Sorry folks, I never quite got to writing up parts 3 and 6. Maybe next time I do the alignment myself I’ll take photos and write them up. I wouldn’t hold my breath though, it might another few years :-).
I downloaded the spreadsheet so I could try to understand the exact process you use and thus understand how I could use this method on my car which has the suspension modified with aftermarket parts which enable more adjustment from stock.
All good until I get to the alignment formula. It has 2 parts or triangles to solve. The first is obvious and the second seems to be an adjustment for the camber of the wheel on the length from “triangle one”. I would have thought that triangle 2 is made with the tire radius and the length to the centre of the car, not as the formula states – to the tape measure. Also, there is an adjustment by “8”. Can’t figure what that number refers to.
Appreciate any insight to help me understand your work.
Rob, here is some info for you. Better late than never, huh?
You are right, the second part of the equation for the toe calculation is to compensate for the camber of the wheel. This is because the laser level is mounted over the wheel’s center but is pointing downwards just slightly so that the beam hits a tape measure on the ground. This slight angle off horizontal introduces a very small angle change in the laser pointing due to the wheel’s camber. For our typical negative cambers, pointing the laser downwards to hit the tape measure will push the beam outwards away from the car’s centerline, and the amount it pushes out depends only on the camber and the distance the beam travels. Now, those two are closely related since if the tape measure is really far away, the laser pointing direction downwards will be very small. So it’s possible the formula could be simplified. But I think it would not be related to the distance from the center line.
The most accurate distance measurement from laser level to the tape measure is, from my experience, from the end of the level where the laser is shining from, since I was able to use a T-square to make sure I drew a vertical line to the ground. That means that half the length of the laser level needs to be added to this measurement to get to the center of the wheel hub. This is the “8”, i.e. 8 inches, for the laser level model I used whose total length is 16″.
Thanks! I know this is an older post but I’m glad that I found it. My A35 was all but useless and now it’s back to tracking focus. Not sure how it could get that much out of adjustment but it did. Much appreciated.
I’ve found this schematic very useful. I have to fix my old SE240 Hafler power amp, but I cannot find the values of two burned resistors, marked at PCB as R24 and R59 (left and right channels), one of them it’s broken and the other measures 24?, but the colour of the stripes are changed, may be the value is 18? or 100?
Somebody can help me with this?
But surely you are supposed to tighten the eccentric nuts with the weight of the car on the wheels, otherwise the arm is always twisted in the rubber bush.
In El Capitan (at least on my system), this fix doesn’t work, as the ssh config location has moved. Simply add the fix to /etc/ssh/ssh_config instead of /etc/ssh_config
Thanks Pete. For a reason I can’t explain, the link table on the first (default) page was missing one link, to the “Calculate adjustments” page. I fixed that. The remaining missing steps 3 and 6 are, in fact, missing! Sorry I just have not had the time/inclination to complete them. I figured the key steps were published and people could “fill in the rest” but that’s probably just laziness on my part. Thanks for the nudge. I will try to get to them.
I have a Sony A700 and I had to calibrate the sensor recently because of a back focus problem. After I’ve done this, the focus is very sharp but I now have a new problem. I can’t focus on infinity using wide angle lenses. Even on manual focus. I’ve tried different lenses with the same result. It seems that the lens stops in a break before it can reach focus. Can you help me on this?
Looking at your drawing it looks nearly identical to the PRO2400 schematic, but without the balanced input and level control board. The PRO2400 manual is available on HiFi Engine.
Brilliant solution, and exactly what I was planning when I found rsyslog slowly killing server performance.
If you are interested, there seem to be 2 issues that cause this problem:
1) syslog using TCP so *.* @@serverip -> this says you MUST get a response in order to keep logging and can also be related to the main issue below
2) Syslog has a default of “NEVER LOOSE ANYTHING”, so if it runs out of queue space etc, is starts slowing everything down by not responding.
It is by design, for places like banks who would rather have everything logged and slow performance than “loose logs”.
So you have a few possible options:
1) use UDP for syslog and risk some loss there
2) Configure syslog to drop some logs should it become swamped
3) Configure syslog to use cache, logging etc better
4) don’t send log heavy stuff through syslog
Again though your solution solves the issue, and for servers where performance and uptime is more important than logs a great fall back even if you do all/any of the above.
Thank-you for your patience and effort to trace-out the 240 schematic. It is very helpful. (and tedious, I know!)
It seems the 2SK163 could be a JFET, which is used sometimes instead of the trusty 2SK170 as a complement to the 2SJ74, when a higher working voltage is required.
The 2SK1636 often shows up in a search as a 2SK163, with the final “6” dropped for some reason. The 2SK1636 is , indeed, a medium power MOSFET; but this would be unusual in a symmetrical and balanced diff input stage…
Hello! I just wanted to ask if you ever have any
trouble with hackers? My last blog (wordpress) was hacked and I
ended up losing a few months of hard work due to
no backup. Do you have any solutions to protect
against hackers?
Basic stuff: use a high-quality provider, use high quality plugins for spam/hacking/etc., update WordPress frequently, use different user names from the defaults and complex passwords.
Great detective work! The 20 mins that Matthew refers to is because there is a commented out line in the ssh_config file for the ForwardX11Timeout 1200 which must be in seconds dan default figure. I just uncommented that line and changed the figure to 596h. Thanks as this was really bugging me trying to test a python app.
Hi, thanks for the guide! I have been here before, but didn’t have much luck with adjusting the AF module (it did work perfectly only with the lens I was using to fix it and very bad with the others).
I came back here recently, modified your protocol and added few more control steps here and there and now all my lenses work perfectly!
Thanks for sharing your version, it’s been a very good starting point and to know what the screws do in the first place! 🙂
I followed the instruction here to adjust the AF module of my Sony A55 and it was a great success. Now my Minolta 50/2.8 Macro focusses very precisely.
An alternative method to that described is to just adjust left and right until both are in focus. You will then find that the center is in focus too.
BIG thank you for this! You should be awarded a medal!
I was prepared to toss my a580 with serious backfocusing issues but with these instructions I managed to get the camera tack sharp, irrespective of lens. Sony service could not do anything about it.
