Microadjustment with and without TC question...

Obviously, each lens is calibrated a bit different within factory specs but I was wondering if it even seems possible to have a situation like mine. Basically, my Sigma 70-200 OS lens needs +8 natively on my 1D MkIII body. So far, no big deal. When I put the Canon 1.4x tc (MkII) on, I am finding that I am more like -4. Could the specs between the camera and two lenses be such that this is possible? I guess, I am pretty sure it is but it just seems odd to me for some reason.

Sounds quite possible to me. So my question is, do you have to change it manually when switching between with the TC and without? Or, does the camera recognize these as different lenses?

Bob

Camera should recognise the difference. 1D4, 5D2 and 7D definitely do

The TC also has tolerances which may be different to the tolerances of the main lens hence the difference. You are effectively creating a new lens by adding the TC

You are introducing another interface that involves dimensional tolerances, that being the TC to the camera body and changing the interface of the lens, which at +8 was accounting for the lens to body tolerances. Now that interface is controlled by tolerances between the lens and TC. So changes would be expected.

I might also add that the Canon TC is "optimized" for specific lenses in the Canon line and certainly not a Sigma.

Thanks all. I definitely understand that each piece in the "equation" has its own tolerances but my real question is whether it seems unreasonable that eh TC could take you from an adjustment of +8 all the way to -4. I don't know, just seems like a lot for a tc which really doesn't have moving optics like a true lens.

Moving optics doesn't have as much to do with things as the physical dimension from the lens to the focal plane, pretty much all of which have been varied and likely provide what is termed a tolerance stackup.

John from PA wrote in post #14378368
Moving optics doesn't have as much to do with things as the physical dimension from the lens to the focal plane, pretty much all of which have been varied and likely provide what is termed a tolerance stackup.

I'll have to research tolerance stackup as it sounds like that may be exactly what I am experiencing from a logical pov. Thanks very much!

I seem to remember that my 300 f/4 IS with and without 1.4XII had different values on the 7D, the camera recognises and stores these separately.

Tolerance stackup is simply the effect of multiple dimensions that are each within tolerance, but the cumulative effect is outside a specified range. Lets say we put down bricks in a walkway. Each brick is 12 inches long "nominally" and I decide I can lay 20 of them in a 20 foot space (I'm not making things difficult with spacing. Lets further say every brick is made to be 12 inches long plus a tolerance of 1/16 inch, minus nothing. If I had a batch of bricks, every one 1/16 inch oversize, then the 20 bricks would be 20/16 or 1-1/4 inch too long and wouldn't fit due to the tolerance stackup. Each brick is in tolerance, but the total is outside my desired boundary.

With machined parts dimensions tend to the high side. In other words, say I tell a machinist I want a series of round bars and I'd like the diameter to be 5.000 inches. I can tolerate anything between 4.999 to 5.001. Since the bars would be machined from large to small, in a 100 given samples more would be 5.001 than would be 4.999.

The opposite is true if a machinist is making holes in something. Lets say I want a 5.000 inch hole and I'll settle on anything from 4.999 to 5.001. Since the machinist is making a hole, going from small to big, more will come out at 4.999 than at 5.001.

Now, if I select a hole at 4.999 and a bar at 5.001 I won't be able to put them together because they interfere. If I pick a hole at 5.001 and a bar at 4.999 the combination will be loose. The difference between the fits is because of tolerance stackup.

John from PA wrote in post #14380124
Tolerance stackup is simply the effect of multiple dimensions that are each within tolerance, but the cumulative effect is outside a specified range. Lets say we put down bricks in a walkway. Each brick is 12 inches long "nominally" and I decide I can lay 20 of them in a 20 foot space (I'm not making things difficult with spacing. Lets further say every brick is made to be 12 inches long plus a tolerance of 1/16 inch, minus nothing. If I had a batch of bricks, every one 1/16 inch oversize, then the 20 bricks would be 20/16 or 1-1/4 inch too long and wouldn't fit due to the tolerance stackup. Each brick is in tolerance, but the total is outside my desired boundary.

With machined parts dimensions tend to the high side. In other words, say I tell a machinist I want a series of round bars and I'd like the diameter to be 5.000 inches. I can tolerate anything between 4.999 to 5.001. Since the bars would be machined from large to small, in a 100 given samples more would be 5.001 than would be 4.999.

The opposite is true if a machinist is making holes in something. Lets say I want a 5.000 inch hole and I'll settle on anything from 4.999 to 5.001. Since the machinist is making a hole, going from small to big, more will come out at 4.999 than at 5.001.

Now, if I select a hole at 4.999 and a bar at 5.001 I won't be able to put them together because they interfere. If I pick a hole at 5.001 and a bar at 4.999 the combination will be loose. The difference between the fits is because of tolerance stackup.

Well said.

The same way, a within tolerance body and a within tolerance lens may yield an outside tolerance combination if let's say the body is -3 MFA and so is the lens, so there aggregate is -6 which, on a non-MFA body is significant. So Canon (or Sigma or Tamron etc.) may state that it is within spec but it may be out of spec on certain bodies. And this is also why some lenses work better on some friend's identical camera body but not on yours (hypothetical example).

Oh, and as for the original post, it's quite possible, I also experience this phenomenon with my TCs.

Perfectly normal and not unusual.
For example my MFA setting with 5DMKIII
70-200 F2.8 IS - W:-3, T:+15
70-200 F2.8 IS + 1.4x TC - W:-5, T:-5
70-200 F2.8 + 2x TC - W:-5, T:-10
As you can see it's all over the board.

Very good info, thanks guys. At this moment, it seems like I am -11 with the TC (and might still be just slightly backfocused) and +5 natively. What's interesting is that I did a bunch of focus tests using charts and whatnot but couldn't get it right. It was while I was actually shooting a game (and admittedly chiming between plays since this issue was driving me nuts) that I saw how far off things still were. I made some adjustments through trial and error during the game and think i got it right. Well, the images looked good, focus seemed consistent, and the keeper rate went way up so I'm pleased.

Thanks again everyone.