Project 01 – Yashica T4

Project 01 – Yashica T4

Cult camera with a reputation for sharp lens and unique rendering, made famous by fashion and portrait photographer Terry Richardson. This is a compact camera of repute and many many articles have been written about it. There are a few links at the bottom of this article if you are not familiar with it. I’d like to focus this article on the lens conversion aspects.

My particular copy belonged to my father and came with a stuck shutter. Destined for the bin I decided to see if I could salvage it. The following project documents my efforts to do so. I have a compulsion fettle so, there will be updates to this article as I go through variations.

Lens: Carl Zeiss Tessar 35mm, f/3.5 lens, 4 elements in 3 groups, T* (multicoated)

Optical system construction: Lens is encased in a mono block plastic cylinder with 3 lugs which hold the lens block in place on a focusing platform. Also attached to the focusing platform is the combination aperture and shutter mechanism.

Shutter and Aperture: Combined Shutter/Aperture mechanism, single set of blades

Common Failures:

  1. Shutter failure – replaceable with donor shutter cable
  2. Motor drive failures: motor not working (replaceable with donor motor), motor not advancing full frame (often it’s a logic board problem and non economical to repair)

Lens Removal:

This is guide on how to remove the lens out of a Yashica T4. It is NOT a repair manual. All the cameras I extract lenses from are beyond repair and my method of removing lenses is based on the easiest way of removing the lens at the expense of the donor camera.

It is fairly straightforward to remove the lens off the Yashica T4. All you have to do is remove the bezel, remove 3 screws and extract the lens. Whilst this appears to be a non destructive, the lens is mounted immediately in front of the shutter mechanism; which is a known area of weakness. Damaging the shutter mechanism is a risk of removal. The bezel itself is held on by 3 prongs on to the focusing platform. These prongs are extremely fragile and may not survive the removal process. However, this is not necessarily a bad thing as, if you intend to follow on by modding the lens, you can use the bezel as a protector and lens hood.

guide_img_15Step 1 – Expose the lens

Insert battery into the camera and turn it on. The lens protector should slide down and expose the lens bezel and lens. If you are unable to power on your camera, you can manually slide the lens protector downwards and out of the way. However, the lens and bezel will be recessed into the body and it may be difficult to remove the bezel. If you have any difficulties, refer to the addendum of this guide and remove the camera casing.

guide_img_17Step 2 – Remove bezel

Once exposed, place a small screwdriver at the 12 o’clock position of the lens and apply gradual pressure to the bezel until the retaining prong releases from it’s socket. Repeat the process at the 4 and 8 o’clock positions until you remove the bezel.

guide_img_13-1Step 3 – Remove lens

With the bezel removed, all you have to do is remove 3 screws to remove the lens itself.


The extracted lens is approximately 20mm at it’s widest (not including the lens mounting lugs) and 12 mm high. It’s Zeiss T* coated and comprises of 4 elements in 3 groups


If you have problems extracting the lens, you may still access it by removing the camera’s outer casing.


Remove two screws on the right side of the camera as indicated.



Next, remove 2 screws from the bottom of the camera.

Finally, remove the 3 screws within the film chamber.

The Yashica T4 is weather sealed so it is not easy to pry the front and rear casing apart but it is possible.


Method of conversion:

  • Simple 3d printed mount with lug for printed aperture plates
  • Lens mounted behind the base of the mount
  • 3D printed fixed aperture
  • Mount coupled to Leica M mount close focus helicoid

Rear of mount with lens in place



Rear of mount with aperture plate in place


Front of lens with lens and bezel in place

Sample Images

Sony A7M2, 35mm f3.5, 1/3200th sec, ISO 100
Sony NEX-6, 35mm f3.5, 1/200th sec, ISO 100
Sony A7M2, 35mm f3.5, 1/250th sec, ISO 100
Lens Modding Overview

Lens Modding Overview

My particular journey into experimentation with lens modification started when I chanced upon an article form a blog by Yu-Lin Chan, a long time lens modder, on his blog – The article showed samples of images taken with an Aires Coral 45mm f1.9 lens from an Aires rangefinder camera which showed amazing rendition and bokeh – a lens character, if you will. Something which you would struggle to find in modern day lenses with their in camera image processed output.