A580 has screws in the bottom under the sticker in the middle, but otherwise the same procedure. Found the latter procedure much better, i.e. screw all 3 fully in, and then 1/2 turn open and then start adjusting. Mine was an additional 3/8 turns on every screw.
Same problem with my old alpha 100, will adjust it next.
I joyfully (sorta) found your article on back-focus problems with the Sony A55. I too am a lucky owner 🙁 of a A55. I also own a Panasonic LX -3 which I dearly love.
A few week ago, I was scrutinizing the same landscape shot with both camera. The A55 lens was a Tamron 10-24mm f3.5 set to 10mm, aperture pryarity mode and stopped down to f5.6, image quality: fine, image file: large, steady shot off, auto focus: automatic AF, AF area: wide, ISO 100, HDR off, files saved as jpgs, cameras were tripod mounted. The LX-3 was set to approximately the equivalent settings.
After making the photos, I opened them and looked at the two images. The LX-3 was tack sharp, in fact, it made hash out of the A55 in just about every respect: contrast, dynamic range, sharpness, etc. Here is a little camera, with a small chip costing 1/4 the price of my DSLR and pounding it into the ground. The A55 was soft except for the extreme right side of the image. The left was very soft, but there looked like more was going on. There as more chromatic aberration with the Tamron lens, which I expected, I also expected some softness in the corners, it’s a Tamron . . . and a 10mm. But this didn’t seem right. The camera has never been dropped, so I ruled out physical damage by me. I know that Sony cameras using a prism instead of a mirror can be prone to “ghosting” but I wasn’t about to let this slip by.
I began searching and found your article, then read Jeff Fridel’s article and printed his focus charts. I’m ready to begin your process of aligning the focus sensors, but before I start (a process I’m not looking forward to) I wanted to get your opinion. I tried my other lens, a Sony 18-55mm just to make sure it wasn’t the Tamron. It was a tad better, but still with all the same basic defects as the Tamron. I read the reviews in DPReview of the Tamron before I bought it and the comments were very favorable.
Given your experience, do you think I have a miss aligned focus sensor? I manually focussed the image using the focus magnifier and it looked go. I was a photo major at RIT so I know a bit about lens aberration, mirror misalignment, etc. But my training is 1980’s vintage. film cameras are gone, people don’t even know what view-cameras are. It’s a new photography world.
If you have time and are willing, could I send you some images to look at?
A misaligned focus sensor module would not explain what I believe you are seeing, which if I can paraphrase is a difference in focus across the frame on a single exposure, i.e. sharp on one side, soft on the other. A crooked focus sensor would affect all parts of the frame the same way. I would have to assume that either your sensor is crooked, or the lens mounting is crooked, or the lens itself is crooked within the barrel.
Thanks for the article. I have A55 and recently bought Minolta 50/1.7. Several tests showed that I have a front focus with this lens. Before that I used 18-55 kit and Minolta 35-70/4 lenses. They are not that fast, so I did notice the problem. Anyway, I’ve tuned the AF, thanks to your article, and here are my observations:
1. After tuning the AF for 50/1.7, I’ve tested other two lenses and there seems to be some back-focus problem on 18-55 at 18 (at 55 it is OK), and there is a huuuuuuuge back-focus on 35-70 at 35. Strange…
2. I’ve first tried to play with left and right bolts with no success. Then I’ve tried to screw them (all 3 bolts) tight and release half of the turn. This position was used as a reference point and what was really strange – AF did not react on those two bolts (left and right). But I was able to move focus plane front and back using the third bolt (Focus height adjust). I do not know what is wrong with my actions, but I’ve tried several times to adjust it with left and right bolts – no success. When I turned left or right bolt too much, it would stop focusing at all, I had to compensate it with “Focus height adjust” bolt, which also affected the front/back focus, which makes it even more difficult to adjust something with left and right bolts.
Do you have any ideas why it was like that? I’d expect it to work in a different way: left and right bolts are used to adjust front/back focus, and third bolt is used to move sensors up and down (so that actual sensor position will match rectangle in the viewfinder).
Thank you for this excellent article and explanation of how to adjust the auto focus. Worked well on my SLT-A33. Fairly easy to do and it sure beats sending the camera into Sony with only so-so expectations of a good result.
Somehow, I had developed what I suspected to be a serious AF problem. Using the autofocus test chart from Jeffrey Friedl’s Blog ( http://regex.info/blog/photo-tech/focus-chart ) I was able to confirm that a pretty large amount of front focus was the problem. I didn’t have time to do a precise correction, but did a rush job before going to an indoor volleyball shoot and had a significantly better time getting usable shots. I’ll go back and fine tune things next, but I just wanted to go ahead and say now how much I appreciate the help provided here.
That would be good too, as I tried searching Google docs to find it.
Your method appears to be the most fool proof and easiest of the ones I’ve seen so far.
A few more parts have been published, now parts 1, 2, 4, 5, 7 are available. Missing are camber measurements (part 3) and front wheel adjustments (part 6). The most complex parts are all published. 3 and 6 are still coming….
/etc/ssh_config is available and has been at least since Lion, I believe (and maybe before that), but of course you must have administrator permissions to edit the file.
You can type “ssh -Y” to get around this whole problem, but there may be some security issues with that. I would read up on the tradeoffs. Excerpt from the ssh man page for the -Y option says: “Enables trusted X11 forwarding. Trusted X11 forwardings are not subjected to the X11 SECURITY extension controls.”
I just got my A 55 back from Sony (Feb 2013) and the dumb asses are still saying no problem found. These people are pathetic at best. You would think after three years of experience with the well documented back focus problem they could hit their ass with both hands, but I guess not.
No kidding. It’s so blindingly obvious if they would just use the far left and right local focus points in the test, in addition to the wide area which is the factory default setting.
It’s taken me too long to find you! Thanks for a great post.
My CZ 16-80 appears to have more backfocus than the original kit lens. Now I’m going to spend some time tweaking
Thank you for nice article! I have the same problem with my Sony A380. It’s so irritating! Newer had problem like this with it before, so i just bought a new nikon D3100.
