Rolf-Werner
This weekend, I’m going to start working with Duka again (after a break of about 25 years...). I’ve ordered some standard paper to start with (a few sheets of Foma PE with a fixed gradient).
What lighting is recommended for this? Is red sufficient, or should it be green-brown? (I’ve still got both lying around.) I haven’t been able to find any information on this.
Thanks for your tips!
Rolf
Renate
I would definitely recommend red. You should test green-brown first. With red, there’s less chance of problems, but even then a quick veil test would be advisable.
€
Best regards
€
Renate
Rolf-Werner
OK, thanks for the tip!
So what are the green and brown ones actually for?
Rolf
Rolf-Werner
So, I’ve just had another look. I’ve still got a three-colour lamp there; it’s got orange, green and ‘Colour’ (really dark).
Then two bulbs in – well – a sort of green-brown colour.
My green-brown Duka lights aren’t here; they’re stored in my old flat.
So there’s no pure red.
What would you go for? Just stick a lamp in a red plastic box? That’s what I did in my first darkroom when I was 14: the developing bottle (Tetenal?) was made of red plastic, so I stuck a Fischertechnik lamp in it, and there you go – darkroom lighting sorted :-)
Rolf
Wolfgg
Hi Rolf,
You can test any Dukalamp yourself: place a small strip of paper under the enlarger in the dark, switch it on briefly so that the paper comes out of the developer slightly grey, but first place a coin on the paper; now switch on the Dukalight and leave it on for as long as it takes to take a sheet of paper out of the box, place it under the enlarger, perform exposure and then add another 30 seconds. After developing and fixing, nothing from the coin should be visible on the paper. Then the Duka lamp is suitable. Done.
Regards, Wolfgang
pittyman
Hi Rolf,
You can test any Dukalampe yourself: place a small strip of paper in the dark under the enlarger, switch it on briefly so that the paper comes out of the developer slightly grey, but first place a coin on the paper; now switch on the Dukalampe and leave it on for as long as it takes to take a sheet of paper out of the box, place it under the enlarger, perform exposure and then add another 30 seconds. After developing/fixing, nothing from the coin should be visible on the paper. Then the DuKa lamp is suitable. Done.
Regards, Wolfgang
That is, of course, nonsense. The test as such is perfectly correct, but the enlarger remains switched off the whole time! After all, the point is to see whether the DuKa lighting fogs the area around the coin, not the enlarger. You can also handle the whole thing like a test strip. The long side is completely covered for half the length, and the short sides are gradually exposed to the DuKa lighting, just like with a test strip. This way you can see when the paper starts to become fogged. For me, it takes more than 10 minutes.
Best regards
Dirk
TobiasCallenius
You must pre-expose the paper; otherwise, there is a risk that, whilst the red light on its own does not cause any fog, the exposure of the image in the enlarger will cause the subliminal pre-exposure from the DuKa lamp to become supraliminal. Therefore, the fog test is completely worthless without pre-exposure.
Best regards,
Tobias
Renate
The test described by Wolfgg is a simple version of the test that Ilford experts consistently and strongly recommend in their publications. The test is certainly not nonsense. Pre-exposure with the enlarger is necessary because it exposes the paper beyond the threshold value. Only once the threshold value has been reached does any further light cause the paper to turn grey. The paper has thus become ‘more sensitive’. As one usually works with exposed paper in the lab, one should also test the darkroom lighting with exposed paper.
?
With fixed-grade paper, as was common in the past, green-yellow is certainly a good choice, as the paper is only blue-sensitive. However, as multigrade paper is also blue-sensitive, one must now use orange or red. Since the eye does not perceive these secondary colours, a test is essential. My red LEDs, for example, also contain a small amount of green light, which can then cause a fog on multigrade paper. An additional red filter solves the problem without making the LEDs appear significantly dimmer.
?
?
Best regards
?
Renate
Rolf-Werner
Thanks for all the tips! I think I’ve read something like that somewhere before. Was it here on the forum?
