Light meter calibration

I've been thinking a lot lately about how to correctly calibrate my various light meters. I have several, of both the incident/lumisphere and reflective/spot types, and they're all over the place. Not only do I want them to match, I'd also like them to all give the "correct" reading based on standards and specifications.

The Internet is chock full of people sharing their rules-of-thumb and get-it-in-the-ballpark methods, but pretty much none of those even vaguely resemble what a professional might do. The closest you'll ever get to that, is recommendations for specific labs that can do calibration.

What no one ever talks about, is how those labs actually do it.

That being said, I've recently decided to tackle the problem myself. While I'm never going to match the setup of a full-blown metrology lab, I should at least be able to attempt a somewhat calculated approach.

The Printalyzer Densitometer Project

The Printalyzer Densitometer project is my attempt at building a new budget-friendly densitometer to make available to the resurging analog photography community. My ultimate goal with this project is to eliminate the cost and convenience barriers that currently stand in the way of many, who might want to try using such a device.

Its actually a project idea I've been kicking around for a long time, but initially sidelined due to concerns over optics and calibration. However, I've recently become motivated to revisit the idea.

Printalyzer Project Update

Since I wrote my original introductory blog post on this project, I've made considerable progress. The first prototype is now fully constructed, and I'm making significant headway in the development of the firmware. As such, I think its time for another project update.

Given that the project is being developed entirely as open-source hardware and software, the latest schematics and code can always be found in the project's Github repository.

Main Unit Construction

The main unit consists of a rather large PCB, with a double-sided design. The reason it is large is that all the components and most of the ports of the Printalyzer are directly mounted to the board. The way I decided to do this, the board is actually mounted in the enclosure upside-down. This puts the buttons and display on the "bottom", while everything else is on the "top". I took this approach because several of the components are rather tall, and there wouldn't be enough clearance if they had to fit within the height limits of the buttons. From my teardowns of other pieces of darkroom electronics, this construction approach is actually quite common.

For the enclosure, I designed something using Front Panel Express and their housing profile script. While rather expensive for building in quantity, at the prototype stage this gave me very nice results.

For the top of the unit, I installed a piece of Rosco Supergel #19 (Fire Red) sandwiched between some transparent material in the gap between the top of the display and the enclosure. This both protects the display and keeps its output spectrum paper-safe.

Finding a nice metal knob for the encoder, without a pesky direction-mark, took some effort. Ultimately I went with a type of knurled guitar knob (from mxuteuk), which seems to be the magic set of keywords to find these things.

For the keycaps, I ultimately went with black. This was mostly because it was the only color for which I had all the necessary parts at the time, given that these are often backordered or special order with lead time. When looking at the unit on the desk, black also seems like the nicest choice. However, in the darkroom, they are hard to see. I also have these keycaps in dark gray, and they can be special-ordered in light gray. (Having that array of LEDs around the keys really helps, of course.)

Meter Probe Construction

The meter probe is a much simpler device. It essentially consists of a light sensor, a button, and minimalist support circuitry to handle power regulation and voltage level shifting. Right now the sensor I'm using is the TCS3472, which has well-documented lux and color temperature calculation guides and might be useful for color analysis as well.

(One benefit of my design here, is that I can make future versions of this meter probe using basically any sensor I want. The only requirement is that the sensor, or really just the front-end of this circuit board, has an I2C interface.)

Since the enclosure requires a somewhat custom shape, I had to go with 3D printing for this one. For the two prototypes I've built so far, one was self-printed and the other I had printed by Shapeways. The later looks a lot nicer, so that's what I'm showing here.

The hardest part of assembling this was probably the fiddly process of connecting the cable. For the cabling, I went with a 6-pin Mini-DIN connector. In practice, that meant finding someone selling M-M black PS/2 cables and slicing one in half.

One last thing this meter probe needs, is some sort of cover for the sensor hole. My plan is to have something printed/cut for me out of something like polycarbonate, from a place that makes panel graphic overlays. Ideally this overlay will be light in color, with markings pointing to the sensor hole. Thus far I've been putting off this part, but getting it in place is going to be essential to finalizing my calibration of the sensor readings.

Firmware Development

I'm now at the stage in firmware development where I have all the hardware inside the Printalyzer working, and have a stable base from which to build up the rest of the device's functionality. My initial goal here is to fully flesh out the functionality of a mostly-full-featured f-stop enlarger timer, then to later follow on with the exposure metering functions. I'm going about it this way because metering requires a lot of testing that's hard to do "outside the darkroom," and I'll need the timing functions regardless.

