Anatomy of the Leica M8

[albertknappmd]

Advertisement (gone after registration)

An incredible anatomy lesson.... far more interesting than those in Med school....

[minglancelot]

Great Article！

 

Here is the chinese edition:Leica M8¡°½âÆÊ¡±È«Í¼½â - Leica»úÉíÎü¶¾Çø - LeicaÖÐÎÄÂÛ̳ Leica,RangerFinder,Leica M8,Leica M6,Leica M7,Leica M3,Epson R-D1,Summicron,Summilux,Noctilux...

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[sol_summilux]

amazing!!!

thanks to sharing.

sol:)

[Guest sirvine]

amazing!!!

thanks to sharing.

sol:)

 

Finally another Sol around here.

[eudemian]

I was so taken with Mark's skill that I have done a collage of double A3 size and collected most of the pictures taken and displayed them on my wall.

 

I find it fascinating and inspiring to look at. Leica does have a habit of attracting oddballs so I think I must have joined the club.

 

When you view all the photo's together it is a really amazing dissection lesson, and an amazing technoligical feat.

 

Tom

[Fotogo]

Mark, Bravo.

 

I enjoyed your reviewing your photos again, and numerous comments on all the issues that were raised.

 

I have a rather simple question about construction of the camera. Can you advise me what is holding the Hot Shoe frame in place to the camera's top? Many thanks

[jlm]

Advertisement (gone after registration)

another ? for Mark:

 

any ideas about the gritty shutter button action? lubrication, for example?

[marknorton]

The hot shoe is held in by 4 screws from inside the top cover. Three are underneath the flex print and one goes through the flex print and makes contact with it. The hot shoe contacts are soldered to the flex print and the hot shoe is also held in place by silicone sealant. Not sure why this should be - might be insulation, might be anti-tamper, might be to make the camera more water-tight.

[marknorton]

John, in an earlier post, I showed the shutter release switch with 4 contact balls and springs which are quite weak and are there to make contact, not profile the release action. This is done by the two adjustable detents.

 

The feel depends on how far in the screws are - the further in they are, the greater the resistance to the shutter release moving - and the profile of the plunger inside the switch which you cannot see.

 

When I re-assembled the camera, I backed off the screws by half a turn which has given me a smoother - if slightly less well defined - release action. I do not think a lubricant is the answer because it could interfere with the electrical contacts. Coating the tips in PTFE might be interesting.

 

One part I didn't show was a tiny plastic disc which fits inside the top of the shutter release switch and is there to cushion the contact between the shutter release and the switch and prevent the feel of metal contacting metal.

[ho_co]

Mark--

Any idea where the orientation sensor is or how it functions?

 

Thanks!

 

--HC

[Paul Hart]

Absolutely fascinating. Thank you for the time and expertise put into this.

 

Note to Leica: "I think at least some of the excessive shutter noise comes from vibrations from the motor being amplified because it is directly mounted on the camera body. A compliant mounting to reduce transmission might reduce the noise."

 

Note to self: if you see M8 serial number 3105477 on eBay, give it a miss...

[wlaidlaw]

Mark--

Any idea where the orientation sensor is or how it functions?

 

Thanks!

 

--HC

 

Probably looks something like this:

 

Wilson

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[marknorton]

Note to self: if you see M8 serial number 3105477 on eBay, give it a miss...

 

You won't, it's my daily camera...

[marknorton]

Mark--

Any idea where the orientation sensor is or how it functions?

 

Thanks!

 

--HC

 

HC, interesting question!

 

We can hear it rattling of course but I have no idea where it might be.

 

I imagine it's something like a gold ball rolling around in a square of contacts inclined at 45 degrees and parallel to the plane of sensor (so, for example, mounted flat on the main DSP board); depending on the orientation, the ball makes contact with two sides of the square. I'm wondering if the white square on the DSP board is that, I thought at the time it was the "clicker".

 

You're making me want to take the camera apart again!

[jonoslack]

You won't, it's my daily camera...

Hi Mark

Wonderful to hear that - does it have any left over problems from it's traumatic experience? If not, I can imagine you liking it more for the effort it's taken (rather like children!).

[marknorton]

Jono,

 

No residual problems aside from a spec of dust inside the rangefinder window glass which I will deal with at some point. It's not a good idea to be constantly taking the thing apart.

 

Actually, the biggest single part of putting the camera back together (once I had found the broken flexprint track and repaired that!) was cleaning the inside of the glasses in the top cover. Normally, I wore latex gloves when handling the camera but the one time I didn't, I got fingerprints on the inside of the glass which for some reason were tough to shift.

 

I think Leica will benefit greatly from their new cleanroom - in lens assembly as well.

 

You're right about my attitude to the camera. I do prefer it to the other two, not least because I backed off the shutter release adjustment screws which has removed the grittiness of the release. Now, it's a much more fluid, silky, release even if the pressure points are "smeared". When the warranty expires on the other two (no point voiding 3 warranties!), I'll do the same on those, it's only 7 screws and 2 minutes to get the top off.

 

As you say, I also like the camera because of what it's been through and I'm less worried if it takes the odd knock. I suppost that if it's possible to bond with a bunch of camera parts, I have done so!

[M8China]

For those who look at the circuit boards in awe, may I offer an anecdot -

 

Many years ago I attended an electronic circuit board class, when the lecturer was asked how does a section of the circuit diagram work, he thought for a while and said, it works by "FM". Naturally we asked, what does FM mean, and he replied, "F***king Magic". We thought that was a good answer.

 

I can only say I am amazed by Mark's skill !!!

 

Many thanks:D

[hugov]

Mark, I couldn't make out the part number of the part number of (what I presume is) the ADC in your pictures (above the two sets of unpopulated pads to the right of the DSP). It looks like a Philips logo on it, but I can't read the part number.

 

Also, is the FPGA an XC3S200? I can't quite figure out the part number on that either; if it is an XC3S200, that is quite interesting, as that device is a fairly small FPGA (second smallest Spartan3 that Xilinx make). I don't think there'd be room in there for, for example, hardware JPEG - they must use the DSP for that.

[marknorton]

That Philips chip is a USB controller. I'd expect the ADC to be on the sensor daughter board close to the sensor itself.

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

The Xilinx is indeed the XC3S200 and I agree, it's likely the JPEG is not done in hardware.

 

[Advertisement]

Hallo,

Look here: Leica M

[hugov]

Thanks for that Mark (and thanks for your very quick reply)!!

 

Interesting they use a discrete USB chip, considering the PXA270 has on-board USB.

 

On closer inspection, you can see the 30-pin sensor board connector appears to be directly connected to the FPGA, implying the ADC is on the sensor board like you said.

 

From what I gather from your photos of the sensor board, it's not possible to see the other side of the board (i.e. underneath the sensor) without destroying it... pity! I would be very interested in what sort of ADC they use. I don't really know much about CCD digitization, but I assume some sort of programmable gain amplifier is used (to allow different ISO sensitivities), and possibly other analog front end circuitry (anti-aliasing filters for the ADC inputs, scaling etc.), but there doesn't seem to be very much room at all on that sensor board. And we can probably assume from what we can see about the rest of the hardware design that they probably use some kind of off-the-shelf ADC/analog front end IC, not a custom ASIC. 16-bit dual input 20+ MSPS ADC's with good noise characteristics and low power consumption are few and far between.

 

p.s. Mark, your macro technique is awesome - I know from experience how hard it is to get good photos of PCBs and small mechnamisms without nasty reflections or shadows!