Tamron 17-50mm f2.8 non VC Wobbly lens

[rbarry]

I have owned this lens for about a year. It was a second hand Ebay purchase, described as being in excellent condition. This was true other than for a small amount of play in the end of the lens, which wasn't made known to me during the Ebay auction.

I was aware that this particular lens from Tamron is prone to this problem, but understood that it can be easily fixed. The "fix" involves removing the dust seal from the front of the lens which then exposes three screws underneath that become loose. Just tighten these screws down and re-assemble the lens. see here: http://www.youtube.com/watch?v=IiOj-7MAO_w

Now I'm not so sure that this is the answer to the problem. What has moved to allow these screws to become loose in the first place? If you tighten the screws and re-seat them are you effectively changing the lens position and possibly altering the focal plane of the lens?

I love this lens, it's done me well and can produce good sharp images. I have no qualms in the straight forward dis-assembly of the dust cover. However, I am reticent to go further without some explanation as to why these screws have become loose and what are the consequences of reseating them. For example, could I end up altering the front/back focus setting of the lens?

Thanks for reading,

Rick.

[Markulous]

Originally Posted by rbarry

Now I'm not so sure that this is the answer to the problem. What has moved to allow these screws to become loose in the first place? If you tighten the screws and re-seat them are you effectively changing the lens position and possibly altering the focal plane of the lens?

Well the lens moves back and forth and presumably the vibration of the lens motor loosens the screws. I see one of the comments mentions avoiding rotating the lens which would alter the AF - presumably the plane of the lens isn't altered by this mounting.

I'd say go for it - after all the lens is not particularly useable if it wobbles and it'll only get worse

[rbarry]

Thanks for the reply Markulous.

I had a look at the lens over the weekend. It seems that the front element sits on a plastic ring that will move with the glass when the front element is rotated. This ring in turn is seated in such a way that when it is rotated with the glass, it will physically move the element out if rotated anti-clockwise, and the reverse if rotated clockwise. I would guess that the total movement will alter the the postioning of the lens by about 3mm.
When the lens is rotated you can see it comes up against a stop, in either direction of rotation. The total amount of rotation relative to the diameter of the lens is under 20%.

When I first looked at the lens with the dust cover off, I was fairly sure the lens was at the end of it's rotation, but I can't remember whether that was anti-clockwise or clockwise! I have stated before that I love this lens for it's sharp clarity, but I can't say that it crossed my mind to ever check for any front or back focus issues. I haven't been able to find out for sure if this movement in the front element has any bearing on front/back focus adjustment, other than unsubstatiated reference to the effect that it does on snippets of information I have gleaned on the internet.

Calibration:

I have printed off an Amsler grid (like graph paper) onto a sheet of A4 paper. Starting at the centre of the grid, I have inserted the numeral 0. All points immediately above and below, and left and right of the 0, I have added the numbers 1 through 10. This grid is then mounted on a cork board with the aid of pins in the four corners of the paper grid to keep it taught.
I intend to place this contraption at 45 degrees off plane and take a series of pictures, focusing at the centre of the grid on the 0. I will start with the moveable element fully clockwise up to the stop, then counter clockwise up to the stop, then centred, then a 1/4 of the way, followed by 3/4's of the way. I hope that sudying the results should give me an idea of where the best position to tighten the element in place will be.

However, having tried a few dummy runs it seems that the results I'm getting are anything but consistent. I think there must be an optimal distance between the lens and the grid that I'm using, and at an optimal focal length, and an optimal, or maybe a series of different aperture settings. My grid pattern is 5mm x 5mm for each square.

Does anyone have any ideas or advice?

Many thanks for taking the time to read this

Rick.