(In all the screenshots below, I apologize for the "gunky" picture quality. The red gel sandwich over the display tends to reflect a fair bit of dust, which ends up looking far more distracting in pictures than it does in real life.)

Home Display

The home display shows the current time and selected contrast grade. It also has a reserved space for the tone graph. While I have figured out what that graph will look like, its not easy to show here until I do actual metering.

From this display its also possible to change the adjustment increment, and obviously to start the exposure timer.

Another feature of the home display (not shown here) is fine adjustment. Using the encoder knob you can enter an explicit stop adjustment (in 1/12th stop units), or even an explicit exposure time. I expect this capability to be useful for repeating previous exposures, in lieu of re-metering.

Test Strip Mode

The test strip mode supports both incremental and separate exposures for test strips, and making strips in both 7-patch and 5-patch layouts. The design places the current exposure time in the middle, and then makes patches above and below this time in units of the current adjustment increment.

Another nice feature, made possible by having a graphical display, is that it actually tells you what patches should be covered for the next exposure. This way its harder to lose your place during test strip creation.

Settings Menu

The settings menu is fairly rough at the moment, and probably does need some user interface improvements. That being said, it shows another benefit of using a reconfigurable graphical display. I can show nice text menus that actually tell you what each setting is, and then return to a full-screen numerical time display when done.

Enlarger Calibration

This little tidbit is the first feature I've worked on so far that actually uses the sensor in the meter probe. Its not yet complete, from a user-interface and settings point of view, but the underlying process is fully implemented now.

The basic issue is that any enlarger timer is effectively just "flipping a switch." It can control when that switch turns on, and when it turns off, but little else. Unfortunately, lamps typically do not turn on or off instantly. On both ends, there is a delay and a ramp time. While these are typically short, not accounting for them can cause consistency issues with short exposures. This problem becomes especially troubling when doing incremental short exposures, such as with test strips and burning.

The goal of the calibration process is to make sure that the "user visible exposure time" is the time the paper is exposed to the equivalent of the full light output of the enlarger, rather than simply the time between the on and off states of its "power switch."

Calibration works by first measuring reference points with both the enlarger on (full brightness) and the enlarger off (full darkness). It then runs the enlarger through a series of on/off cycles while taking frequent measurements. When complete, it generates 6 different values:

• Time from "power on" until the light level starts to increase
• Time it takes the enlarger to reach full brightness (rise time)
  
• Full brightness time required for an exposure equivalent to what was sensed during the rise time
• Time from "power off" until the light level starts to decrease
• Time it takes the enlarger to become completely off (fall time)
• Full brightness time required for an exposure equivalent to what was sensed during the fall time

These values are then fed into the exposure timer code, and used to schedule the various events (on, off, tick beeps, displayed time, etc) that occur during exposure.

Using the above numbers as a rough example of what this all means, without profiling a simple 2 second exposure would actually expose the paper to light for about 2.4 seconds but only a 1.88 second equivalent exposure.

These numbers come from a simple halogen desk lamp, which is more convenient for basic testing. I've also run the same calibration cycle on my real enlarger. It takes longer to rise to full brightness, but is faster at turning off. (Its similar case is 2.2 seconds with a 1.78 second equivalent exposure.)

Now I'll admit this doesn't seem like much, but enlargers do vary and it can suddenly become critical if you're making <1s incremental exposures for test strips or burning.

Blackout Mode

This feature is a little "quality of life" nicety that I haven't seen anyone else do. Every once in a while, some of us want to print on materials, such as RA-4 color paper, that has to be handled in absolute darkness. This means once your enlarger is all setup for the print, you need to turn out all of the lights in the darkroom before taking out the paper. Having to manually turn off your safelights and throw a cover over your illuminated enlarger timer, gets a bit annoying.

This is why I added the "Blackout" switch to the Printalyzer! Flipping this switch turns off the safelight and all of the light-emitting parts of the device. Of course everything still works when in this mode, so you can still make prints and test strips in complete darkness. (Eventually I'll probably add additional audio cues and lock some ancillary features to improve usability.)

Next Steps

At the outset of this project, I collected a long list of "good ideas" for things the device should do. That list is still growing, of course. Eventually I will get to them, but right now my priority is to get the standard features implemented and stable.

I have a few more things to do for the basic f-stop timer functionality, such as burn exposure programs. After that, I need to work on an actual user interface for adding/changing/managing enlarger profiles.

Once these things are done, I'll finally move on to paper profiles and print exposure metering. I expect that to be a long-tail item, because it will require a lot of work on sensor calibration and learning about sensor behaviors. However, I plan to do as much this as possible via first converting raw sensor readings into stable and standardized exposure units. This way, it will be possible to experiment with different sensors (as needed) without dramatic changes to how the metering/profiling process works.