Flash forward a few months later and my dad surprises me with a cache of cameras belonging to the family from my grandfather onwards. The choice items include a Yashica T4 and an Olympus Trip 35, both cult cameras, but sadly dead. This started me on my journey to see if I could salvage the lenses from compact cameras and rangefinders known to have good lenses and create a set of lightweight prime lenses I could carry around with me when travelling light.

This overview is particular to my experience and covers areas which I have explored so far, and that is very much based on the fact that I am a big advocate of 3D printing. There are other ways of modding such as gluing filter rings on to lenses and attaching them onto the camera using macro reversal adapters but that’s something I’ve tried… yet.

Why bother?

Let’s get one thing straight up front. Transplanting a lens which has been scientifically calibrated to perform on a dedicated, aligned and optimized to a specific camera is never going to yield as good a result as for the camera it was meant to serve. If you enter into this

However, given the right approach and care, a transplanted lens can yield results which produce interesting and rewarding results.

My experience to date has been mixed. In general, the transplanting of the lens and creating a mount and aperture for the lens has not been difficult. However, the performance of the transplanted lenses has been a bit of a mixed bag. The majority of the lenses I have transplanted have lived up to the expected performance in the center of the lens but not able to provide good edge definition.

As the title of this article suggests, this is a learning experience for me and one which I will document and share with those who are interested.

I have, to date managed to transplant over a dozen lenses and ended up with a whole lot of very usable pancake type lenses, mostly weighing under 100g, which I carry around with me to supplement my usual shooting set up and had a lot of fun along the way.

Types of lenses

There are probably technically accurate terms for these, but these are the 3 most common lens configurations I have encountered so far when extracting lenses from compact cameras and fixed lens rangefinders are:

ov_img_12Multiple lens groups, separated by an aperture and/or shutter – this configuration is the most common for fixed lens rangefinders and high end compact cameras. These lenses are the most challenging to modify as you will need to maintain register between the different lens groups, remove or modify the shutter/aperture and find a way of implementing an aperture between the lens groups.

Failure to maintain correct register and distance between the lens groups will result in aberrations and a distorted image. If you’ve ever used a lens such as the Lensbaby Composer or tried ‘lens bashing’, you will be familiar with the effects.

The next challenge will be to either remove or disable the shutter mechanism, leaving a clear light transmission path. In most cases, disabling the shutter mechanism by jamming the shutter wide open is the easiest way if you can locate the relevant part of the shutter mechanism. However, if your shutter mechanism does not work in the first place, then your only option is to remove it completely.

All this, bearing in mind that the next considerations – how to throttle light transmission via an aperture and allow the lens to focus.

ov_img_06Single lens block, usually sealed – this configuration is the easiest to transplant. Once extracted, the lens group is easily mounted. As these lens type is designed sit in front of their camera’s shutter/aperture mechanism, it’s also easy to create an aperture and focusing solution.

In some instances, the entire lens block, aperture and helicoid can be extracted as a single unit, making the transplanting of the lens even easier.

The only drawback is that image quality is not as great as the multi lens group with aperture in between. In my experience, some monoblock lenses are meant to be only shot wide open, with no aperture and using only shutter speed to control exposure. In these instances, adding an aperture mechanism has little effect on the depth of field, and can even cause vignetting.

There is no easy way of telling, you just need to try various options.

Multiple lens groups, floating lens elements, separated by an aperture/shutter – the floating elements in this configuration pose a serious challenge when transplanting recreating the focusing mechanism usually involves trying to replicate intricate mechanicals.

I’ve not tried to transplant such lenses, yet. My Olympus XA will remain in storage until I have a little more experience under my belt.

All the above configurations are based on compact cameras and rangefinders with fixed focal length lenses. I’ve tried to transplant zoom lenses from compact cameras but, with a couple of exception, they have not been successful. This will be covered in my future articles.

Mounting considerations

The object here would be to create a platform in which to mount the lens and, subsequently on to the camera.

In terms of creating a platform for the lens, there are 3 methods which I have tried to date:

Simple 3d printed mount –mounts lens block, simple add on aperture, attached to close focus helicoid. This is the simplest way of transplanting a lens, and works particularly well as a quick and easy way of transplanting lenses.