Thanks for the post. I just ran into this installing a remote Oracle database and it kept failing half way through because it could send out the display. Nice job!
It’s not a bug, it’s how OpenSSH works. and you don’t have to set ForwardX11Timeout to something stupid long, just set ForwardX11Trusted yes in your ~/.ssh/config or /etc/ssh_config files
ForwardX11Timeout
Specify a timeout for untrusted X11 forwarding using the format described in the TIME FORMATS section of sshd_config(5). X11 connections received by ssh(1) after this time will be refused. The default is to disable untrusted X11 forwarding after twenty minutes
has elapsed.
Interestingly, I couldn’t put the option in my (NFS shared) ~/.ssh/config, as some linux ssh (e.g. OpenSSH_4.3p2) didn’t recognise ForwardX11Timeout.
I have another issue. On my A55 when I zoom I have back focus, but when full wide I get front focus. I have tried this with a few lenses. I have calibrated mine for the Sony SAL 28-75 f/2.8 at 50mm as it is the lens I use the most, but I am still unhappy. Zooming to 75mm will give me back focus, but one wide (28mm) I get front focus.
Any ideas?
While I can’t answer your questions specifically, I recently went through a little more learning about the AF system theory and construction and found an excellent article on understand autofocus: http://www.clarkvision.com/articles/understanding.autofocus/. This may or may not be the case for the a55 but there are different groups of AF sensors that can be used depending on the f/ number of the lens on the camera. Wider apertures (f/2.8 or lower) are supposed to able to take advantage of AF sensors further apart which therefore have better distance discrimination and are faster and more accurate to focus. I am wondering whether maybe f/2.8 is right on the borderline and at one focal length extreme the f/2.8 sensors are used and at the other they are not.
Another thing I am noticing with my lenses after revising this post is that not only does AF accuracy differ from lens to lens, but also for different focus distances. I tuned my AF sensors for a Minolta 135/2.8 and it reliably and consistently snaps into essentially perfect focus at a range of distances from about 2 meters on up. I have an old, cheap Minolta 50/1.7 that shows significant front focus when the test chart fills the field, which is at about 0.7 meters distance. The front focus reduces as distance increases, but never quite gets to accurate focus before the focus mark in the center becomes too small and the AF sensor starts picking up the patterns on the sides of the chart to focus on.
Dear Riaan, did you ever manage to sort out what was wrong here? I have exactly the same problem, except I get back focus when fully wide and no problems when fully long.
Thank you very much for this post. I have been fighting this exact issue since I was forced into a 10.7 upgrade in September. I had the parameter name correct, but kept trying to make it work with 30d.
Awesome article – I was able to fix the backfocus on my a33 a few months ago.
Just got an a57 and I’m going through the backfocus issue again! Any ideas on how to adjust the focus sensor? There doesn’t seem to be any access panels near the flash anymore and I’ve tried removing some of the top screws but I still can’t find those adjustment screws.
I’m reluctant to bring it back to Sony. They kept my a33 for 2 months just to tell me there were no backfocusing issues.
I just stumbled upon your EXCEPTIONAL post concerning the Sony SLT-a55 focus issue. This was after having visited my local camera shop Sony expert today for the same problem; and leaving with the feeling that I should consider sending my a55 to Sony for repair. After reading your post and others’ comments, I think that would be a waste of my time and money! So, I think I want to try performing your adjustment myself — as I apparently have the same problem you experienced with your camera.
I looked at my camera and tried to determine where the two covers are under the pop-up flash and where to pry to pop them off so I can access the three adjustment screws. It’s not obvious to me. Could I impose on you to describe in detail how to do this — so I hopefully won’t damage my a55 while trying to fix a problem I should never have had?!!
Thanks in advance for any additional guidance you can provide — AND, for the EXCELLENT description of the problem, solution, etc.!!!
Oops! I must have stayed up too late last night. Re-reading your original post this morning revealed that I had somehow missed your link to the DPReview.com post that showed pictures of the cover removal process. As a result, feel free to remove both of my comments to eliminate the clutter.
Many thanks again!! I feel you have done ALL Sony a33/a55 users a HUGE favor in making this information available!
I *REALLY* wish someone actually finally went ahead and peel the darn shell of the body of an A57 or A65 and show the world where in the bloody hell those darn adjustment screws are. It’s *REALLY* getting annoying at this point !
I bought a A65 in november of 2011 and by december it didn’t turn on then the thing was sent in and with in a month the batteries were draining all the while talking with sony tech’s and having focusing problems which they said was me then finally in late march did some testing and sent the tech tewst pictures they told me to send it in with the new lens i had bought 70-400G and i got it back 5 weeks later and it was just as bad and contacted them again and sent in test .Now they have it again and we will see . But 6 months of provlems with a brand new sony A65 and a brand new 70-400G and talking on my cell cost tons of money you would think they would get it right and compensate people some how for their mistakes. I think i will run the tests when i get it back right away again and if it has problems i think i will sell all my sony stuff off and go to nikon what a shame . For right now i’m using my A330 and get better pictures then with the a65 and my 70-400g lens wow that says something and they don’t sell DSLR’s any more only SLT”S , Bad move.
I have had this same issue with my A55 since I got it. I have an A200 and never had this problem.
Now the camera is out of warranty. Can you suggest where to take it for repair since I am not very savvy when it comes to doing my own adjustment?
First, there are times the image is very sharp, and other times, it is less than satisfactory. I have tried different focus modes (tend to prefer single auto focus so that I can frame the image after selecting a focal point).
Second, when focusing on faraway objects, many are out of focus.
Third, when using manual focus for distant images, I have to back up a smidge from the infinity setting.
Are these issues common to the focus problem you are describing here? And where would you recommend a novice take a camera to to make the kinds of adjustments addressed here. Thanks.
It is hard to say whether you are seeing the same issues I’ve described without some testing along the lines of what I did. The default (wide area) auto focus mode takes focus readings from a number of sensors and selects one that it thinks is the subject. From the Sony rep I talked to, the most likely selection mechanism is to use the one corresponding to the sensor that sees the closest part of the subject. But it adds uncertainty into the process. One hint is whether the default focus consistently chooses a sensor that is off-center and does not seem like the right selection.
