[rdsi]

RSDI, you mentioned earlier that the top of the ETS is threaded on, then held by a set screw. Is it possible this could be an adjustment to accomplish what I'm trying to do by massaging the height setting?

Not really, it's just how the unit is assembled. It is a good idea to loosen the set screw & make sure the top is fully screwed on tight, then tighten the set screw. Some people have complained about the top being loose causing erratic reading because of this.

I did get my ETS working and it seems to be very repeatable between activation's with the same tool in the spindle, but I was a little disappointed in the overall accuracy. In the 2-4" range where most of my tooling lives. I was getting a consistent error (long ) of around 1
to 1 and 1/2 thou. This was in comparison to my Fowler mechanical height gauge, which is pretty much dead on in this range with gauge blocks.

So I confirmed this error on the mill by setting gauge blocks on top of the ETS and doing a set tool length probe. I then adjusted the probe height setting on the machine to 0.0014 higher than the measured size. I had set the size using the procedure outlined by Entirely Crimson in his first video with the 1-2-3 block.

So the adjustment to the height setting gives me good tool lengths, to within just a few tenth's.

But the reason for this post is to ask everyone if I'm fooling myself here or is this a good idea?Terry

I'm not following you here?

Maybe this is related but the top carbide surface is not necessarily parallel to the base. Depending where you press on the ETS you can get a few thou variance across the surface of it. You can correct for this using the set screws mounted in the base. I just mounted the ETS in my grinder and made the two surfaces parallel so I wouldn't have to fuss with it. Also, the probing speed does affect your results. The faster you go the more offset you get although it is consistent. Somewhere I recorded these values but cannot find them at the moment...

[kstrauss]

Following up on RDSI's comments:
Is it possible that when measuring an actual tool it is near the centre of the top while with the spindle it is pressing on the edge of the top? I solved this problem with mine by making a spindle probe with a ball at the end. I carefully measured this "probe" and use it rather than the actual spindle. I wrote a little piece of gcode to automate the setup of tools. It first trips the ETS with my "probe" and then measures the actual tool and sets the tool table entry appropriately.