Densitometer design, construction and calibration

Hi,

I would like experiment with a transmission densitometer, mainly to have a more solid/scientific approach when comes to development times (N, N+ or N-). However the cost is putting me a bit off and that's why I am thinking to go DIY. This article HERE seems to provide a good step by step guide to build your own densitometer and components seems to be broadly available online (there is a components list at the end of the article).

Do you think it will work? Looking at densitometers online that costs well above £500 it seems to good to be true that with about £50 of components you can built something similar...or am I missing something?

Ferru

1. Having a step tablet of known density is essential so you can calibrate the densitometer. A local photofinisher or radiography department my allow you to read your step tablet.

2. A diffuse densitometer will closely represent the results from a diffuse enlarger if you take flare into account. A condenser enlarger will have quite different results because of Callier Q Factor.

DarkroomAutomation.com sells an enlarger meter for a bit under US$100. It doubles as a transmission densitometer when used as follows:

1. Put the meter on the easel or baseboard of your enlarger, turn off the safelight, remove any filter in the enlarger and turn it on.
2. Put a clear (unexposed) part of the negative over the meter's sensor, and press the right button to put it into relative mode.
3. Place the spot of the negative you wish to measure over the sensor, and the meter will give you its relative density in stops.
4. If you want logarithmic density, multiply the measurement above by 0.301.

This meter is made by Nicholas Lindan, who posts here regularly. The easiest way to buy it is to email him and pay with Paypal.

I know the trend is to make densitometers with electronics and computer modules.

But it's not hard to make a densitometer from an ordinary meter. All you need to do is make a window you can put the film sample over, and then "null" the meter without film... put the sample in and count how many f/stops of light were blocked by the light. Each stop is 0.30 density. Some meters are calibrated in third-stops, so each mark is 0.10 density.

I have an old book that gives instructions to make a densitometer with a GE meter, and those are very cheap.

One option is to use standard flatbed scanner (see page 5):

https://www.scotty-elmslie.com/uplo...tical_guide_to_film_characteristic_curves.pdf

I've build DIY meter for direct film measurement but I think it was a bit difficult to use so I typically measure my negatives with enlarger: https://www.photrio.com/forum/threads/diy-light-meter-for-darkroom.169398/page-4

FerruB said:

Hi,

I would like experiment with a transmission densitometer, mainly to have a more solid/scientific approach when comes to development times (N, N+ or N-). However the cost is putting me a bit off and that's why I am thinking to go DIY. This article HERE seems to provide a good step by step guide to build your own densitometer and components seems to be broadly available online (there is a components list at the end of the article).

Do you think it will work? Looking at densitometers online that costs well above £500 it seems to good to be true that with about £50 of components you can built something similar...or am I missing something?

Ferru

Click to expand...

just my opinion:
Yes, a good densitometer,such as the Heiland, will cost about$800 unless you buy used but, I seriously doubt a £50 DIY job can compete. The good news is: the Heiland is a pecise high-quality tool lasting decades and once you have a densitometer ,You'll find numerous uses for it in analog as well as in digital photography.And no, I'm not in any way sponsored by Heiland Electronics,sust a satisfied customer and frequent user.

I think a transmission densitometer is less expensive than a scanner.

Just depends if you want a toy or something serious. It's a lot easier just to find a good used one. Sometimes I'm reminded of some old Popular Mechanics-style article explaining how to build your own nuclear submarine using spare washing machine parts. I'm quite a jerry-rigger myself, and have made the majority of my own darkroom equipment. But when it comes to two categories - optics and electronics - well, there's that old Clint Eastwood movie remark, "A man's got to know his limitations".

I’m using the Marshall Studios Densitometer tonight. Precision to 0.02 and a primary instrument. Never have to worry about a part burning out or having it go out of calibration.

Can you actually measure 135 film frame details with densitometer? Or how small area can you spot and meter from negative with commercial products?

ic-racer said:

I think a transmission densitometer is less expensive than a scanner.

Click to expand...

interesting point.

vedostuu said:

Can you actually measure 135 film frame details with densitometer? Or how small area can you spot and meter from negative with commercial products?

Click to expand...

looking at mine, I'd guess about 1mm diameter.

RalphLambrecht said:

looking at mine, I'd guess about 1mm diameter.

Click to expand...

