Panasonic G7 Conversion

[Nemo Andrea]

Hello All,

 

With the recent arrival of my UV filter I figured I need to match my camera a bit better to the filter. The canon 1000D I have produces the right kind of pictures (or well enough for now), but the process is rather tricky and less enjoyable than I'd like. As my camera will do double duty as IR camera, I figured an upgrade is in order. Since I use micro-four-thirds (panasonic) cameras as my current system, I think a panasonic G7 will work best for my needs (electronic viewfinder, 4K photo (ive done some testing with this to see how well this can enable handheld UV photography), higher max ISO, video etc).

 

I'd like to lay out my plans and hope to hear some of your thoughts on this.

 

I was initially planning on doing a primitive conversion (bare sensor), as the camera only has contrast detect. This may mess up panasonic depth from defocus systems, but that wouldn't be the biggest loss imo. However, after going through the database of sensor stack thicknesses https://www.lensrentals.com/blog/2014/06/sensor-stack-thickness-when-does-it-matter/, I found that MFT cameras have relatively thick (4mm) sensor stacks, and as such this would probably make most native lenses unable to properly focus (infinity problems and such). So I figured I order some replacement glass. While it will likely be rather pricey (more than 50% of the camera), I think this would be the best way to go.

 

Kolari vision (as well as lifepixel) offer pre cut glass for this purpose. Now here is my main question that I'd like to hear your take on:

 

I would like to test the STC clip in filters. https://shop.stcoptics.com/product/clip-panasonicm43/ I have seen another post on this site asking if anyone had tried them, but I have not been able to find much on mft results. For me the advantages of this would be: not having to deal with step up rings, using older lenses without filter thread, and in general easier lens swapping. Naturally the downside is that for ''multispectral' photos (swapping filters for the same scene), it would be more cumbersome.

 

As these go behind the lens, they effectively increase the filter stack thickness. As such I was thinking perhaps I should request/ask kolari if they can cut glass to the right size that is the usual thickness of the camera (4mm) minus that of the glass used by STC (so probably about 2-3mm, which is nikon/canon thickness). I have enquired at STC what the thickness of the glass is for their product (I read both 1mm and 2mm on the same page at a third party retailer ). What do you think about this? Most of the time there will be a STC filter in place (with possible exception of UV, see below).

 

I suppose another thing that would be interesting to test/investigate would be whether its possible to reseat a different filter into the STC holder. That might allow for making a UV pass clip filter. I have asked STC if they would consider selling just the clip without a filter (I assume they wont, but it could be interesting). Cheapest clip in filter is 80 dollar, which is rather substantial if you only want the frame. I suppose once I know the thickness of the STC clip, the feasbility of this could be considered. Based on STC images, the filter seems to be held in place with metal tabs at the front, which should allow for user reseating and some tolerance to varying thicknesses. One challenge would be getting a UV pass filter cut to the right dimensions and shape. I don't have the funds to go cutting up and buying extra UV pass filters, so I suppose it'll remain an open question for a while. But perhaps a braver soul might want to pick that up if I can get some of the dimensions figured out in the future.

[Andy Perrin]

I wonder if you could 3D print those STC frames somehow?

 

I’m not sure about the rest of it. I don’t know of anyone who has adapted their camera glass to an STC filter like that.

[nfoto]

Not familiar with the model you intend to use, but you do have ensured it is without any pesky IR LED monitor, yes?

[Nemo Andrea]

Not familiar with the model you intend to use, but you do have ensured it is without any pesky IR LED monitor, yes?

 

Yep, the G7 is not listed as having the IR shutter problem on Kolari Vision's website, and it supports electronic shutter if such a trouble would turn out to exist.

[ulf]

One potential problem with clip in filters is the increased risk of contaminating the sensor.

Every time you remove the clip in filter-assembly, the sensor is exposed to dust.

 

I have four clip in filters for my EOS 60DAstronomik.de that I have stopped using.

When changing filter the sensor is facing up. Big dust particles tend to fall down on the sensor.

 

In a really clean environment changing the filter can be done without problems, if you are careful, but not in the field.

It is bad enough exchanging lenses.

[Nemo Andrea]

One potential problem with clip in filters is the increased risk of contaminating the sensor.

Every time you remove the clip in filter-assembly, the sensor is exposed to dust.

 

I have four clip in filters for my EOS 60DAstronomik.de that I have stopped using.

When changing filter the sensor is facing up. Big dust particles tend to fall down on the sensor.

 

In a really clean environment changing the filter can be done without problems, if you are careful, but not in the field.

It is bad enough exchanging lenses.

 

Hmm, I do expect dust to get onto the sensor, but if I have transparent glass layer on top of the sensor, cleaning should be equivalent to how one would ''clean'' a stock camera right? Unless the remounted glass does not produce a tight enough seal to prevent dust from actually getting directly on the sensor.