So for the standard-grade paper I’m planning to use for now, the colour doesn’t really matter… Does it?
I’ll give the test a go anyway, though – I’d be interested to see the results. I’ve never done it before. That reminds me, it obviously depends on the distance from the lamp as well. If, like me, you’re working in a larger room with perhaps several lamps, you’d need to test it in several different spots.
Rolf
Morte
I’d say: you can’t go wrong with a red lamp. Even if you’re using fixed-grade paper now, who knows what you might want to do in the future?
I’d also recommend a new bulb, even though you already have some. Over the course of 25 years or more – given that they’ve been in use for a while now – filters, housings and so on can change their properties, colours can fade, and so on. A new (darkroom) lamp (not a party light or similar) doesn’t cost that much and is a good and important new investment to avoid hassle and failures. You don’t use 25-year-old paper, do you...
I myself use the red light bulb available from FOTOIMPEX. It is of good quality and does the job well.
http://www.fotoimpex.de/shop/fotolabor/dunkelkammerbirne-rot-230v-15w-e27.html
Incidentally, contrary to what is stated in the shop, it is no longer from the Narva range.
Wolfgg
Dirk surely knows who’s talking nonsense and just wants someone to make the effort to lay out the exact background :).
So: we’re talking about the keywords density curve, characteristic curve, densitometry (search engine).
Here is an example of a characteristic curve, which, incidentally, applies to both film and paper:
http://de.wikipedia.org/wiki/Dichtekurve
(then click on ‘Density (photography)’)
€
The X-axis represents the amount of light falling on the paper (or film), the Y-axis the density (=amount of silver). If the Duka lamp test is carried out with completely unexposed paper and the light contains spectral components to which the paper reacts, then the curve is traversed slowly from left to right, starting approximately at the point where the axes intersect. The crucial point now is this: as long as the amount of light reaching the paper is not yet sufficient to exceed point A, the paper remains white after development and the misleading impression arises that the Dukalamp is safe! If the paper is now exposed to additional light from the enlarger, areas appear on the paper that lie on the curve between B and C, where the paper reacts completely differently – namely, much more sensitively to harmful light than before point A – and immediately adds any Duka light to the enlarger light if the light colour falls within the paper’s speed range! This is precisely the aim of the coin test: to use brief pre-exposure with the enlarger to shift the paper beyond point B into the steeper range, where it reacts immediately to the smallest amounts of light (as will be the case during enlargement), before the coin is placed on top as a ‘light barrier’ for the light from the Duka lamp. Only when the paper shows no trace of the coin do you know that the Duka lamp is safe.
Regarding light colour: red suits all papers (to my knowledge, Panchromatic paper for CN films has not been available for a long time), the only drawback being that the image appears slightly higher in contrast than it will later in daylight; you simply need to take this into account. In my Duka, I use yellow LEDs (12-volt plug-in power supply, potentiometer as a series resistor, set so dim that I can just about see what’s where; this even works for RA4).
In case you didn’t know: never develop paper by eye, but by time; ideally with the emulsion side down, so it receives hardly any Duka light in the tray. If the image is too dark, *always* reduce the exposure time, never the development time.
Regards, Wolfgang
pittyman
Hello Wolfgang,
I suppose I’ll have to take that nonsense back! I was already aware of the density curve, but in relation
to the DuKa lighting, I’d never considered it before. You never stop learning, thank you for the detailed
explanation. I suppose I’ll have to check my lights again to see if they’re up to scratch, for better or worse, now
with a bit more expertise. Thank you, Wolfgang, and please accept my apologies for accusing you of spreading misinformation!
Best regards,
Dirk
Rolf-Werner
I’ve completely overlooked the fact that there’s a second page again...
OK, I already knew about the characteristic curve and all that, I’ve just never put it into practice. My approach has always been: just enough light so that you can’t walk in front of it. The white base plate on the enlarger is easy to see anyway; the rest is down to playing with the shadows :-)
I’d already thought about yellow LEDs, and red ones too. Do the red ones perhaps not have the right spectral line?