Is it easy to pinpoint that spot from negative?

vedostuu said:

Can you actually measure 135 film frame details with densitometer?

Click to expand...

With a conventional bench densitometer? Not really. The measurement spot is a few millimeters across so the "details" have to be pretty big.

The alternative is to use an "easel densitometer." This is a mode available on many high precision exposure control systems made by ZBE, Wallner, RH Designs and, of course, Darkroom Automation.

The negative is loaded into the enlarger and the density of the negative is determined by measuring the projected image. The meter is zeroed on the projection of the clear film rebate. After that the meter is moved to the desired spot and the density over b+f is read out on the meter. The advantage is that the Callier effect and system flare are both incorporated into the measurement, giving the negative's effective density. This is also the disadvantage to the method.

Nicholas Lindan said:

With a conventional bench densitometer? Not really. The measurement spot is a few millimeters across so the "details" have to be pretty big.

Click to expand...

I built myself own densitometer with quite small measuring spot and got realiable results from it. I faced the same issue with that I couldn't really measure any particular smaller spot from negative on purpose. I then switched to analyzing my negatives on enlarger which works so much better. I thought commercial products were somehow better on this like easier to somehow see what spot are you measuring etc.

Let me be clear with the scanner reference previously: if you already own scanner, you might have densitometer available already (with much more precise spot measuring - it seems).

Along with a regular XRite transmission densitometer, I have the ZBE easel densitometer for reading projection densities. It' s deliberately engineered to be color blind or neutral, so can be used for either black and white film or color film, provided you want basic neutral density values. In other words, it doesn't have give separate CMY readings. Nor is it an enlarging meter, but reads in logarithmic units like other densitometers, which can then be translated into time if needed with a simple math calculator. It's also way more accurate than enlarging meters (and I have an especially good one of those too), with a far wider density range. But it works accurately only for on-axis readings; there is no cosine correction for off-center portions of the projected image. To read those, you'd have to shift around the negative in the carrier itself. Despite those limitations, it's still one of the most useful, and most used, darkroom devices I've ever purchased. It makes a lux meter seem like a cheap bathtub rubber duck by comparison. Very few of these portable units were ever made, and they were expensive, and differ in design and purpose from their enlarger feedback units.

I can't even measure details on my 8x10 negatives with my densitometers. They even have the 'optional' small 2mm measuring windows.

vedostuu said:

Can you actually measure 135 film frame details with densitometer? Or how small area can you spot and meter from negative with commercial products?

Click to expand...

If its a normal photograph? Probably not. But if you're trying to calibrate a process, then you'll either expose the film with a sensitometer or by photographing a large frame-filling target. In those cases, it should absolutely be doable.

dkonigs said:

If its a normal photograph? Probably not. But if you're trying to calibrate a process, then you'll either expose the film with a sensitometer or by photographing a large frame-filling target. In those cases, it should absolutely be doable.

Click to expand...

For sure it is possible if the whole frame is in same density.

I have always thought densitometers are used on normal photographs And I just wondered how the densitometers work to get such accurate reading..

RalphLambrecht said:

just my opinion:
Yes, a good densitometer,such as the Heiland, will cost about$800 unless you buy used but, I seriously doubt a £50 DIY job can compete. The good news is: the Heiland is a pecise high-quality tool lasting decades and once you have a densitometer

Click to expand...

Like many of Heiland's products, I'll bet that the actual "bill of materials" for their $800 densitometer is closer to $50. Okay, the nice metal shell could make it closer to $150 or more (enclosures are often underestimated in cost) but still. What you're really paying for is R&D and calibration on a low volume niche product.

The hardest part about making a densitometer, IMHO, is calibration and eeking out accuracy at the extremes of its range.
Also keep in mind that with any electronics project, you're absolutely going to spend a lot more yourself on developing it than the cost of components in the end result.

So do I think its possible to construct such a thing for <$50 worth of parts? Probably. But you'll likely spend more money on the project itself than the cost of buying an older used unit off eBay or elsewhere. (just keep in mind that you still need a calibration reference, no matter what you buy.)

I do it with sheet film actual images rather frequently. No problem. 1mm aperture (the aperture size is interchangeable in my densitometer) In theory, it's no different with a 35mm original. It's just that 1mm diameter overlaps a bigger portion of a very small negative. Same when a piece of dust is on your film when it gets exposed. With 8x10 film the effect might be so small overall that nobody notices it; in a 35mm shot with the sky included, it proportionately looks like the Goodyear Blimp.