Another thing I'll likely do interspersed with this, is start to take advantage of the USB port I put onto the device. There are a few features I want to implement with this, including:

• Backup/Restore of user settings and profiles with a thumbdrive
• Firmware updates with a thumbdrive (right now it requires a special programmer device, which isn't practical for "end user" use).
  
• Keyboard entry of profile names (I'll make sure they can be entered without a keyboard, but being able to use one is a nice thing.)
• Connection to densitometers to help with paper profiling (not sure how this will work, but it can't hurt to try).

I'm still not sure whether, or at what point, I'll proceed with transforming this project into an actual "product" or "kit." The biggest hangup is really that the device necessarily plugs into mains power and switches mains-powered devices. That means it may be hard to safely sell/distribute units without first forming a company and going through various painful and expensive certification processes. However, I'm not going to worry too much about that until I have something rather complete.

Regardless, all the necessary data to build one yourself will always be freely and publicly available. However, actually assembling one of these does require circuit board assembly skills and tools everyone isn't likely to have.

Introducing the Printalyzer!

The Printalyzer is a new project I've been working on for several weeks now. It aims to become a modern full-featured darkroom enlarger timer and exposure meter.

 

What is it?

At its core, this is a timer for a darkroom enlarger. That means it has a switchable outlet on the back, and can turn an enlarger on and off for a set amount of time. However, it is going to be far more sophisticated than a simple clock. Its going to allow the adjustment of exposure time in "stops," making it what is known as an "f-stop timer." That means you adjust time in logarithmic units of exposure, rather than linear increments. Here's an article that attempts to explain the concept.

Of course it'll also include features to help with making test strips, and calculating dodge/burn exposures in stop units as well.

In addition to all this, the device is going to include an exposure metering probe. This probe will let you measure and visualize the contrast range of a negative before printing it. The idea is that you can figure out a fairly decent choice of both contrast grade and exposure time, without even making a single test strip.  Of course you still can make the strip, but it becomes more about fine-tuning than making your initial decisions.

Why are you doing it?

Why have I chosen to take on this project? Don't similar devices already exist? Well, technically they do. Several, in fact. You can find everything from polished commercial products to crowdfunding campaigns to homebrew hobbyist projects.

Okay then, why am I bothering?

Well, in short, I've found myself growing increasingly frustrated with what is currently out there.

Most of the commercial products in this space are quite expensive, and were designed and built in the mid-to-late 1990's. While they do work, they tend to suffer from all the limitations of "built to cost" embedded devices from that era. This often means a limited user interface that can be difficult to use without constantly referencing a manual, firmware that is difficult or impossible to update, features that are constrained by the capabilities of an 8-bit microcontroller, fixed peripheral choices, and a general lack of new development work.

Many of the more modern projects tend to forget things that made those old devices great, like a sturdy case, real buttons, and a no-nonsense primary interface. They also sometimes try to do too much, such as attempting to be a general purpose darkroom timer. Finally, its rare that they tackle the exposure metering problem.

 

What are your project goals?

Basically, to take what I like from those older products and to bring it up to date with more modern embedded technology.

Fundamentally, I like to think of this project as building a platform, rather than a static appliance.  Sure, its going to have all the necessary hardware to be an enlarger time and meter. However, its also going to have enough excess capacity so that it can be continuously updated to increase its functionality.

My goals for the Printalyzer include:

• Use real buttons for the user interface, and have enough of them that the need for awkward combinations is minimized. A rotary encoder knob may be used where it makes sense, but not as a catch-all.
• Use a graphical display whose layout can be changed depending on the device mode. It can emulate the look of 7-segment and bar-graph LEDs when that makes sense, but it can also display nice menus for setting things up.
• Use a flexible interface for connecting the metering probe, such that the choice of light sensor isn't baked into the device. It should be possible to use different sensors as desired, simply by plugging in a different probe. While I'll initially focus on B&W metering, being able to function as a color analyzer is absolutely a stretch goal.
  
• Have enough program memory that all desired features can be implemented. There should never be a reason to need different versions of the device to offer different feature sets.
  
• Have enough user memory that settings and calibration profiles are not arbitrarily limited, and can even be accompanied with meaningful descriptions.
• Include a USB port, so that settings can be saved/loaded from thumbdrives. It should also be possible to use this port to connect to other peripherals. This could include keyboards, for typing profile names, or even densitometers to automate the process of creating profiles.
• Include a "Blackout" switch that turns off all illumination on the device, enabling the user to perform basic functions by relying entirely on audio cues. This will be of great benefit when doing color printing, where paper must be handled in total darkness.
• Able to run off all common mains voltages, without modification (except maybe a different fuse).
  
