Using an Enlarger Meter?

[albada]

Are you using an enlarger meter? It sits on the easel and tells you the brightness of a spot in your image.

While I'm waiting for the meter from Darkroom Automation to arrive, I hacked together a meter out of a photodiode taken from a 1990's plastic SLR connected to a decent voltmeter. I convert millivolts to EV numbers using a table I printed. Here are three levels of sophistication I can think of:

Level 1: Use the meter to get a tone correct. You must determine (once) from a test-strip that a certain EV (at a given exposure-time) yields the correct skin-tone (or some other tone, such as near-black or near-white). From then on, for any negative, turn the aperture ring until the meter gives you that EV, and then that metered spot will print at that tone. Simple.

Level 2: Measure nearly-black and nearly-white points in your image, average them, giving you the mid-tone EV. Turn the aperture until that mid-tone is at the correct EV, thus ensuring the overall exposure is correct.

Level 3: Use the nearly-black and nearly-white numbers measured above to select the grade. To avoid the calculations and looking up a number in a grade-table, an analyzer such as from RH Designs can do this work for you.

I'm at level 1 (getting one tone right). Even that takes experience, as shown by my stupid mistake from yesterday: I measured a bright spot on somebody's forehead and used that as the skin-tone, which made everything else too dark. Duh. But even at my basic level, I have eliminated most test strips.

If you have a meter, how are you using it?
Mark Overton

 

[MattKing]

I do make use of an Ilford EM-10, but really only as a comparator. By that I mean that I have recorded information about what meter reading gives a useful mid-tone with a particular time of exposure with a particular paper. This allows me to adjust the light intensity and narrow down the range for my test strips - I just eyeball what I think would be a good mid-tone in a negative and then adjust the light intensity for a 16 second exposure at that point. My initial test strips are then made with half stop steps centred around 16 seconds (5.6, 8, 11, 16, 22, 32, 45).
It is also really good for adjusting exposure when magnification changes.
Finally, it makes preparation of contact proof sheets from my rolls of film much more predictable.

 

[distributed]

I am using a self built meter along with some calibration data I have recorded. This is still in the prototyping stage, so at the moment I only have one calibration dataset for the paper I typically use together with the developer I use.

With the enlarger stopped down, I measure the brightness of the projected negative and find minimum and maximum illumination. With these numbers, I look up the required contrast and required blue/green ratio to use with my LED light source. Given the contrast I look up the "toe point", i.e. the needed exposure for highlights and use that for exposure. In many cases the exposure proposed that way is a good starting point for dodging and burning. Sometimes a bit of fine adjustment, especially for highlights, is needed and easily done with a test strip with small increments, say 1/6 stop.

The meter could also be used to determine how much I need to burn in (extreme) highlights lying outside the selected paper range, but I haven't tried that yet.

You can find some HD curves used in the preparation of my calibration data in the following link. The information about the toe points is missing there, though.
Trying out Stouffer T2115 wedge on darkroom papers

 

[bdial]

I use a meter occasionally for exposure adjustments when I change print size, and I want to minimize test exposures for the new size.

 

[radiant]

One of my measurement problems were that it was difficult to find a good spot for measurement.

Let say if your calibrated measurement is based on highlights or shadows reading - what if there isn't such tone available in negative you are trying to measure? Or the area where you should measure is really small?

 

[distributed]

> One of my measurement problems were that it was difficult to find a good spot for measurement.

Sometimes I have trouble spotting extreme points, but I take multiple measurements across the baseboard and my setup only accepts the minimum and maxium values. One issue is that my update rate is low, about 2 Hz and thus I have to move slowly. I would like to add acoustic feedback so I realize when a min or max has been added. This is nice to know and would make redoing the measurement simpler.

This being said, there are situations where I can't measure what I would like to. One instance was specular reflections on small leaves that I couldn't accurately target. These were fortunately easy to correct with a bit of soft burning. In the case of not having a suitable highlight or shadow measurement because there is no such tone, I would revert to doing test strips, manually. I estimate that these situations are either easy to correct, like the specular highlight, or rare, like in the latter case. If they happen to be common, a min/max measurement system does not make much sense to use, of course.

 

[radiant]

Sometimes I have trouble spotting extreme points, but I take multiple measurements across the baseboard and my setup only accepts the minimum and maxium values. One issue is that my update rate is low, about 2 Hz and thus I have to move slowly. I would like to add acoustic feedback so I realize when a min or max has been added. This is nice to know and would make redoing the measurement simpler.

I have min/max too WITH audible signalling When my meter finds new min/max it makes different beeping sound. If I cannot hear the different beeps I know I have found all local min/max'es. And as we are using same sensor, mine is 2Hz too. It is a bit slow yes!