Transmission densitometers conventionally have cross hairs on an illuminated area to help you guide your film into exact position. One trick is to circle the exact spot you want to read using removable ink on the non-emulsion side of the film, then remove that with film cleaner. Or else you can tape your film to a larger piece of frosted mylar, which itself shows your circle, and either null out the density of the mylar itself first, or subtract it from your reading afterwards. Easy enough, even for 35mm negs.

Phil Davis has plans for building a simple densitometer using a Pentax spotmeter in his "Beyond the Zone System" (Appendix 2). You can kind of see the setup in this article on the BTZS website here

https://www.btzs.org/Articles/Sensitometry Part 2.pdf

I couldn't afford a dedicated densitometer but was lucky enough to find a working spotmeter that I set up with an old Olympus lens for focusing (I think it was a 49mm adapter ring but not sure). Anyway, it was pretty straightforward and works fine

Regards
Dave

Like others, just out of curiosity, I tried adapting one of my Pentax digital spotmeters for use as a densitometer and enlarging meter - NO WAY. Not even remotely in the league or utility level of the real deal, and I'm pretty darn good at jerry-rigging. It would need to read far lower light levels and have a far wider range, plus more accuracy. To do a densitometer plot, you need complete linearity over the whole distance. Once again, my wise crack about trying to build a nuclear submarine out of spare old washing machine components applies, Phil Davis or not. If you want a fun clever so-so DIY project, why not? But if you want a real densitometer, simply buy one!

Incidentally, quality densitometers are not just overpriced, cheaply manufactured items. A high level of quality control and skilled calibration protocol is involved. Companies like MacBeth and X-Rite knew what they were doing. In fact, my wife worked with a 6 million dollar propriety XC-Rite spectrophotometer in Biotech quality control, when designer vaccines and pharmaceutical prototypes were running around 40K per cc !! No room for error. That dang machine was such a trade secret that her workspace was behind four-foot thick concrete walls behind an actual timed bank vault door. The software itself was known by none but the company founders themselves. So real engineering expertise is behind even their far lesser devices, as well as in the mid-level XC-Rite industrial spectrophotometers I used for my day job. None were mere toys, by any means, not even my humble darkroom transmission densitometer.

I used my spot meter for many years as a densitometer. My hand-held meter is very linear and has at least as good of an output of data as a densitometer. Of course it has, otherwise it would be a poor meter. Internally there is no difference in the circuitry, but the densitometer has a way to hold the sensor in the correct place from reading to reading.

FWIW, I'm actually currently working on attempting to build a reflection densitometer (as an accessory to my Printalyzer project, or perhaps as a standalone device). The geometry is different, but the problems do overlap. Also, many off-the-shelf devices do both functions.

In the process of doing this, I have taken a very close look at how my X-Rite 810 and my Heiland TRD-2 are designed. You really can attribute a lot of the differences to the available technology of when they were built.

The X-Rite has 4 channels (R,G,B,V) and was designed in the 80's, when good ADCs (analog-to-digital converters) were still a more novel thing. So it has a very complicated read head design (one or two photodiodes per channel), and dedicates a lot of its circuitry to a not-exactly-integrated ADC and to multiplexing the use of that ADC. The digital side is rather simple.

The Heiland was designed somewhere around 2010-2012 and has only 1 channel. This was after LEDs were able to do some things in this area, but before integrated digital light sensors became available. So it uses an LED in the base and another in the head, and uses a single photodiode connected to an ADC chip. The digital side isn't very different, but uses more modern components (so its smaller and simpler in implementation).
The main failing of the Heiland design is that it doesn't care as much about the spectrum where its measuring, so it will read colored targets with different values than the X-Rite (on its "visual" channel). But for B&W targets, its probably not an issue. Of course this could probably be solved with a different photodiode or the right filtration on top of the one they used.

In my own project, I've found that its actually pretty easy to get good consistent data between the "low" and "high" patches of a typical reflection calibration target (e.g. D=0.10 through D=1.70, give or take). The problem is that once you get into the really dense targets, because its a logarithmic scale, even a minuscule error can seriously throw off your readings.