• All hardware and software for the device will be completely open source, so that anyone can build or modify it.
  

 

My anti-goals, or things I explicitly do not plan to include:

• It will not be a multipurpose darkroom process timer. That means no timing of film development. This timer is meant to sit next to the enlarger, not to be carried around and used for everything.
• It will not require any sort of smartphone app or computer interface to set this thing up. It should be possible to use the device as a completely standalone unit, where all of its capabilities should be locally configurable.
  

 

How will it work?

The device will consist of a main unit that looks somewhat like every other enlarger timer, and a connected metering probe.

 

 

Here's a basic block diagram that shows the main components:

The main unit has buttons, a display, illumination LEDs, a buzzer, and relays to control both the enlarger and the safelights. The metering probe has a button, to trigger readings, and a light sensor.

 

How far along is the development?

The first revision of all the schematics and PCB layouts are finished. I've constructed the first prototypes of the metering probe, and am in the process of constructing the first prototype of the main unit. Once that is done, I'll have to do a lot of testing and begin to write the software. Since this is all open-source, you can see the nitty-gritty of this work on Github.

As far as the theory and process of print exposure adjustment and metering, I've also done a lot of research to prepare for the project. In addition to what I can figure out on my own, it also helps that there are a lot of long-expired patents that go into a good amount of relevant detail on the subject. Additionally, I'm hoping I can make the calibration profiles for this device compatible with other existing devices, to make initial setup as easy as possible.

Will I be able to get one?

That is the hope. While the first prototypes are mostly for me to be able to tinker around with an enlarger timer/meter that I can reprogram myself, I would eventually like to make this into a real product. I'm not yet sure if it'll be a DIY thing or a complete constructed unit. Its also likely that there will be non-trival legal, regulatory, and manufacturing hurdles to make this happen. I'll also need to get the cost down, as nice-looking prototype enclosures can be quite expensive.

Nestronic Input Board (Part 4)

Introduction

When I designed the main board of the Nestronic, I was busy enough that I didn't want to hold up progress on account of other areas of the project. I basically just figured out what interface pins I might need for an input board, brought them out to a connector, and set that part aside.

I figured I'd need an I2C bus to interface with some sort of I/O expansion chip, and a few support GPIO pins. I also thought it might be neat to try experimenting with the ESP32's touch sensor capabilities, so I made sure at least one of the GPIO pins could be used for that.

(For more project background, please read the [Introduction], [Architecture], and [Prototype Assembly] blog posts.)

Building the Nestronic Prototype (Part 3)

(For background, please read the [Introduction] and [Architecture] blog posts.

Introduction

I eventually got to the point where the schematic was mostly finalized and the whole thing was working on a breadboard. While a big achievement in and of itself, it was far from a finished product.

First, breadboards usually look kinda messy. After all, they tend to evolve as you tinker with the circuit design. Also, breadboard wires are never the exact right length for any given connection. While you can make a clean-looking breadboard, it often requires far more time and effort than one is willing to put in for a simple prototype.

Second, breadboard connections are a little flaky. Breathing on your circuit the wrong way can cause this wire or that resistor to have just enough of a contact issue that something starts or stops working reliably.

Third, breadboards are noisy and full of stray capacitance. I could detect clock signals from one end to the other. Digital connections sometimes had just enough noise to make the difference between a high and a low. And to quote Dave Jones, you sometimes have to "hold your tongue at the right angle" to get things to even start up correctly.

Breadboard Nestronic

At this point I exported the netlist for my schematic, pulled it into KiCad's PCB tool, and began the layout and routing process. After a lot of work, I eventually ended up with a complete board design. The core of this layout was the NES CPU section, with all its parallel bus traces. The rest was kinda squeezed in around it. Perhaps not the most optimal of layouts, but good for the general size and shape of the enclosure I wanted to put it in.

PCB Layout

Nestronic System Architecture (Part 2)

(For an introduction to this project, please read the previous blog post.)

System Design

Now that the design of the system is mostly complete, I'd like to present it in a little more of a top-down fashion. First, I'll present a block diagram of the complete system. Then I'll go in-depth on how I arrived at all the relevant elements of the system. Finally, I'll show the detailed circuit schematics this all represents.

The design of the Nestronic consists of two major subsystems. First, there's the NES CPU Section, which builds around the RP2A03 CPU and contains everything necessary to actually synthesize video game music. Second, there's the ESP32 Microcontroller Section, which provided the front-end and completes the project.

Block Diagram