Complex 3d printed mount – mounts multiple lens groups, provides capability to integrate a separate aperture between the lens groups. This is the only way of transplanting a lens with multiple groups.

Donor lens, in which the existing lens elements are replaced with the lens elements to be transplanted.

Which method to use will be led by the construct of the lens and your options for creating a focusing and aperture solution to use on your particular camera.

In some ways, the camera you intend to use the lens on will dictate how you approach the mounting of the lens. If you shoot using mirrorless camera, there are more options including a lot of off the shelf focusing helicoids.ov_img_20

Focusing considerations

Perhaps the most important aspect for modding lenses is how you intend to implement a focusing mechanism. It is possible to shoot wide open, that is to say without an aperture, but if a lens does not focus easily and smoothly then it becomes a lot less pleasant to handle and use.

There are many ways of doing this and doing so successfully will lead to a lens which you will end up wanting to use, rather than one which you build for fun. The two methods I have tried to date are:

Use a close focusing helicoid – these are available in many mount options. The ones I would recommend for lens modding purposes are the Leica M lens mount to your camera mount helicoids if it exists, for the reasons I mention below.

3d printed helicoid, friction focus – this is a very crude way of focusing but is the quickest way for you to adjust distances between lens groups and establish the correct flange distance. It’s just a printed tube to mount lens elements which slides through hole in a printed mount. It’s not a long term solution, but I use this exclusively to determine the lens parameters before creating a more permanent mount to mount on to a focusing helicoid or donor lens shell.

There are other ways of creating a mechanism for focusing your lenses, such as screw thread focusing. This method is commonly used to focus images thrown from projectors. It’s not a method I’ve tried myself so far, but I am certain that I will be doing so in one or more of my future projects.

Lenses from compact cameras and fixed lens rangefinders are only designed to work in their intended cameras, their flange distance is often unique and poses challenges.

The primary reason why lens modification is more suited for mirrorless cameras is because of flange distances. Like the compact camera or or fixed lens rangefinder camera that the lens comes from, mirrorless cameras have a short flange distance. This makes it easier for the donor lens to be attached to a focusing mechanism. DSLR cameras have a longer flange distance to accommodate their mirror box, which makes modifying lenses for them difficult if not impossible.

The reason I recommended using Leica M mount focusing helicoids is because Leicas, being rangefinders, have a short flange distance. This allows for us to reasonably increase or decrease the flange distance for the donor lens by manufacturing a mount by recessing or protruding the lens. Additionally, the short flange distance also allows us to design built in apertures for the lens platform.

Aperture Considerations

When considering an aperture solution, there are a lot of options and most of them are straightforward to implement (corresponding images below, left to right):

3d printed fixed aperture plate – this is the simplest solution. It’s basically a plate with a fixed opening, which sits close to the last element in the lens

3dr party component aperture – these are standalone apertures which are manufactured specifically for scientific optical equipment.

Aperture from the camera the lens came from – in a lot of cases, the aperture from the camera the lens came from can be reused. However, in most cases, the apertures in compact cameras are controlled via internal linkages. The difficulty here would be in making an appropriate linkage to operate the aperture once transplanted.


What’s next?

What I’ve written so far is only a guide to the thinking process involved in modifying and transplanting the optics from compact cameras and fixed lens rangefinders.

As an outline, the process I have found to work is as follows:

  1. Choose your camera, do not expect to salvage any parts. It’s possible but more often than not, the camera will end up being next to useless.
  2. Do your research – check for repair manuals/online repair guides, it will give you an idea of how to dismantle the camera, lens may not be worth extracting if the camera has a floating lens elements for example.
  3. Open the camera – avoid the capacitor, tape over exposed contact points.
  4. Extract lens
  5. Choose mounting, focusing and aperture approach
  6. Print necessary components in draft quality to check tolerances and fit
  7. Assemble the set up and make adjustments to the components
  8. Repeat until you have a satisfactory fit

The last 3 steps require a lot of perseverance and patience and I do not expect anyone reading this to be proficient just from reading this article.

If you are still interested in doing so yourself, keep visiting this site. I will be documenting the various projects I embark on.

