LensRentals.com

Panasonic Pancake Dissections

Posted by

Panasonic makes an excellent pair of micro 4/3 pancake lenses, the 14mm f/2.5 and 20mm f/1.7. While they look similar, there are several differences.

The optical formulas are quite different, for example.

 

Optical diagram of the 20mm (left) and 14mm (right) pancake lenses

While the performance of both is excellent, like most pancake lenses there is a bit of softness in the corners wide open.

MTF charts for the same two lenses

 

They also differ in focusing method: the 14mm is internally focusing; that is one internal group moves when the lens is focused. The 20mm focuses by moving the entire lens assembly, making the lens physically longer at closer focusing distances.

Both, however, are 'focus-by-wire', meaning that when you turn the focus knob you aren't actually focusing the lens; you are sending a signal to the electronic motor that actually moves the focus element. Probably because only one group is moving, rather than the entire lens, the 14mm has the reputation of focusing more rapidly and being quieter during focus.

There is one other difference that has gotten some attention lately. The 20mm lens is known to cause banding when used on the Olympus OM-D camera. I entered a discussion about this on DPReview, and ended up volunteering to open the lenses and see if we could find anything that might contribute to the banding.

If you don't like reading geeky stuff and looking at undressed lenses, the answer is we didn't, so you can move along now. For my fellow geeks, though, let's have a look inside.

Open Them Up Already

Looking from the mount side, the lenses show a lot of similarities. Same 3-screw light baffle, same 3-screw mount, and the same electronic connector (DUH).

 

20mm (left) and 14mm (right)

 

Removing the light baffle led to our first surprise, and we've now renamed the 20mm lens The Jack-In-The Box.

 

The spring has something to do with the fact that the lens assembly extends for close focusing. Whether it helps the motor move the lens, or the motor only moves the lens back and the spring moves it forward, I can't say. But if you put the lens back together without the spring, it doesn't focus. (Pretty simple way to figure it out, huh?)

OK, back to work. Removing the mount is straightforward and shows the rear barrel assemblies are about identical except for size.

 

Three screws take the rear barrels off, fully exposing the PCBs (main circuit board). You can see the focusing motors for both lenses in similar locations, although quite different in size, as you would expect. (Remember, the 20mm has to move the entire lens, not that it's a huge thing to move.)

One significant difference does show now. Notice on the 20mm (left) the toothed pattern along the manual focus ring. The 14mm has a similar arrangement, but it's a layer deeper. As we'll see later, the teeth pass through an optical sensor that then sends focusing signals to the AF motor when you turn it.

Here's a close-up of the sensor and teeth in place. The flex cable connects the sensor to the main PCB.

 

Removing the PCB cleans things up a bit but doesn't really show anything new.

Internal view of 20mm (left) and 14mm (right)

 

The 20mm, on the left, shows the large focusing motor in the upper right section and the top of the smaller aperture motor at about 4 o'clock from the lens elements. With the 14mm we can also see the focusing motor at the top of the image, and 3 screws that are next in line for removal.

Removing the 3 screws from the 14mm lets us remove all of the optics and motors in one piece; they drop right out of the housing. If we flip it over we're looking at the front element. Now we can see the focusing motor at the top, and the brass gear of the aperture motor. The screws tell us there's some further disassembly we could do, but that would almost certainly involve taking out the aperture blades and other things that are really time consuming.

The 14mm internals

 

At this point, though, I notice the motor bodies and their electromagnets are towards the front of the lens. I really doubt the motors have anything to do with the banding reported with the 20mm. If they did, though, the smaller motors put toward the front of the lens might be helpful in preventing that with the 14mm.

As an aside, now that the main lens assembly is removed we can see an identical optical sensor and toothed focus ring, like the one on the 14mm. It's simply set forward and inside the lens case, so it's less apparent from the back.

Optical sensor from behind.

 

20mm Internals

The 20mm has another 'something different' for us: the focusing ring is held on by a large retaining ring that easily unclips, letting us slip the external focus ring off.

