown. Stuff that was spotted on the small plate seems to spot the same on the good spot of the medium 600 x 900 plate I know that is not proof but it is a good indication that there is not a major discrepancy at least not in local flatness. To properly test I should set the square up on gauge blocks (which i don't have) at its airy points and test it like that. But to answer the question no I have not done the math.

This touches on the reason why I would never call this square a master no matter how well it measures up, the error accumulates the error in the plate, plus the error in the gauging of flat, parallel and the square.

That being said. Some of the error averages out, and this is the major advantage of the box square, If it passes the flatness and parallel test you can test the same error in angle 8 times. and if you get the wrong numbers you know you have an inconstancy in the surface plate or flatness/parallel or too much hysteresis in the gauge.

-Josh

Originally Posted by Dave J

I am just wondering since it's getting down to this sort of high precision, would a nice timber ring in the center to stop you hands from heating it, or a piece of timber in 4 spots help?
I do think a nice timber ring about 20-25mm wide would look nicer though.

Dave

That would be do-able, In fact it is such a good idea I think it will bug me until I make it.

If were me Josh, to grind that edge I'd fix the square flat to a plate, set it at 45 degrees and touch off the corner. Use a dial indicator to step over 0.50 mm then plunge grind with very short traverses. Rotate the square 90 degrees and repeat. Stand it up off the fixture plate if you don't want to touch it, but it's just a fixture, so dealer's choice.

Yes headroom is a PIA, I ran out of room on mine last week, and no amount of head scratching, cursing, and cups of tea while staring at it was going to change the fact I needed to make up a shorter fixture .... Or buy a different grinder. I chose plan B!

Pete

Not being able to stand not having the short edges chamfered here is the setup. I was just able to get it under the wheel with about a 1mm or 2 to spare. and there is just enough travel to get the wheel to come over enough to touch the entire edge with about 10mm to spare.
The nice thing about the setup I was able to just rotate the fixture by 90 degrees each time, no need to un-clamp and reposition.


DSCN2862.jpgDSCN2863.jpgDSCN2866.jpgDSCN2867.jpg

-Josh

What, you're not going add another facet where the three chamfers meet?
GWJ (feeling lazy tonight) Thats one good looking tractor weight

Stuart

Wondering how long (how much $$) it is going to take for some forumite to convince you they need it more than you do. You have made yourself a great bit of kit and there be plenty greeneye monsters out there.

Maybe I can send you all my casting for you to practice on

Originally Posted by DSEL74

Wondering how long (how much $$) it is going to take for some forumite to convince you they need it more than you do.

No, just reminded me that I need to get back to pattern making and one like this would be a good addition to the collection.

PDW

I have been meaning to to an update on this thread.


I ending getting it to better than 2 microns in square and parallel all the way around both ways (the limit of repeatable measurement of the 00 plate is about 2 micron so there is some uncertainly in how good it is). It was quite a challenge getting there, grinding, scraping and finally lapping. I would have to leave it "settle" between working it and measuring it. When I got to the limit of reliable step grinding I started what I will "step lapping" ie using a 5000 grit stone to take off the smallest fraction and then measure the corrections, once the desired result was measured it was then put back on the grinder and parallel ground the opposing side and then flippy flopped to grind the lapped face. I would then re-check it both way and repeat as required.

I have done all that, and I'm still unhappy with it.... The heat treatment made it too soft so it scratches and mark too easily, so it will be back into the furnace to be made harder and then refinished to the same level.

-Josh