Review – Pirate3D Buccaneer

Review – Pirate3D Buccaneer

Overview/Initial Impressions

The Buccaneer is the third 3D Printer I have had long term experience with, the others being build it yourself 3D Printer kits and the XYZPrinting DaVinci. My immediate impression of the Buccaneer was that I sure hope that it preforms as well as it looked. The whole package unashamedly tries to copy the Apple minimalist ethos from the sleek design, reminiscent of the Macintosh Cube (remember those?) to the sparse documentation which accompanies the printer.

It is partly this sparse documentation that my review is longer than I would like, as some of my experiences using the machine will help those of you who already have the machine or about to receive it from Pirate3d as they fulfil their Kickstarter back orders.

The rest of the package includes a printer platform, 2 sticky printing pads and a 400g reel of white PLA. In their Kickstarter campaign, Pirate3D goal aimed at raising funds to build a under USD$400 3D printer which can be set up and printing in less than half an hour. Certainly the first impressions were positive, but the proof of the pudding is in the eating.

Specifications bucc_review_img_1

Buccaneer’s basic specifications are as follows:

Printing Technology: Fused Filament Fabrication
Highest Layer Resolution: 50 microns (0.05 mm)
Filament Diameter: 1.75 MM
Cartridge Capacity: 400 G
Max Print Size: 130 mm x 96 mm x 139 mm
Nozzle Diameter: 0.4 MM
Product Weight: 8 KG

The Buccaneer only works with PLA and Pirate3D recommend only using their own, whilst not ruling out using generic PLA.  Certainly their own PLA spools are the only ones which will fit within the built in filament spool cavity at the top of the printer. More importantly, however, is the fact that PLA printing parameters are preset and printing using ABS is not supported. The reason for this is most likely because there is no user control of the extruder temperature. This means that, aside from ABS, printing using exotic PLA filaments such as Laywood or flexible PLA (for example) has to be done at your own risk.

The build envelope puts it in the small print capacity category, which is something crucial to take into consideration. It just about suits my needs and there are plenty of ways to be creative with the objects you print so that this is less of a limitation.

Maximum print quality is an impressive 50 microns, which makes this one of the few FFF printers with this capability and is only exceeded by a couple of, much more expensive, FFF printers which can print at 25-35 microns

Build Quality

The build quality of the Buccaneer is on the whole good, but let down by some poor design choices and production issues.

Focusing on the positives, the print mechanism is a very nicely finished. Centered on stamped Stainless Steel frame, which keeps the main stepper motors tensioned around the belt motors, all the printer components are thoughtfully integrated into the polycarbonate body.  The extruder head assembly is built around a standard MK7 extruder head and extractor fan. The whole assembly is neatly wired and insulated with tape and a heat shield. The controller board and electronic circuitry are protected by heat shielding and all cables are neatly cable tied. It is an example of a machine built around a consumer mindset and a far cry from other entry level 3D Printers, which are largely evolved from early kit form 3d Printer aimed at enthusiasts.

However, all is not rosy as there are many build quality issues. I have an early production model, probably from the very first batch to be shipped. This would explains some of the more basic build quality issues like the silicone pads at the base of my printer becoming detached because of poor adhesive or the 2 of the 3 magnets which hold the build platform onto the printer mechanism dethatching because they were not properly attached to the platform. In both cases, a spot of industrial strength glue solved the problems but not before causing misprints. There are, however, more fundamental issues caused by bad design.

The first and most immediate of the design flaws is the magnetic platform itself. It fails, even after regluing. One particular magnet has even detached again after gluing. The magnets themselves are just strong enough to hold the platform in place and sit in grooves so when it fails, it’s not immediately apparent. And when it does, the platform does not sit perfectly level, but it’s not visibly apparent until you start to print and get edge curling or print head clogging because the printer is printing across a horizontally uneven surface. It took endless unsuccessful calibration attempts before the problem became evident and I’ve had to reinforce the magnetic base as a result. I will detail how I did this in a future post.