Distributed you are correct; min/max doesn't solve the problem still completely; you might measure "too bright" or "too dark spot" which in realy I don't want to use as reference.

No enlarging meters for me. No use for that. None of the best printers I know of have ever used them. Test prints. Use eyes.

I used meter but then went back to test strips and now we have had discussions with distributed I'm thinking should I try metering again But in any case I've found that a good eye and ability to analyze withouth devices is really needed skill. No meter can replace this.

 

[mrosenlof]

I have the enlarging attachment for my Gossen Luna-pro SBC. So far I have "used" it by forgetting that I have the thing. I'm possibly too attached to test strips.

 

[distributed]

@vedostuu: You're halway to having a Heiland Splitgrade system then! :-D

Regarding the sensor, a TSL 2591, there are faster integration modes, but I haven't tried them yet as I don't have a lot of dynamic range to spare.

 

[logan2z]

I use the RH Designs Analyzer Pro. While it's certainly not necessary, it does allow me to get into the ballpark on exposure/contrast much more quickly. I rarely make test strips anymore.

 

[radiant]

@vedostuu: You're halway to having a Heiland Splitgrade system then! :-D

Regarding the sensor, a TSL 2591, there are faster integration modes, but I haven't tried them yet as I don't have a lot of dynamic range to spare.

With my pre-LED conversion halogen-setup enlarger the light was so dim that for dense parts I could get only few digits apart so I didn't even try the faster integration parts. I needed to take every performane from the meter. BTW: this might be issue for other meters too if you like dense negatives.

 

[Sirius Glass]

I have one. I read the instructions. Turned it on. And I have never used it, because when I get a perfect print, I move on the next negative.

 

[dkonigs]

I have an RH Analyser Pro, and I absolutely love it. My approach then is probably what the OP calls 2/3. I use it to get a feel for the contrast range of the negative, pick a starting time, and see what contrast grade works the best. While I do still often make test strips, its extremely common for me to actually not adjust my base exposure much (or at all) based on them.

Of course when I'm printing color, the device gets reduced to what is essentially just an f-stop timer. However, even without being able to leverage metering, the f-stop timing approach seriously helps narrow down the exposure quite quickly. (Processing overhead with color is greater, so reducing the number of "attempts" is of higher priority.)

Of course I'm also actually in the middle of developing my own enlarger timer/meter based off of my experiences with the RH product. First and foremost, I see a lot of little opportunities to improve things when not hamstrung by the limitations of built-to-cost embedded technology from 1996. Beyond that, I'd love to tackle the color problem, but not yet sure how to do it best, or which sensors will give the best data.

 

[R.Gould]

I have used an RH designs analyser/pro for many years now, best piece of darkroom kit I ever bought, 99% of prints are correct first time, and I haven't made a test strip for many years, only time I make test strips is if I need to calibrate a new paper, save me a lot of wasted paper, hence a lot of money and it doesn't take over for you, I still use my own input based on the way I want the print to look, gives me dodging/burning times using the grey scale

 

[Frank53]

No enlarging meters for me. No use for that. None of the best printers I know of have ever used them. Test prints. Use eyes.

That doesn't mean these gizmos are bad. I've just never seen the point.

yes, they do use them, here’s one:
The best meter there is imo

 

[albada]

I'm also actually in the middle of developing my own enlarger timer/meter based off of my experiences with the RH product. First and foremost, I see a lot of little opportunities to improve things when not hamstrung by the limitations of built-to-cost embedded technology from 1996. Beyond that, I'd love to tackle the color problem, but not yet sure how to do it best, or which sensors will give the best data.

I'm impressed with your Printalyzer project. I read over your web-page for it, and have a few suggestions:

1. LED-based lamps are becoming more common in enlargers, and they provide new capabilities that a modern analyzer should take advantage of. Instead of merely turning the lamp on or off, you can (1) set the brightness level of some LED-controller chips using resistance or voltage, and (2) they can dimmed using PWM. I think PWM with TTL-level signals is more common than analog dimming. Anyway, the ability to set the power-levels of individual LEDs allows one to select any contrast in B&W, and to correct for color when printing color. The UI should let the user set individual LED power-levels. Actually, I use log2 attenuations, where 0=full power, 1=half power, etc. Such attenuations blend well with the EV system.

2. Allow different times for the three LEDs, and allow the LEDs to run sequentially or concurrently (to save time). With f-stop timing, the user can adjust an individual color with LED-time as well as LED-brightness. In sequential mode, allow the red LED to turn on for a few seconds before green or blue turns on. This feature will give the user time to position a dodge/burn tool, and also makes split-grade printing effortless.