20mm with retaining clip removed

Which then lets us remove the focusing motor. It's rather unremarkable.

 The main part of the lens still sits in the lens housing.

Notice the small, square posts with holes in the center that are set in the focusing ring; you can see 4 of them in the image above. We can lift the central part of the lens out of the housing now, which shows us the grooves in the focusing ring. Pancake lenses don't have helicoids; the lens slides forward and backward during focusing because when the ring rotates, curved grooves inside move the lens backward and forward.

 

As with the 14mm, we chose not to disassemble the internal elements. There are no apparent optical adjustments that can be made and disassembling the diaphragm is time consuming.

Conclusion

I don't think our little dissection shows anything that is directly attributable to causing banding, because I would expect it is some electronic component on the main circuit board, which is well beyond our capabilities to analyze. The 20mm does have a larger AF motor that is less well shielded, but I doubt this has anything to do with the problem - the motor isn't running when people are taking photographs.

Roger Cicala

Lensrentals.com

October 2012

10 Responses to “Panasonic Pancake Dissections”

Joachim said:

To me, the spring looks like an antenna. If the camera is set into continuos or automatic decision whether focus method to use, it could be the motor is powered, although not moving. You can use a motor as a brake as well. this kind of waves are difficult to measure, especially in a small device like a camera. Does banding happen with all focus- and ISO-settings? Higher ISO is an amplification of the sensor's signals. And light is also just a kind of waves in a different spectrum. Interesting to see how much electronics and mom-optical stuff they packed into a small volume.

Sean Esopenko said:

I noticed the COPAL stamp on one of the AF motors. Do you guys see that stamp on camera parts often? I have view camera lenses with "Copal" shutters. Is there a correlation?

A said:

Actually I'd say the motor could very well be the source - a poorly chosen/constructed motor can kick out a fair amount of noise on a whole host of frequencies... Banding basically *is* noise.

As a matter of interest, is the "mirror(less) box" on the OMD any wider than other similar m43rds cameras which don't exhibit banding? Or does it have more plastic where there's metal in the front of the body.

Also, I wonder whether a lens (or camera) gets FCC tested in isolation, or when used with a particular camera/lens?

questionlp said:

Nice to see these lens teardown articles.

NB: The link to DPReview is broken.

David said:

Those ribbon cable connectors are conveniently close to the rear lens mount. I'm curious if you could disconnect one (say to the focus motor) and test the lens for banding, or if this would just give you a lens error. Not as much fun as dissecting a lens, but this kind of troubleshooting can be fun too if you have the patience and equipment (would need an oscilloscope). I agree with A that the motor could be the cause, bet it is powered during focus lock. The motor would be easier to isolate than a component on the PC board, if you have the patience to try it out Lend me an OM-D and I'll help you out!

As an aside, I once had (still have actually) a turntable that developed a bad rumble (low frequency hum), and it turned out to be a ribbon cable was the culprit. I fixed it using shielded wires jammed manually into the connectors at both ends ( couldn't get a replacement cable). I thought this was a bogus place to use a ribbon cable, especially given the premium paid for the turntable, which was programmable and allowed random play of songs on an LP.

tmei said:

The spring is probably to eliminate mechanical backlash, so that the movement of the lens when focusing does not depend on previous direction of travel. Very thoughtful design!

Michael Jelks said:

There was a thread recently on DPReview in the micro 4/3 forum that seemed to suggest the banding goes away after the lens has had a chance to "warm up". Noise and/or electrical interference was a possibility as well.

http://forums.dpreview.com/forums/thread/3285963

LensRentals Employee

Roger Cicala said:

Michael, that's the thread that started all this . . . .

Michael Jelks said:

This is what happens when reading articles on an iPhone screen. It's hard to keep the entire article in context. :)

Philip Johnson said:

I managed to blow dust inside the front element. Is it possible to go in through the front of the lens?

Leave a Reply