How many LEDs have you connected together? In parallel or in series? What value are you using for the potentiometer? Maybe I’ve got something suitable in the workshop...
Wolfgg
Hello Rolf-Werner,
These are some ancient 5mm LEDs from the 1970s with the low brightness typical of that era (3 mcd at 20 mA); they are connected in series: 4 yellow LEDs, a 150-ohm resistor and a 1 kΩ potentiometer. Connected to this is an unregulated 9V DC mains adapter, which rises to around 12V under the low load.
If you were to build this using today’s much brighter LEDs, you’d need to increase the resistance values by about tenfold. It’s best to test this in the Duka; for example, I would simply use a 1kΩ fixed resistor and a 10kΩ or higher potentiometer and try out which range of resistance works so that the brightness can be reduced to a level where ‘larger objects are just about recognisable’. Of course, you should wait until your eyes have adjusted to the darkness. If there are more than 4 LEDs, the voltage must be increased or a second chain of 4 must be connected in parallel, as the forward voltage is 2.7V per LED.
Red LEDs work for B/W, but not for RA4, if only because the eyes’ speed in detecting red light drops significantly at night compared to green and yellow, which must be compensated for with correspondingly brighter red light; see:
http://de.wikipedia.org/wiki/V-Lambda-Kurve
Regards, Wolfgang
Rolf-Werner
Great, I'll have a go at that – thanks a lot!
Rolf-Werner
I’m reviving this thread because I’ve just had an idea. So please don’t be angry... :)
€
Using my digital camera (Nikon D50), I took a photo of a red light bulb (not a Duka bulb) and a red LED, then checked the histogram in a good image-processing programme. And lo and behold: you can actually see just how narrow-banded the light is.
€
Here are a few screenshots:
€
[ATTACHMENT NOT FOUND]
€
These are four test images: for comparison, the laptop screen, then a normal red bulb turned into a lamp and viewed directly, then indirectly, and a red LED very close up (some overexposure).
€
[ATTACHMENT NOT FOUND]
€
The histogram shows only the blue component. Here, of course, it is very high.
€
[ATTACHMENT NOT FOUND]
€
There is also quite a lot of blue in the red bulb, probably partly because the colour isn’t applied so evenly at the very top. Furthermore, the bulb has slight overexposure, which naturally also creates blue echoes on the camera sensor.
€
[ATTACHMENT NOT FOUND]
€
When it is merely reflected, the blue light is no longer as strong.
€
[ATTACHMENT NOT FOUND]
€
However, the LED shows even less blue, even though it had overexposure.
€
[ATTACHMENT NOT FOUND]
Right, I’ll add this here too: this is the LED measured again by the camera, so it’s not overexposed, and you can see how little blue there is.
€
Wolfgg
Hello Rolf,
Digital cameras aren’t suitable for serious spectral measurements, not even the expensive ones. The built-in colour filters are on a par with gelatine filters, and, more importantly, the spectrum needs to be divided into far more than just three bands.
Here are a few ‘proper’ measurements for comparison, taken with professional equipment – namely the GretagMacbeth i1 Pro, which is so revered by the EBV crowd. This spectrophotometer does not divide the spectrum into just 3 bands (“red, green, blue”) as is the case with digital cameras, for example, but measures from 350 to 740 nm at intervals of approx. 3.3 nm with an actual resolution of 10 nm, making it extremely accurate. Technically, this is achieved using a diffraction grating and a CCD array positioned behind it. In case anyone wants to verify the measurements: the software used was “Argyll”, a very flexible freeware command-line programme (exact command for light measurement: illumread -v -S -H Licht.dat)
Here is the homepage:
http://www.argyllcms.com/
Here is the spectrum of a red LED:
[ATTACHMENT NOT FOUND]
here is that of a yellow LED:
[ATTACHMENT NOT FOUND]:
here is that of a green LED:
[ATTACHMENT NOT FOUND]
and here, for comparison, the Rowi Duka lamp:
Red filter:
[ATTACHMENT NOT FOUND]
Green filter:
[ATTACHMENT NOT FOUND]
Orange filter:
[ATTACHMENT NOT FOUND]
And here is the dark olive filter for RA4:
[ATTACHMENT NOT FOUND]
Under ‘Fact Sheets’ here, you’ll find the data sheet for Multigrade IV with the spectral sensitivity:
http://www.ilfordphoto.com/products/product.asp?n=26
As you can see from the comparison, the yellow LED starts where the paper ends, i.e. at approx. 565 nm.