From a big picture point of view there are two possible significant focus issues. One is that the focus sensor block is mounted straight in the camera but the whole sensor is not correctly adjusted, so no auto-focused images will be sharp. The other is that if one of the sensors is correctly located but the block is crooked (that was my problem, and this kind of misadjusted sensor passes Sony’s focus test). To figure this out you will probably need to use local area focus and try focusing using the leftmost sensor, center sensor, and rightmost sensor and see if there is a consistent pattern at each sensor. In my case I finally figured out that using the rightmost sensor always produced sharply focused images, but using center or left sensors produced poorly focused images, and the further left the further out of focus.
I don’t know if this has anything to do with needing to back up a smidge from infinity using manual focus, since the auto focus is not used at all and it may have to do with the lens. My guess is that most lenses are designed to allow focus to go slightly past infinity so that normal manufacturing tolerances and wear over time will still let the lens focus at infinity without running out of focus range before infinity is reached.
I had my A55 in to Sony for something else asked them to take a look at the focus. They fixed the other problem but didn’t even respond to my comcerns about focus. Now I’ll send my camera to someone else to get it fixed and try to get Sony to reimburse.
Surprisingly, I still see the crash if I use XQuartz 2.7.1. http://xquartz.macosforge.org/trac/ticket/553 was opened, but closed with a message it should be filed at freedesktop.org. a search there for “ForwardX11Timeout” returns no results, so I’d guess it’s not bee reported to them. I was going to file it myself, but I don’t care enough to create an account.
Outstanding and in depth analysis, I am facing the same issue with my A55, I did give my camera to Sony repair center, they updated the software but it has still not solved the issue, As I am not as technical as you, I have now your email to back my concerns.
Thank you
Nice atricle was going to write one just like it as I experienced the same situation with the alpha 580, such a great camera (when you sort this little problem out) but Sony really letting themselves down by not paying enough attention to the autofocus allignment.
On the non SLT models like the a580 the screws are at the base of the camer under the L shaped sticker.
Same experience. Sony, give us micro-adjustment in all cameras so we don’t have to deal with support centers unwilling to spend the time it takes to properly adjust the AF.
Thanks for the article. I have A33 and exactly the same situation. Camera has been sent to service center and today received it back with “we could not repeat the problem”. Back focus is pronounced with wide angle – 18mm with kit lens or 28/2.8, but at the same time focus is perfect with 50/1.7 lens.
Paul
/ Sep 8, 2023Hey is the pro2400 the same as the se240 ?
E
/ Feb 28, 2021Are you still alive mate?
E
/ Feb 27, 2021What happened to parts 3 & 6?
Mike
/ Mar 1, 2021Sorry folks, I never quite got to writing up parts 3 and 6. Maybe next time I do the alignment myself I’ll take photos and write them up. I wouldn’t hold my breath though, it might another few years :-).
Ramachandran Ammapet Vijayan
/ Sep 24, 2020Thank you very much!
rob
/ Apr 19, 2020Mike, this is very clever, thanks.
I downloaded the spreadsheet so I could try to understand the exact process you use and thus understand how I could use this method on my car which has the suspension modified with aftermarket parts which enable more adjustment from stock.
All good until I get to the alignment formula. It has 2 parts or triangles to solve. The first is obvious and the second seems to be an adjustment for the camber of the wheel on the length from “triangle one”. I would have thought that triangle 2 is made with the tire radius and the length to the centre of the car, not as the formula states – to the tape measure. Also, there is an adjustment by “8”. Can’t figure what that number refers to.
Appreciate any insight to help me understand your work.
Mike
/ Mar 1, 2021Rob, here is some info for you. Better late than never, huh?
You are right, the second part of the equation for the toe calculation is to compensate for the camber of the wheel. This is because the laser level is mounted over the wheel’s center but is pointing downwards just slightly so that the beam hits a tape measure on the ground. This slight angle off horizontal introduces a very small angle change in the laser pointing due to the wheel’s camber. For our typical negative cambers, pointing the laser downwards to hit the tape measure will push the beam outwards away from the car’s centerline, and the amount it pushes out depends only on the camber and the distance the beam travels. Now, those two are closely related since if the tape measure is really far away, the laser pointing direction downwards will be very small. So it’s possible the formula could be simplified. But I think it would not be related to the distance from the center line.
The most accurate distance measurement from laser level to the tape measure is, from my experience, from the end of the level where the laser is shining from, since I was able to use a T-square to make sure I drew a vertical line to the ground. That means that half the length of the laser level needs to be added to this measurement to get to the center of the wheel hub. This is the “8”, i.e. 8 inches, for the laser level model I used whose total length is 16″.
Raph C
/ Oct 21, 2019This is amazing! Thank you so much!!
Christopher M Jacobs
/ Jul 25, 2019Thanks! I know this is an older post but I’m glad that I found it. My A35 was all but useless and now it’s back to tracking focus. Not sure how it could get that much out of adjustment but it did. Much appreciated.
PGV
/ Sep 27, 2018Thanks a lot for the fix!! It works perfectly now on my mac.
zz
/ Feb 3, 2018Thanks. that really help!
gio
/ Nov 2, 2017Thanks !!!!!!!!!!!
Cisco
/ Sep 7, 2017Thanks
Petri
/ Aug 14, 2017Thanks for the solution! This has bothered me for the past week 🙂
Ricardo (Madrid, Spain)
/ Feb 16, 2017I’ve found this schematic very useful. I have to fix my old SE240 Hafler power amp, but I cannot find the values of two burned resistors, marked at PCB as R24 and R59 (left and right channels), one of them it’s broken and the other measures 24?, but the colour of the stripes are changed, may be the value is 18? or 100?
Somebody can help me with this?
Andrew
/ Sep 22, 2016Great article, I’m rebuilding my suspension at the moment and would like to do this, can you re link steps 3 & 6 as they are missing. Thanks.
Anna
/ May 28, 2016Hi. I need help with my Sony Alfa 100. Where to find micro adjustments?