3. Make the front part of the metering probe (containing the sensor) very thin. When making small prints in 35mm, the enlarger's head is so low that a sensor-height of 10 or 15 mm results in a significant error. Also, the top surface of the probe (surrounding the sensor) should be white to make the image visible to the user.

4. The sensitivity of a photodiode is proportional to its area, so for highest sensitivity, I suggest that its area be at least 5 mm^2.

5. If the display has a backlight, I suggest that a vertical shield be located behind it and to its sides to prevent its light from affecting the metering probe. When metering a dense area on a negative, such a light would cause a large error. An alternative would be to place an EV-display in the metering probe itself, which can be lit by a backlight (shadowed from the sensor), or by the enlarger's light (that's what I'm doing, but it's often hard to read the digits).

Anyway, I hope your Printalyzer goes into production!
Mark Overton

 

[dkonigs]

Its probably time for a dedicated thread on the project, and another blog post, but I'll answer these initial comments/questions here:

1. LED-based lamps are becoming more common in enlargers, and they provide new capabilities that a modern analyzer should take advantage of. [...]
2. Allow different times for the three LEDs, and allow the LEDs to run sequentially or concurrently (to save time). [...]

Sophisticated handling of LED enlarger heads makes more sense as a follow-on project. I'm not aware of any sort of "standardization" in how these are controlled, and I think many of them are likely to come with their own controller units. Thus, doing this requires working against a bucket-list of "What specific enlarger heads am I controlling?" and possibly even making a head-specific "adapter box." Its not a bad idea, but its a lot more complicated (and hard to generalize) than simply turning an outlet on/off.
Similar arguments could be made for multigrade heads, and anything where the timer interface is more complicated than on/off.

3. Make the front part of the metering probe (containing the sensor) very thin. When making small prints in 35mm, the enlarger's head is so low that a sensor-height of 10 or 15 mm results in a significant error. Also, the top surface of the probe (surrounding the sensor) should be white to make the image visible to the user.

The meter probe used by the RH Analyser is actually 25mm thick. Mine, around the sensor, is only about 15mm thick. The distance from the bottom to the top of the sensor itself is a tad under 10mm. I'm not sure how much lower I can go without causing problems for mounting hardware. The top of my current prototype may be black, but the plan has always been to cover it with some sort of sticker that has a transparent window over the sensor itself (like the RH probe has). That sticker will be a lighter color, regardless of what color I pick for the probe body.

4. The sensitivity of a photodiode is proportional to its area, so for highest sensitivity, I suggest that its area be at least 5 mm^2.

The window above the sensor is 5mm in diameter. The sensor itself is a modern digital light sensor, where I'm limited to what's actually offered. There are many choices here, and my design intentionally allows me to build a variety of meter probes with different sensors as I see fit. However, the sensor itself is a very tiny photodiode array encased within a tiny transparent rectangular package.

5. If the display has a backlight, I suggest that a vertical shield be located behind it and to its sides to prevent its light from affecting the metering probe. When metering a dense area on a negative, such a light would cause a large error. An alternative would be to place an EV-display in the metering probe itself, which can be lit by a backlight (shadowed from the sensor), or by the enlarger's light (that's what I'm doing, but it's often hard to read the digits).

The light from the display does not get picked up by the sensor, so this is not a problem. (It can detect safelights, but those are turned off during metering.) I do have a filter on top of the display so its spectrum won't fog paper. For the indicator LEDs around the device, I can just pick them in parts of the spectrum where I shouldn't have to worry about this. The brightness of everything is also adjustable.

Anyway, I hope your Printalyzer goes into production!

So do I. I've actually built the first prototypes of the unit, so its now a functional piece of hardware I'm working with. Right now I'm neck deep in working on the software for this thing, starting with the basics of f-stop timing, test strip modes, and related functions. Obviously the to-do list is pretty long, and I'll be working on this for quite some time.

My greatest worry with production is that "certification and liability protection" (UL/CE/CSA/FCC/etc) may end up to be a big enough stumbling block that its hard to justify for a small-run "boutique" product. However, that's probably a subject for different forum.

 

[radiant]

I'm not aware of any sort of "standardization" in how these are controlled, and I think many of them are likely to come with their own controller units.

I would build "intelligent" port with following possibilities at the same connector (configured at software):

- R/G/B relay control; 5V out to external relays
- R/G/B PWM; similar to relay control and 100% software
- UART out; one signal is UART TX pin, simple 9600bit/s ascii protocol "R;G;B\n"
- WS2812 LED control (basically one digital pin is needed for this)

So you can tackle a huge amount of "standard" and make it possible to others integrate to your Printalyzer. If you really want to modernize / update you aren't only controlling line voltage to halogen lamp.