The second most annoying issue I’ve had with the printer is the amount of tension the filament feed mechanism places on the filament. I have occasionally had filaments fracture, yes fracture, into multiple pieces on its own accord so much so that my standard practice is to eject the filament after use and snip off the parts which were still in the feeder tube when printing, just in case. When a filament fractures, you will need to dismantle the extruder unit in order to feed filament in from the extruder end of the filament feeder tube to flush out all the bits of filament which is a lengthy process.

The third, and somewhat unsubstantiated, design flaw is that the beautifully crafted transparent polycarbonate base is not adequately braced. I say somewhat unsubstantiated because in the 2 months I’ve had the printer, the base has already bowed inwards by about a centimeter, across the front opening. The bottom of the opening needs to be braced or else the combination of the weight of the printer and resonant frequencies generated by printing will cause the base to fail. It’s just a question of time. Ironically, polycarbonate stress fractures were also the cause of the Apple Cube to be withdrawn from the market and that machine was a lot lighter and had no moving parts.



Setting up starts fairly straightforwardly. Attach the printing pad onto the print platform with the double sided sticky tape on the bottom of the pad and attach the platform to the printer via the magnets at the bottom.

Downloading and installing the Windows/iOS/Android software is the next step which leads to the first connection directly with the printer, which acts as a WiFi access point when switched on.  Once connected, the next step is to set up the printer on your wireless network which also makes the Buccaneer a shared printer.

So far so good but then the whole experience sours. The next steps to set up the printer in order to be able to print is needlessly convoluted.

Loading, unloading the filament and calibrating the extruder to print accurately on to the platform are software controlled. In other printers, this is achieved by physical buttons but in an attempt to out-Apple, Apple, the Buccaneer has no physical buttons or interface save the software controls over WiFi. So simple tasks like loading the filament and telling the extruder how far it needs to be over the platform is both tedious and annoying.

For example, load filament – push filament through until it reaches the extruder unit, activate software feed cycle, confirm platform is clean, wait for platform to raise to the extruder (30 seconds), wait for extruder to heat (30-45 seconds), wait for filament to extrude a small amount and another 30 seconds for the platform to be lowered to its base position and the extruder head to cool down.

OK, so it takes a longer to feed a filament, why would I complain? Well, if the filament does not catch and extrude, you have to repeat the entire process again. Yes – confirm platform is clear, yes – wait again for the platform to rise, yes – wait again whilst the extruder heats, yes – pray that the filament catches this time. Repeat process as often as needed for successful feed.

With other printers, the head remains heated, the platform (if applicable) remains elevated until you confirm successful feed and you can control the feeding duration manually, i.e. keep pressing the filament feed button until you see filament oozing out. The Buccaneer software goes through a predefined feed attempt cycle with a set period of filament feed, automatically cools down the extruder head and lowers the platform. If the filament feeder mechanism does not catch the filament, you’re locked into another mindless cycle of playing platform yo-yo and waiting for something which was perfectly hot to heat up again and again.

The same convoluted, mindless, multi-step process applies to both the unloading and calibration apply. Thankfully the calibration only needs to be done once in a while. There are other reasons for disliking the software controlled interaction which I will cover later.

Once you are familiar with the processes to set up the printer for successful printing, it becomes instinctual and easy (even if a bit un-necessarily lengthy) to get printing.


Printing and Print Quality

On the printing side, there is only one option and that is to use the Buccaneer Windows/iOS/Android software. The printer has no direct external connections like USB or SD Card slot and can only be accessed via WiFi, either as a standalone Access Point or as part of a WiFi network. This is potentially a drawback as it means that 3rd Party software such as Cura cannot currently be used.

Thankfully, the Pirate3d Buccaneer software is really well developed and easy to use. The software itself is integrated around Pirate3d’s Treasure Island repository. The repository itself is there to serve as a ready source of 3d models which have been tested and known to work on the Buccaneer.

Treasure Island also exists as a repository website,, which accepts user contribution and (eventually) as a marketplace of 3d objects. It is currently underdeveloped and only hosts free objects. I say underdeveloped as all descriptions of models are sparse, non-categorized and only contain basic meta tags. It is also only searchable in the iOS/Android versions but even then, because of the brief descriptions and basic tags, searching is a fairly hit or miss affair.