Alan
/ Mar 14, 2016But surely you are supposed to tighten the eccentric nuts with the weight of the car on the wheels, otherwise the arm is always twisted in the rubber bush.
Pete
/ Feb 27, 2016looks great, but could you please post the missing steps?
sephamorr
/ Feb 14, 2016In El Capitan (at least on my system), this fix doesn’t work, as the ssh config location has moved. Simply add the fix to /etc/ssh/ssh_config instead of /etc/ssh_config
Pete
/ Jan 21, 2016Great posts Mike,
The links to steps 3, 5, and 6 appear to be missing, or am I missing something?
Mike
/ May 28, 2016Thanks Pete. For a reason I can’t explain, the link table on the first (default) page was missing one link, to the “Calculate adjustments” page. I fixed that. The remaining missing steps 3 and 6 are, in fact, missing! Sorry I just have not had the time/inclination to complete them. I figured the key steps were published and people could “fill in the rest” but that’s probably just laziness on my part. Thanks for the nudge. I will try to get to them.
Enci
/ Dec 17, 2015Thanks very much. Very helpful
Bill W.
/ Aug 26, 2015Very helpful, thx!
Sam
/ Jul 16, 2015Thanks for sharing this! It’s helpful!
sebastian
/ Jun 25, 2015Genius!!! thanks a lot!!! That was a very annoying bug
Alexandre Campos
/ Apr 5, 2015Hi
I have a Sony A700 and I had to calibrate the sensor recently because of a back focus problem. After I’ve done this, the focus is very sharp but I now have a new problem. I can’t focus on infinity using wide angle lenses. Even on manual focus. I’ve tried different lenses with the same result. It seems that the lens stops in a break before it can reach focus. Can you help me on this?
Curt_C
/ Feb 9, 2015Nicely done, Mike.
Looking at your drawing it looks nearly identical to the PRO2400 schematic, but without the balanced input and level control board. The PRO2400 manual is available on HiFi Engine.
C
MisterG
/ Jan 9, 2015Brilliant solution, and exactly what I was planning when I found rsyslog slowly killing server performance.
If you are interested, there seem to be 2 issues that cause this problem:
1) syslog using TCP so *.* @@serverip -> this says you MUST get a response in order to keep logging and can also be related to the main issue below
2) Syslog has a default of “NEVER LOOSE ANYTHING”, so if it runs out of queue space etc, is starts slowing everything down by not responding.
It is by design, for places like banks who would rather have everything logged and slow performance than “loose logs”.
So you have a few possible options:
1) use UDP for syslog and risk some loss there
2) Configure syslog to drop some logs should it become swamped
3) Configure syslog to use cache, logging etc better
4) don’t send log heavy stuff through syslog
Again though your solution solves the issue, and for servers where performance and uptime is more important than logs a great fall back even if you do all/any of the above.
Thanks againa
Jeff
/ Nov 28, 2014Thanks, this has been annoying me for years!
DavidT
/ Nov 24, 2014Thank-you for your patience and effort to trace-out the 240 schematic. It is very helpful. (and tedious, I know!)
It seems the 2SK163 could be a JFET, which is used sometimes instead of the trusty 2SK170 as a complement to the 2SJ74, when a higher working voltage is required.
The 2SK1636 often shows up in a search as a 2SK163, with the final “6” dropped for some reason. The 2SK1636 is , indeed, a medium power MOSFET; but this would be unusual in a symmetrical and balanced diff input stage…
The 2SK163 JFET can be purchased here, as of 11/24/2014:
http://www.ampslab.com/components_fets.htm
Thank-you for sharing the 240 schematic, and hopefully this note helps you, as well as those who land here…
Kind Regards,
David
VittorioC
/ May 10, 2015It is confirmed that 2SK163 is a JFET. If somebody is still looking for the datasheet, he can find it here:
http://www.datasheets360.com/part/detail/2sk163/-2713613957636864927/
Dave
/ Nov 14, 2014I experienced the same problem with XQuartz and Mavericks and this seems to have solved it. Great detective work!
Anthony
/ Oct 30, 2014Had same issue with Cygwin. Thanks very much
Liz
/ Sep 26, 2014Thanks so much!
Lowell
/ Aug 7, 2014Saved my day!
Rodney
/ Jul 6, 2014Thanks for the work-around. This has been bugging me for months, couldn’t find a solution.
Wei
/ Jul 2, 2014I was troubled by the same problem. Thank you very much !
Eric
/ Apr 30, 2014Thanks! That timeout was really annoying.
???????
/ Mar 21, 2014Hello! I just wanted to ask if you ever have any
trouble with hackers? My last blog (wordpress) was hacked and I
ended up losing a few months of hard work due to
no backup. Do you have any solutions to protect
against hackers?
Mike
/ Mar 22, 2014Basic stuff: use a high-quality provider, use high quality plugins for spam/hacking/etc., update WordPress frequently, use different user names from the defaults and complex passwords.
Antonio
/ Mar 7, 2014Great fix ! Thanks a lot sir, you have earned my gratitude and you have avoided me a big headache.
Rob
/ Feb 24, 2014Great detective work! The 20 mins that Matthew refers to is because there is a commented out line in the ssh_config file for the ForwardX11Timeout 1200 which must be in seconds dan default figure. I just uncommented that line and changed the figure to 596h. Thanks as this was really bugging me trying to test a python app.
Max
/ Feb 9, 2014Hi, thanks for the guide! I have been here before, but didn’t have much luck with adjusting the AF module (it did work perfectly only with the lens I was using to fix it and very bad with the others).
I came back here recently, modified your protocol and added few more control steps here and there and now all my lenses work perfectly!
Thanks for sharing your version, it’s been a very good starting point and to know what the screws do in the first place! 🙂
kat
/ Jan 29, 2014thankyou soooooo much!! sanity saver
jerry
/ Jan 9, 2014Thanks I had same issue with cygwin, trying the same
Steve
/ Dec 30, 2013Love the writeup – any ETA on parts 3 & 6?
Nik
/ Dec 17, 2013Thank you. I was not looking forward to resolving this, you likely saved me hours of frustration.
