[Fenwoodian]

.
I'm currently working on a project that requires me to shoot a mirror lens. I have not shot one in over 40 years.

I recently purchased a used, like-new 500mm f/8 Kalimar mirror lens. When I put it on my K-1, the shake reduction (SR) went wild, with the sensor constantly dancing and chattering all over the place - even though I was not significantly moving the camera.

At first I thought that there was something wrong with the lens. But when I put other normal and short telephoto lenses on the K-1, the SR returned to working normally.

Then I tried a little experiment. I tested a 35mm lens, then a 50mm, then a 90mm, and finally a 180mm. In each case I had SR on and was gently holding the camera. Well, I noticed that with each longer lens I put on the K-1, that I could hear/feel the sensor moving to a greater and greater extent. So, I figured that with a really long lens like a 500mm on, the SR would move the sensor much more to compensation for the same degree of camera movement. If that's true, I believe that in the future I'll only use the mirror lens with the SR off in order to prevent from "overstressing" the sensor movement mechanism.

Does this make any sense? Have you experienced this yourself? Or is it something about mirrorlenses themselves that make SR mechanisms go nuts?
.

[ismaelg]

Are you entering the focal length at startup?
It would make sense that with a 500mm the SR will be far more active than with shorter focal lengths, but I would think it has nothing to do with it been a mirror lens.

[barondla]

.
I'm currently working on a project that requires me to shoot a mirror lens. I have not shot one in over 40 years.

I recently purchased a used, like-new 500mm f/8 Kalimar mirror lens. When I put it on my K-1, the shake reduction (SR) went wild, with the sensor constantly dancing and chattering all over the place - even though I was not significantly moving the camera.

At first I thought that there was something wrong with the lens. But when I put other normal and short telephoto lenses on the K-1, the SR returned to working normally.

Then I tried a little experiment. I tested a 35mm lens, then a 50mm, then a 90mm, and finally a 180mm. In each case I had SR on and was gently holding the camera. Well, I noticed that with each longer lens I put on the K-1, that I could hear/feel the sensor moving to a greater and greater extent. So, I figured that with a really long lens like a 500mm on, the SR would move the sensor much more to compensation for the same degree of camera movement. If that's true, I believe that in the future I'll only use the mirror lens with the SR off in order to prevent from "overstressing" the sensor movement mechanism.

Does this make any sense? Have you experienced this yourself? Or is it something about mirrorlenses themselves that make SR mechanisms go nuts?
.

Don't think it has a thing to do with being a mirror lens. The SR doesn't know what type of lens it is. It doesn't even know the focal length unless the lens supplies the info via electrical contacts.

I wouldn't worry about wearing the SR out. It is magnetically actuated similar to a speaker. Look at how long it takes to wear out a speaker.

Thanks,
barondla

[micl161]

I have shot both Tamron 500mm (55BB) mirror and Ohnar 300mm mirror on my K1 and they both looks 'good' (sure the SR move more than the shorter lens) with the SR on K1.

I also shot the Sigma 50-500mm zoom on K1 which have the in-lens SR, you can feel the K1 SR move more than the in-lens SR. (I turn off either one) and through the
viewfinder, the view move more rapidly on K1 SR.

[photoptimist]

I just tried a little experiment:

1) put on a tiny little lens (a petite Pentax M 20 f/4), input 2000mm as the focal length, put it liveview, and the SR system certainly does jump a round a lot.

2) put on a big, long, heavy lens (an 80-200 f/2.8), input 2000mm as the focal length, put it liveview, and the SR system wasn't as twitchy, especially if I firmly held both the lens and body.

With SR set to 500mm and a center of gravity only a few inches from the sensor, it only takes a 0.009" dip of the camera body to require the SR system to drive sensor to the limit of travel. But if you think about the chain of events: the camera rotates down, the front of the lens rotates up, more sky shows in the top of picture, but the image is inverted on the sensor, so more sky is projected on to the bottom of the sensor, which requires the SR system to push the sensor down to recenter the image. Next Newton enters the scene with his third law. Whilst the sensor is getting shoved down, the body is getting shoved upwards -- it's that equal-and-opposite force thing. So the SR system sends the sensor down and the body up, the SR system then detects that upward motion of the body, and twitches the sensor in the other direction that which sends the body down again. Depending on the springiness of your grip, it can start to resonate up-down-up-down-......

A big long heavy lens can reduce the effect both by increasing the mass and rotational inertia of the system and by moving the center of rotation farther from the sensor.

Short, lightweight, but long-focal length lenses would be likely to show much more twitchiness.

P.S. This same effect can happen with some tripods and some lenses -- the system can resonate. It's even more likely with a tripod because there's not much damping.

[Fenwoodian]

.
Thanks for the great responses everyone!

I just programmed my f-2 button to control SR. Now when I have the 500mm lens on my K-1, I'll only turn the SR on just before I shoot; and will then promptly shut if off when done shooting. I noticed that just walking around the back yard with the mirror lens on and SR on that the whole while I had the camera in my hands the SR was going nuts. So in the future, I'll only use SR on the 500mm mirror lens when it's actually needed.

[Not a Number]

I haven't noticed anything out of the ordinary using a 1000mm mirror lens on my K10D. But then there have no doubt been a great deal of changes in SR since the K10D was made.

[microlight]

Hm, interesting. I have a Samyang mirror lens that I occasionally use on the K-3II but haven’t noticed this effect. I’ll try later.

[reduno]

Hm, interesting. I have a Samyang mirror lens that I occasionally use on the K-3II but haven’t noticed this effect. I’ll try later.

The K-3 II has a much smaller sensor than the K-1, so you may feel less vibration, even if the sensor is moving the same distance.

[Not a Number]

The K-3 II has a much smaller sensor than the K-1, so you may feel less vibration, even if the sensor is moving the same distance.

Good point. Smaller sensor, less mass too.

[kiwi_jono]

Have not noticed any SR issues with a Tamron 500 mirror on my K-1.
As long as I have correct FL length set at start up (have seen some issues with wrong FL set - but even then its usually just ends up with a bit blur to the images).

Is there anything loose in the lens (would imagine that a loose element or center reflector could cause this)?

[reduno]

Is there anything loose in the lens (would imagine that a loose element or center reflector could cause this)?

I would not expect a loose element to confuse the SR system, since I believe it is based on detecting movement of the camera body and moving the sensor to compensate. If it was relying on the light path through the lens it could potentially be a different story.

[kiwi_jono]

Very good point. You are right and in fact lens should not have any influence at all.

[Gimbal]

I haven't noticed anything out of the ordinary using a 1000mm mirror lens on my K10D. But then there have no doubt been a great deal of changes in SR since the K10D was made.

The big difference here is that it is only the later models (after the K5 or so) that has SR active in such a way that it moves the sensor as soon as you halfpress the shutter button. The earlier models did not activate the sensor coils until the mirror started to flip up, thus there is no active movement inside the K10D until you actually take the picture.

[fs999]

Looking at the title, i've decided to use less mirror lenses