 

Did you ever notice any chance in sharpness or otherwise changes in optical performance with your clip in filters? (I suppose it's hard to tell without setting up a dedicated test before and after conversion)

[enricosavazzi]

[...]

Did you ever notice any chance in sharpness or otherwise changes in optical performance with your clip in filters? (I suppose it's hard to tell without setting up a dedicated test before and after conversion)

I am using a rear-mounted Baader U with lenses like the legacy AI Nikkor 24 mm on full-spectrum converted Sony A7 II, and it works fine. In fact, using a rear-mounted filter (in a clip-in adapter or mounted within a lens adapter between lens and camera body) avoids the color shifts frequent when front-mounting dielectric coated filters at the front of wideangle lenses.

 

With very fast lenses, the added thickness of the rear-mounted filter can slightly reduce image sharpness, but this is mainly noticeable with very fast legacy lenses designed for film cameras, and when these lenses are used with aperture fully or almost fully open.

 

You are correct that just removing the built-in IR and UV-cut filter in a Micro 4/3 camera prevents infinity focus with most lenses. This either requires that the filter is replaced with a window of a similar (optical) thickness and transparent to UV and IR, of the use of a lens adapter between lens and camera about 4 mm shorter than standard adapters. The shorter adapter is of course only feasible when using e.g. DSLR lenses on a mirrorless camera.

[dabateman]

I thought the G7 had an IR shutter problem. But can't find that now. I might have it mixed up with the GX7.

This is how to take it a part if yoir brave:

 

https://www.lifepixel.com/tutorials/infrared-diy-tutorials/life-pixel-panasonic-lumix-g7-diy-digital-infrared-conversion-tutorial

[ulf]

Hmm, I do expect dust to get onto the sensor, but if I have transparent glass layer on top of the sensor, cleaning should be equivalent to how one would ''clean'' a stock camera right? Unless the remounted glass does not produce a tight enough seal to prevent dust from actually getting directly on the sensor.

 

Did you ever notice any chance in sharpness or otherwise changes in optical performance with your clip in filters? (I suppose it's hard to tell without setting up a dedicated test before and after conversion)

I also have a replacement window of correct thickness in my camera, so dust on the sensor is double wrong. Not dust on the sensor or even the sensor's cover glass, but on the replacement window. I still find it annoying to clean the replacement window too frequently.

 

No, I have not seen any decrease of sharpness with the clip-on filters from Astronomik.de

[Nemo Andrea]

Alright, so the conversion was a success. First a post with some comments on the guide by lifepixel. (https://www.lifepixe...ersion-tutorial).

 

Step 10:

Also remove a grounding (I think thats its function anyway) screw. This will be important when separating the assemblies later.

 

(image credit lifepixel of course)

 

 

Step 16:

Separating front and rear assemblies is a bit trickier than one might expect. I had to look at the G7 service manual (which can be found online) to find how to do it. The procedure is to first undo the top panel (contaning dials and viewfinder, which can be lifted off. Then to separate the assemblies, you have to pry the right side outwards a bit to allow the battery compartment to come out.

 

note: when you lift the top panel, you will get close to the flash capacitor circuit. It is covered by plastic so it should be fine, but keep that in mind before you start poking with screwdrivers in that region.

 

Step 17:

Here it becomes a bit tricky. If you remove the black inner screws, the whole optical stack on top of sensor (sheet, rubber, hot mirror, dust shaker) will be loose on top of the sensor and just fall off. I would personally indeed remove them and just have the optical stack fall out as youll be reusing none of the glass. It might also be smart to first only remove the outer screws, move the sensor to a different area, and then remove the black screws so that you can forget about the camera body for the while. keep the sensor facing into the table to minimise dust getting on it.

 

Step 18:

The dust filter is glued onto the black plastic (appears silvery in lifepixel image) that holds the hot mirror in place. From my understanding it has to go, and since its stuck on there really tightly, I just decided to break the glass with my screwdriver. One might try to heat the adhesive, but the risk of melting the plastic support frame was too great in my opinion (this would be disastrous). Safer to break it I'd say.

 

Step 20:

the hot mirror is weakly bonded to a rubber support frame that sits between it and the sensor. Be careful in removing this as it tears easily... To reassemble just replace the glass with the new filter (in my case kolari vision's clear glass for full spectrum), place the black plastic frame on top (with the dust filter in a thousand pieces somewhere else) which should yield a general alignment on top of the sensor. Fine alignment is guaranteed by some pins so that will always be fine.

 

Other notes:

[1] reconnecting those ribbon cables is going to be a real pain in the ass.

[2] My replacement filter had a bit of residue on it (not sure if it was because I somehow touched it), but simply a wipe with microfibre and 70% ethanol seemed to fix that well enough.

[3] Compared to the canon 1000D conversion, I found this one a bit easier. The 1000D required more dangerous actions in the proximity of the sensor; here the sensor is pretty safe I feel.