Jeremy
/ Dec 13, 2013I followed the instruction here to adjust the AF module of my Sony A55 and it was a great success. Now my Minolta 50/2.8 Macro focusses very precisely.
An alternative method to that described is to just adjust left and right until both are in focus. You will then find that the center is in focus too.
Joonas
/ Oct 27, 2013BIG thank you for this! You should be awarded a medal!
I was prepared to toss my a580 with serious backfocusing issues but with these instructions I managed to get the camera tack sharp, irrespective of lens. Sony service could not do anything about it.
A580 has screws in the bottom under the sticker in the middle, but otherwise the same procedure. Found the latter procedure much better, i.e. screw all 3 fully in, and then 1/2 turn open and then start adjusting. Mine was an additional 3/8 turns on every screw.
Same problem with my old alpha 100, will adjust it next.
Mike
/ Nov 28, 2013Great to hear this was useful!!
Gary
/ Sep 20, 2013Mike,
I joyfully (sorta) found your article on back-focus problems with the Sony A55. I too am a lucky owner 🙁 of a A55. I also own a Panasonic LX -3 which I dearly love.
A few week ago, I was scrutinizing the same landscape shot with both camera. The A55 lens was a Tamron 10-24mm f3.5 set to 10mm, aperture pryarity mode and stopped down to f5.6, image quality: fine, image file: large, steady shot off, auto focus: automatic AF, AF area: wide, ISO 100, HDR off, files saved as jpgs, cameras were tripod mounted. The LX-3 was set to approximately the equivalent settings.
After making the photos, I opened them and looked at the two images. The LX-3 was tack sharp, in fact, it made hash out of the A55 in just about every respect: contrast, dynamic range, sharpness, etc. Here is a little camera, with a small chip costing 1/4 the price of my DSLR and pounding it into the ground. The A55 was soft except for the extreme right side of the image. The left was very soft, but there looked like more was going on. There as more chromatic aberration with the Tamron lens, which I expected, I also expected some softness in the corners, it’s a Tamron . . . and a 10mm. But this didn’t seem right. The camera has never been dropped, so I ruled out physical damage by me. I know that Sony cameras using a prism instead of a mirror can be prone to “ghosting” but I wasn’t about to let this slip by.
I began searching and found your article, then read Jeff Fridel’s article and printed his focus charts. I’m ready to begin your process of aligning the focus sensors, but before I start (a process I’m not looking forward to) I wanted to get your opinion. I tried my other lens, a Sony 18-55mm just to make sure it wasn’t the Tamron. It was a tad better, but still with all the same basic defects as the Tamron. I read the reviews in DPReview of the Tamron before I bought it and the comments were very favorable.
Given your experience, do you think I have a miss aligned focus sensor? I manually focussed the image using the focus magnifier and it looked go. I was a photo major at RIT so I know a bit about lens aberration, mirror misalignment, etc. But my training is 1980’s vintage. film cameras are gone, people don’t even know what view-cameras are. It’s a new photography world.
If you have time and are willing, could I send you some images to look at?
Thanks in advance for your help,
Gary
Mike
/ Sep 20, 2013A misaligned focus sensor module would not explain what I believe you are seeing, which if I can paraphrase is a difference in focus across the frame on a single exposure, i.e. sharp on one side, soft on the other. A crooked focus sensor would affect all parts of the frame the same way. I would have to assume that either your sensor is crooked, or the lens mounting is crooked, or the lens itself is crooked within the barrel.
Sergii
/ Sep 6, 2013Thanks for the article. I have A55 and recently bought Minolta 50/1.7. Several tests showed that I have a front focus with this lens. Before that I used 18-55 kit and Minolta 35-70/4 lenses. They are not that fast, so I did notice the problem. Anyway, I’ve tuned the AF, thanks to your article, and here are my observations:
1. After tuning the AF for 50/1.7, I’ve tested other two lenses and there seems to be some back-focus problem on 18-55 at 18 (at 55 it is OK), and there is a huuuuuuuge back-focus on 35-70 at 35. Strange…
2. I’ve first tried to play with left and right bolts with no success. Then I’ve tried to screw them (all 3 bolts) tight and release half of the turn. This position was used as a reference point and what was really strange – AF did not react on those two bolts (left and right). But I was able to move focus plane front and back using the third bolt (Focus height adjust). I do not know what is wrong with my actions, but I’ve tried several times to adjust it with left and right bolts – no success. When I turned left or right bolt too much, it would stop focusing at all, I had to compensate it with “Focus height adjust” bolt, which also affected the front/back focus, which makes it even more difficult to adjust something with left and right bolts.
Do you have any ideas why it was like that? I’d expect it to work in a different way: left and right bolts are used to adjust front/back focus, and third bolt is used to move sensors up and down (so that actual sensor position will match rectangle in the viewfinder).
Steve M
/ Sep 1, 2013Thank you for this excellent article and explanation of how to adjust the auto focus. Worked well on my SLT-A33. Fairly easy to do and it sure beats sending the camera into Sony with only so-so expectations of a good result.
Somehow, I had developed what I suspected to be a serious AF problem. Using the autofocus test chart from Jeffrey Friedl’s Blog ( http://regex.info/blog/photo-tech/focus-chart ) I was able to confirm that a pretty large amount of front focus was the problem. I didn’t have time to do a precise correction, but did a rush job before going to an indoor volleyball shoot and had a significantly better time getting usable shots. I’ll go back and fine tune things next, but I just wanted to go ahead and say now how much I appreciate the help provided here.
Gerrit
/ Jun 16, 2013That would be good too, as I tried searching Google docs to find it.
Your method appears to be the most fool proof and easiest of the ones I’ve seen so far.
Gerrit
/ Jun 15, 2013Hopefully you’ll publish Pt 6 first, like today, as my front alignment is awfully off!!
Mike
/ Jun 16, 2013Yes, it’s coming :-). More important though is the spreadsheet posting which has been some time in formulation, but I think it’s close to ready.
Tony
/ May 10, 2013where are the other parts? I can only find 1,2 and 7…
Mike
/ May 10, 2013They are coming — slowly….
