Fun with Canon focus micro-adjustment

Tobers

Tobers

Messages
3,288
Name
Andy
Edit My Images
No
Just posted this on another forum but thought I'd share...

I had a nice order from a polo club based in Barbados for shots of their team that I took the other week, so decided to splash out on a Sigma 100-300 F4, from Jessops no less. I figured as I can get the VAT back it's about the same price as buying from Hong Kong, and I get a 28 day no quibble money back as well. Better than hiring a lens I reckon! Also got another £35 off by ordering online with "pickup in store" and coupon code "SUMMER35". If all works out well I'll put my 70-200 F4L up for sale after the weekend and break even quite nicely.

I decided not to wait to get enough money for the 120-300 2.8 as I reckon I need some decent DOF anyway. Maybe next year...

First things first - onto the tripod to check the focus. Usual procedure - well-lit target taped to a door down the hall, camera on a tripod with mirror lockup, set in manual at wide open (f/4). Camera tethered to pop the pics straight into my PC.

Here's the first non-calibrated shot. All these are rather mega crops by the way - probably 200% or something.

#1 no adjustment
312572820_CELyg-L.jpg


#2 After some fiddling around, I settled on +16
312571968_ykUgf-L.jpg


Nice - I'll have that as an improvement any day!

Most lenses get sharper as you stop them down a bit. So...

#3 at f/8 also on +16 adjustment. I wanted to ensure I didn't screw up the smaller apertures somehow.
312572244_GCpjy-L.jpg


#4 Lastly, here it is with a Canon 1.4x extender added at f/5.6.
312572554_4owwD-L.jpg


Needless to say I think this micro adjustment thing is well worth it. For those wondering, it's on 1D and 1Ds mkIII cameras only at the moment. I'm sure it will filter down into the new 5D whenever it arrives, then the 40D range etc. Canon would be mad not to.

Interestingly though, the lens detecting software didn't recognise that I was using a Canon 1.4x extender, so unlike with the 70-200, I cant have a separate micro-adjustment with and without the extender.

So - hopefully sorted for some good action shots on Sunday. The lens looks great by the way, and I love the way Sigma stuff comes with a really tough case. Build quality v.nice. Zoom ring goes the wrong way which will lead to some cock-ups I'm sure. All good so far...

Tobersssssssss
 
How did you actually decide how much to + or - the focus? Did you just take a range of shots adding 2 incrments or something. This is definately for me. I have a 1D and 2 of my Canon lenses don't focus sharply.
 
The range of adjustment is -20 to +20.

I start by taking shots at -20, -10, 0, +10, +20. You can pretty quickly work out where the best focus is happening. In my case it was between +10 and +20.

I then took another set of shots at 1-point increments from +10 to +20, and picked the best.

Takes about 30 mins.

Make sure you refocus after changing the adjustment and before taking each pic!

Also, be sure to use mirror lockup and a remote release if you have one, or self timer if you dont. At 300mm with a slow shutter speed, the mirror thunk made a lot of difference to the sharpness of the pic.

Note down the settings per lens, though the camera should remember it. It's worth copying your custom settings to a memory card anyway just in case.
 
Thanks Tibers, I'll try it. Do you refocus manually or auto focus? I'm guessing it should be auto so you can let the camera make the adjustment you have selected?

I really notice error on my 70-200 2.8 IS. Focus at 70 and then zoom and the 200 image is pretty blurry. I can only assume that's because the initial focusing error is being magnified?
 
steveophoto said:
Thanks Tibers, I'll try it. Do you refocus manually or auto focus? I'm guessing it should be auto so you can let the camera make the adjustment you have selected?

I really notice error on my 70-200 2.8 IS. Focus at 70 and then zoom and the 200 image is pretty blurry. I can only assume that's because the initial focusing error is being magnified?
Click to expand...

You must autofocus (y). Also, always refocus after you have zoomed as the zooming will alter the focus.

sportysnaps said:
if you have two lenses the same (as i do) can it tell the difference between them or is one canon 24-105 f4 the same as the other as far as the camera is concerned
Click to expand...

Nope - a 24-105 is a 24-105. It doesn't know the difference between two copies of the same model lens.
 
Yes, if the lens models are different it will remember them and apply the adjustment when the lens is attached. I'm sure it can remember more models than anyone can afford.
 
So how do you know it's not not the F stop your using giving you a soft image and not the lens. You see, you have shot a flat surface, so you can't tell if the lens is front focusing or back focusing. It would need to be a 3D subject to work out the adjustments, or am I wrong.
 
PapaLazarou said:
So how do you know it's not not the F stop your using giving you a soft image and not the lens. You see, you have shot a flat surface, so you can't tell if the lens is front focusing or back focusing. It would need to be a 3D subject to work out the adjustments, or am I wrong.
Click to expand...

Because shots #1 and #2 were at the same f stop they ought to have identical depths of field.

It's probably not THE best test, as the camera might be focusing elsewhere on the paper which would give different results with a razor thin DoF. Probably not an issue at f/4, but could well be at f/1.4.

I'm a big fan of this focus test method:
http://www.focustestchart.com/chart.html
 
PapaLazarou said:
So how do you know it's not not the F stop your using giving you a soft image and not the lens. You see, you have shot a flat surface, so you can't tell if the lens is front focusing or back focusing. It would need to be a 3D subject to work out the adjustments, or am I wrong.
Click to expand...

You need to use the widest aperture and keep it constant during the test. Once done, then check other apertures & focal lengths to make sure all is OK there as well.

All sorts of advice on the method, but the general concensus is a vertical flat surface works best. A 45 degree tilted surface was originally recommended but apparently not any more according to Canon. Oh yes - I was using centre point focus only to avoid it focusing on other bits of the page - you need to ensure the tests are consistent.

I dont think it matters if it starts off front or back focusing - the adjustment goes from lots of back-focus to lots of front-focus and you choose the sharpest setting. If the lens was back-focusing, it wont be once you are finished fiddling. I dont really care what the lens was doing once I've done the adjustment as it's academic by then :).
 
Is there any inconsistencies when doing AF micro-adjustments for zoom lenses? what i mean is after adjustments, Do you get different results when using different focal lenghts of the zoom lens.
 
PapaLazarou said:
Is there any inconsistencies when doing AF micro-adjustments for zoom lenses? what i mean is after adjustments, Do you get different results when using different focal lenghts of the zoom lens.
Click to expand...

It should be done at max aperture, initially, and max zoom.

Interestingly, when I checked all my L series after they had been back to Canon for calibration, no microadjustment was necessary.

Hence I always get them checked under warranty as a matter of course.
 
Just done my lenses.

Micro adjustment done on Canon 1DMKII

Canon 50mm F1.2 L needed +18 (pin sharp on my 1DSMKII and 1DMKIIN)
Canon 70-200 F2.8 L IS needed none
Canon 24-105 F4 L IS needed none
Canon 17-40 F4 L needed none
Tried to do my Sigma 150mm macro but didnt seem to work, will it only work with Canon lenses ?.
 
You must log in or register to reply here.
Back
Top