Your only other option for printing your own or 3d models from other repositories is to import this via the, not entirely obvious, “Add New Print” button on the Windows. If you have to rely on solely on the iOS or Android then you are currently out of luck. Pirate 3d have stated they are working on a way for models to be loaded by iOS/Android via Dropbox integration. This would mean that users could save their 3d model files on Dropbox and load them via the iOS/Android apps. At time of writing, this has not been deployed.

Loading STL is only limited by a file size limit of around 22mb, this is undocumented and from my own experience in loading STL files. I presume that the 22+mb limit is a constraint of the on board memory as STL files remain on the machine during the printing process.

Once loaded, the model appears as a visual, scaled, representation on a virtual platform in a perspective view. There is a slider at the bottom right of the screen which allows the user to rescale the model to fit the Buccaneer’s 130 mm x 96 mm x 139 mm build envelope. What’s not obvious, though, is the fact that the software automatically scales any loaded model to fit the build envelope. Again, this is undocumented and caused many wasted prints. If there were a simple warning and override button, the user could use the on screen buttons to re-orient the imported model to fit within the build envelope by flipping the model on the X,Y or Z axis from its default orientation.

Once the imported model is oriented and scaled, the next step is to select the print resolution, adhesion, support and infill options (hidden in the Advanced Settings checkbox). The nice features here are the automatic adhesion (brim, raft) and support settings which do a really good job. I rarely print models with potential for infilling so will refrain from commenting on this feature save that “Advanced Settings” is a bit of a misnomer for this button.

Once confirmed, the printing begins and this is where the Buccaneer shines. Set to the highest resolution of 50 microns, a properly prepared and calibrated platform and using Pirate3d’s own filament; the results are very good. Most of the models I have printed this way require little more than removal of supports and the odd small surface artefact. Layer stepping at 50 microns is so fine that it takes on the appearance of a matt finished plastic object. Only the evenness of the stepped layers gives this away on closer inspection.

I have not done timings or calculated printing speed as this is a criteria which I am not concerned about nor do I have an immediate basis for comparison. I will append this review with this information once I have had a larger pool of printers reviewed.

The only other downside to printing specifically with the Buccaneer that the machine is LOUD.  It is loud because of the way the printer is built. The base is built around a 3 sided wall construction with the front open for platform access. This base also acts as an amplification chamber, magnifying the sounds emanating from the stepper motors and cooling fans to a distractingly loud 82db. However, this being a wirelessly networked printer, it is possible to place it somewhere where the noise would not be an issue. I’ve done just that and placed the base on sound absorbing material, which helps.


In reviewing this printer, I have tried to be as objective as possible and give the reader an idea of what it’s like to use this printer on a daily basis. Most of the negative points I have made are down to early production issues and the software is constantly being updated. I count at least 3 software updates in the last 2 months and when components have failed, Pirate3D have been quick to resolve issues. I had an extruder unit fail on me which was replaced with haste. This is extremely rare and, as the unit itself is a pre-assembled component bought in by Pirate3d, I chalk it down to luck of the draw but I would praise the response of the company in rectifying the issue.

Would I buy the printer again if I had the choice? If they fix the software driven load/unload filament and calibration issues, reinforce the platform and brace the base then – yes, definitely. The Buccaneer is (currently) the most consumer friendly 3D printer I know of with high quality printing capabilities, two of the three criteria I look for in a printer because my aim is to encourage others to adopt 3d printing technology.

The third criteria being price, and now they have gone to market Pirate3d have upped their price to a psychological threshold price of USD999. Former Kickstarters like Robo3D and SL3D have also done the same with their products so we shouldn’t be surprised. But at this price it makes it a little more difficult to recommend to a friend. The question then becomes, do we recommend the Buccaneer for the ease of use and relatively hassle free maintenance or a cheaper enthusiast level 3D printer and risk ending up playing tech support as they potentially deluge you with endless setup and printing questions.

Despite all the media attention, 3D printing is still in its infancy and a long way from general consumer acceptance. The Buccaneer is a step in the right direction as far as far as I am concerned.



Overall nice design, build

Print quality at 50 microns

Ease of set up


Relatively maintenance free



Limited material options

WiFi only, no USB or card slot

Loading/Unloading badly controlled by software

Platform levelling

Does not support 3rd party software