Mike
/ Jun 22, 2013A few more parts have been published, now parts 1, 2, 4, 5, 7 are available. Missing are camber measurements (part 3) and front wheel adjustments (part 6). The most complex parts are all published. 3 and 6 are still coming….
xdeimos
/ Apr 25, 2013Hi, Im little bit confuse with ur post
1. do u have to access /etc/ssh_config through ur mac?
I do not have that path though
2. what do u mean u can also try ssh -Y
please email me with this issues
thank you
Mike
/ Jun 22, 2013/etc/ssh_config is available and has been at least since Lion, I believe (and maybe before that), but of course you must have administrator permissions to edit the file.
You can type “ssh -Y” to get around this whole problem, but there may be some security issues with that. I would read up on the tradeoffs. Excerpt from the ssh man page for the -Y option says: “Enables trusted X11 forwarding. Trusted X11 forwardings are not subjected to the X11 SECURITY extension controls.”
Alan
/ Apr 8, 2013Wow, this was so frustrating. Glad I finally found your tip.
Rick
/ Feb 24, 2013I just got my A 55 back from Sony (Feb 2013) and the dumb asses are still saying no problem found. These people are pathetic at best. You would think after three years of experience with the well documented back focus problem they could hit their ass with both hands, but I guess not.
Mike
/ Jun 22, 2013No kidding. It’s so blindingly obvious if they would just use the far left and right local focus points in the test, in addition to the wide area which is the factory default setting.
Dennis C
/ Feb 18, 2013It’s taken me too long to find you! Thanks for a great post.
My CZ 16-80 appears to have more backfocus than the original kit lens. Now I’m going to spend some time tweaking
David Ruiz
/ Feb 10, 2013Thank you for nice article! I have the same problem with my Sony A380. It’s so irritating! Newer had problem like this with it before, so i just bought a new nikon D3100.
Roland
/ Jan 30, 2013Thanks for the post. I just ran into this installing a remote Oracle database and it kept failing half way through because it could send out the display. Nice job!
John
/ Jan 23, 2013It’s not a bug, it’s how OpenSSH works. and you don’t have to set
ForwardX11Timeoutto something stupid long, just setForwardX11Trusted yesin your~/.ssh/configor/etc/ssh_configfilesCraig Macdonald
/ Jan 16, 2013man ssh_config details ForwardX11Timeout:
Interestingly, I couldn’t put the option in my (NFS shared) ~/.ssh/config, as some linux ssh (e.g. OpenSSH_4.3p2) didn’t recognise ForwardX11Timeout.
Jyri Virkki
/ Jan 11, 2013Thanks!
Riaan Roux
/ Dec 27, 2012I have another issue. On my A55 when I zoom I have back focus, but when full wide I get front focus. I have tried this with a few lenses. I have calibrated mine for the Sony SAL 28-75 f/2.8 at 50mm as it is the lens I use the most, but I am still unhappy. Zooming to 75mm will give me back focus, but one wide (28mm) I get front focus.
Any ideas?
Mike
/ Jun 22, 2013While I can’t answer your questions specifically, I recently went through a little more learning about the AF system theory and construction and found an excellent article on understand autofocus: http://www.clarkvision.com/articles/understanding.autofocus/. This may or may not be the case for the a55 but there are different groups of AF sensors that can be used depending on the f/ number of the lens on the camera. Wider apertures (f/2.8 or lower) are supposed to able to take advantage of AF sensors further apart which therefore have better distance discrimination and are faster and more accurate to focus. I am wondering whether maybe f/2.8 is right on the borderline and at one focal length extreme the f/2.8 sensors are used and at the other they are not.
Another thing I am noticing with my lenses after revising this post is that not only does AF accuracy differ from lens to lens, but also for different focus distances. I tuned my AF sensors for a Minolta 135/2.8 and it reliably and consistently snaps into essentially perfect focus at a range of distances from about 2 meters on up. I have an old, cheap Minolta 50/1.7 that shows significant front focus when the test chart fills the field, which is at about 0.7 meters distance. The front focus reduces as distance increases, but never quite gets to accurate focus before the focus mark in the center becomes too small and the AF sensor starts picking up the patterns on the sides of the chart to focus on.
Peter
/ Jul 24, 2015Dear Riaan, did you ever manage to sort out what was wrong here? I have exactly the same problem, except I get back focus when fully wide and no problems when fully long.
Greg
/ Nov 7, 2012Thank you very much for this post. I have been fighting this exact issue since I was forced into a 10.7 upgrade in September. I had the parameter name correct, but kept trying to make it work with 30d.
Mikey
/ Sep 28, 2012Awesome article – I was able to fix the backfocus on my a33 a few months ago.
Just got an a57 and I’m going through the backfocus issue again! Any ideas on how to adjust the focus sensor? There doesn’t seem to be any access panels near the flash anymore and I’ve tried removing some of the top screws but I still can’t find those adjustment screws.
I’m reluctant to bring it back to Sony. They kept my a33 for 2 months just to tell me there were no backfocusing issues.
Aaron
/ Aug 8, 2012I just stumbled upon your EXCEPTIONAL post concerning the Sony SLT-a55 focus issue. This was after having visited my local camera shop Sony expert today for the same problem; and leaving with the feeling that I should consider sending my a55 to Sony for repair. After reading your post and others’ comments, I think that would be a waste of my time and money! So, I think I want to try performing your adjustment myself — as I apparently have the same problem you experienced with your camera.
I looked at my camera and tried to determine where the two covers are under the pop-up flash and where to pry to pop them off so I can access the three adjustment screws. It’s not obvious to me. Could I impose on you to describe in detail how to do this — so I hopefully won’t damage my a55 while trying to fix a problem I should never have had?!!
Thanks in advance for any additional guidance you can provide — AND, for the EXCELLENT description of the problem, solution, etc.!!!
Aaron
/ Aug 9, 2012Oops! I must have stayed up too late last night. Re-reading your original post this morning revealed that I had somehow missed your link to the DPReview.com post that showed pictures of the cover removal process. As a result, feel free to remove both of my comments to eliminate the clutter.
Many thanks again!! I feel you have done ALL Sony a33/a55 users a HUGE favor in making this information available!