Maybe even ease up on certifications by taking line voltage out of the device and run it with USB power. Maybe build external line voltage relay box using "intelligent" connector.

Good luck with your project, hope it succeeds!

 

[jim10219]

I had an Ilford EM10. I discovered that it wasn't accurate enough to avoid making a test strip, so I just went back to making test strips. I sold it because it took more time to get the results I wanted with the device than it did without.

 

[albada]

Its probably time for a dedicated thread on the project, and another blog post[...]

Please do so; many of us will be interested in this.

The meter probe used by the RH Analyser is actually 25mm thick. Mine, around the sensor, is only about 15mm thick. The distance from the bottom to the top of the sensor itself is a tad under 10mm. I'm not sure how much lower I can go without causing problems for mounting hardware[...]

The RH meter-probe is 1 inch thick?! That will cause prints to be noticeably lighter when working low because the inverse-square law hurts quickly.
Here is a sensor that is 3 mm thick: https://www.digikey.com/en/products/detail/excelitas-technologies/VTB8440BH/5885859
If you support it from above, bend the leads sharply at 90 degrees, and cover the bottom with thin material, I think you could reach 5 or 5.5 mm.
But that's a bare-metal sensor; you would need to supply your own analog circuitry and A2D to read it.

I am presently using a similar bare sensor connected to a Kaiweets multimeter that has a resolution of 0.1 millivolt. This crude arrangement works well and provides sufficient sensitivity. So I think it would not be difficult to design your own analog circuitry around such a sensor. But I wonder if temperature-sensitivity would be a problem for such a meter (winter versus summer in the darkroom).

Mark Overton

 

[Rick A]

I do make use of an Ilford EM-10, but really only as a comparator. By that I mean that I have recorded information about what meter reading gives a useful mid-tone with a particular time of exposure with a particular paper. This allows me to adjust the light intensity and narrow down the range for my test strips - I just eyeball what I think would be a good mid-tone in a negative and then adjust the light intensity for a 16 second exposure at that point. My initial test strips are then made with half stop steps centred around 16 seconds (5.6, 8, 11, 16, 22, 32, 45).
It is also really good for adjusting exposure when magnification changes.
Finally, it makes preparation of contact proof sheets from my rolls of film much more predictable.

+1

 

[eddie]

With the rising costs involved in analog printing, I think any invention which gets you to your desired results, while using less paper, should be applauded.

 

[Craig75]

No enlarging meters for me. No use for that. None of the best printers I know of have ever used them. Test prints. Use eyes.

That doesn't mean these gizmos are bad. I've just never seen the point.

they def arent going to make you a better printer.

but you can spotmeter the scene after the event on the easel and calculate times and grade you need to line up whatever it is yr trying to do quickly without making strips or use the data to feed into development times you might need to copy something or make a mask.

(he says owning one but never having done any of this)

 

[albada]

I don’t think it’s for me. My prints often have a lot of detailed/complicated manipulations and I think I’m as well off the old fashioned way. Photography involves enough junk as it is, so I prefer to keep things simple wherever I can, and when it comes to printing, I just don’t see the value of metering devices.

I often photograph people, and being able to print skin-tones correctly on the first try without a test strip saves me both time and materials.
I also photograph classic cars, and being able to set both exposure and grade based on shadow and highlight measurements saves me one or two test strips, but I've had little practice at this. This is level-3 in my original posting, and I'm still perfecting level-1 (getting a tone right).
Those shadow/highlight measurements also tell me where I'll need to dodge or burn.

BTW, when I need to create a test strip, I do so MattKing's way: Using f-stop timing with the following half-stop times: 4, 5.6, 8, 11, 16, 22, 32, 45 seconds.
Mark Overton

 

[revdoc]

There's no need to overthink this; it just isn't that complicated.

I use an EM10, and I get the right exposure 9 times out of 10. Here's what I do.

1. Set the EM10 to 80. (Chosen to give an aperture between 5.6 and 11 most of the time.)
2. Put the sensor under a highlight.
3. Safelight off, no contrast filter.
4. Adjust aperture to get a green light.
5. From testing I know that exposure for that spot is 10s for slight tone only, 20s for slight tone and texture, 30s for light grey. (Zones IX, VIII and VII, if you prefer.) I make a choice based on how I want the print to look.
6. I choose a contrast filter and make a pilot print on a 1/4 sheet to verify my choice of exposure and contrast, followed by a complete print.

If you meter the highlights, you can change the contrast without altering the exposure.

Even if you try for 100% automation, you still have to make the decisions in steps 5 and 6, because no machine knows how you want your prints to look.