Mike
/ Jul 10, 2012I *REALLY* wish someone actually finally went ahead and peel the darn shell of the body of an A57 or A65 and show the world where in the bloody hell those darn adjustment screws are. It’s *REALLY* getting annoying at this point !
Karl
/ Jul 5, 2012Thanks!
Gabri
/ Jul 5, 2012Many thanks for your solution!
Histrionic
/ Jun 19, 2012I’m also interested in more details or pictures of the covers that were removed to access the adjustment screws.
Eric Stoiber
/ Jun 10, 2012I bought a A65 in november of 2011 and by december it didn’t turn on then the thing was sent in and with in a month the batteries were draining all the while talking with sony tech’s and having focusing problems which they said was me then finally in late march did some testing and sent the tech tewst pictures they told me to send it in with the new lens i had bought 70-400G and i got it back 5 weeks later and it was just as bad and contacted them again and sent in test .Now they have it again and we will see . But 6 months of provlems with a brand new sony A65 and a brand new 70-400G and talking on my cell cost tons of money you would think they would get it right and compensate people some how for their mistakes. I think i will run the tests when i get it back right away again and if it has problems i think i will sell all my sony stuff off and go to nikon what a shame . For right now i’m using my A330 and get better pictures then with the a65 and my 70-400g lens wow that says something and they don’t sell DSLR’s any more only SLT”S , Bad move.
Mike
/ May 14, 2012A65… problem worse I’ve yet seen. Makes the 24mp sensor entirely pointless. Yes, it’s *that* bad. Every single photo spoiled by back focus !
Bob
/ May 10, 2012Thanks so much for that fix!
Cindy
/ Apr 24, 2012I have had this same issue with my A55 since I got it. I have an A200 and never had this problem.
Now the camera is out of warranty. Can you suggest where to take it for repair since I am not very savvy when it comes to doing my own adjustment?
Nancy
/ Apr 13, 2012I have an a33 with multiple focus issues.
First, there are times the image is very sharp, and other times, it is less than satisfactory. I have tried different focus modes (tend to prefer single auto focus so that I can frame the image after selecting a focal point).
Second, when focusing on faraway objects, many are out of focus.
Third, when using manual focus for distant images, I have to back up a smidge from the infinity setting.
Are these issues common to the focus problem you are describing here? And where would you recommend a novice take a camera to to make the kinds of adjustments addressed here. Thanks.
Mike
/ Apr 13, 2012It is hard to say whether you are seeing the same issues I’ve described without some testing along the lines of what I did. The default (wide area) auto focus mode takes focus readings from a number of sensors and selects one that it thinks is the subject. From the Sony rep I talked to, the most likely selection mechanism is to use the one corresponding to the sensor that sees the closest part of the subject. But it adds uncertainty into the process. One hint is whether the default focus consistently chooses a sensor that is off-center and does not seem like the right selection.
From a big picture point of view there are two possible significant focus issues. One is that the focus sensor block is mounted straight in the camera but the whole sensor is not correctly adjusted, so no auto-focused images will be sharp. The other is that if one of the sensors is correctly located but the block is crooked (that was my problem, and this kind of misadjusted sensor passes Sony’s focus test). To figure this out you will probably need to use local area focus and try focusing using the leftmost sensor, center sensor, and rightmost sensor and see if there is a consistent pattern at each sensor. In my case I finally figured out that using the rightmost sensor always produced sharply focused images, but using center or left sensors produced poorly focused images, and the further left the further out of focus.
I don’t know if this has anything to do with needing to back up a smidge from infinity using manual focus, since the auto focus is not used at all and it may have to do with the lens. My guess is that most lenses are designed to allow focus to go slightly past infinity so that normal manufacturing tolerances and wear over time will still let the lens focus at infinity without running out of focus range before infinity is reached.
Jim
/ Mar 7, 2012Thanks for the info.
I had my A55 in to Sony for something else asked them to take a look at the focus. They fixed the other problem but didn’t even respond to my comcerns about focus. Now I’ll send my camera to someone else to get it fixed and try to get Sony to reimburse.
Mike
/ Mar 7, 2012Matthew, thanks for the info. I’ve entered the bug at FreeDesktop.org: https://bugs.freedesktop.org/show_bug.cgi?id=47053.
Matthew
/ Mar 6, 2012Surprisingly, I still see the crash if I use XQuartz 2.7.1. http://xquartz.macosforge.org/trac/ticket/553 was opened, but closed with a message it should be filed at freedesktop.org. a search there for “ForwardX11Timeout” returns no results, so I’d guess it’s not bee reported to them. I was going to file it myself, but I don’t care enough to create an account.
Jack
/ Mar 1, 2012Can you provide guidance or pictures of the covers that you removed to access the adjustment screws?
Thanks, Jack
Matthew
/ Mar 1, 2012Have you filed a bug with Apple? if so, could you share the rdar?
I’ve confirmed that it is a 20 minute timeout, and that ‘ssh -Y’ does appear to avoid the problem.
Kashan
/ Feb 21, 2012Outstanding and in depth analysis, I am facing the same issue with my A55, I did give my camera to Sony repair center, they updated the software but it has still not solved the issue, As I am not as technical as you, I have now your email to back my concerns.
Thank you
Russell Lees
/ Feb 4, 2012Nice atricle was going to write one just like it as I experienced the same situation with the alpha 580, such a great camera (when you sort this little problem out) but Sony really letting themselves down by not paying enough attention to the autofocus allignment.
On the non SLT models like the a580 the screws are at the base of the camer under the L shaped sticker.
Janez
/ Dec 13, 2011I have same problem on Sony a65. Does anyone know where AF focus unit adjustment screws are on sony a65?
Mike
/ Nov 28, 2011Same experience. Sony, give us micro-adjustment in all cameras so we don’t have to deal with support centers unwilling to spend the time it takes to properly adjust the AF.
erix
/ Oct 25, 2011Thanks for the article. I have A33 and exactly the same situation. Camera has been sent to service center and today received it back with “we could not repeat the problem”. Back focus is pronounced with wide angle – 18mm with kit lens or 28/2.8, but at the same time focus is perfect with 50/1.7 lens.