Regards, Wolfgang
Renate
Hello,
Unfortunately, Wolfgang’s measurements aren’t meaningful enough either. There’s no indication on either axis of the graphs as to what the numbers mean. I can guess what the X-axis represents, but the Y-axis is impossible to deduce. Judging by the numbers, however, it’s a linear scale, and that isn’t meaningful for chemical photography. Exposure is always exponential, not linear. Accordingly, the scale of the spectrograms would need to have a logarithmic division. Then you can see that the LEDs in the crystal-clear housing emit small amounts of secondary colours alongside the main colour. In other words, a red LED emits enough green light to cause a fog on multigrade paper, whilst a yellow LED produces a violet fog on colour paper. The only solution is to use additional filters. The LEDs with coloured housings are better, but they do not provide enough light of the primary colour.
Best regards
Renate
Wolfgg
Hello Renate,
I suppose you’re used to more sophisticated technology with lots of display options. Here, things need to be a bit more straightforward, otherwise newcomers will be put off straight away. That’s why everything is kept as simple as necessary.
The graphs correspond to Rolf’s histograms; after all, my measurements are intended to provide a comparison with his. The X-axis can therefore only be the wavelength of the light in nanometres. Although the Y-axis officially shows lux, that is irrelevant, as what matters here is solely the ratio of how the energy is distributed across the spectral range. And as for lin/log: linear Y-scales have become the norm not only in LED manufacturers’ data sheets; across the entire lighting industry, I see almost exclusively linear Y-scales. This makes sense, as we’re not dealing with many orders of magnitude here—which would require a logarithmic scale—but at most two. Although the Gretag i1 has a measurement range of 0.2 to 300 cd/m², it is unclear what resolution it operates at internally, particularly what is achieved with low-intensity Duka light sources. If it is only 8-bit, as I suspect, then a logarithmic Y-axis display provides no additional information. It is clear, of course, that the i1 cannot compete with a spectrophotometer at a university institute costing over €100,000 with a dynamic range spanning many orders of magnitude. But in my practical experience, it is accurate enough to measure Duka light sources.
The important thing is simply to set the Duka lamp as dim as possible in the laboratory. And the coin test should be carried out at least once, just to be on the safe side. In this way, even the yellow LED on the RA4 does not produce a fog for me. And something that is rarely mentioned: you can link the Duka lamp to the enlarger lamp, i.e. when the enlarger lamp is on, the Duka lamp is automatically switched off. This is very advantageous for long exposure times and when measuring the negative.
Nevertheless, the issue shouldn’t be overplayed. Apart from the Duka light, there are other ‘nuisances’ that end up on the paper – just think of the stray light from the enlarger and the walls. Or have you all dutifully covered the walls around the enlarger with black velvet? And do you only stand in front of the enlarger cabinet wearing black clothes? Perhaps someone might even try blacking out their face, like in the military, to see if it improves the contrast :).
Regards, Wolfgang
?
Renate
Hello,
We’re taught at school that a number without a context is meaningless. That’s basic knowledge and has nothing to do with the effort involved. Incidentally, I don’t use an expensive spectrometer either, but a very simple device. It’s a handheld spectrometer that I bought from Brenner for €10.00. I use my eye as the sensor. The logarithmic function is naturally built into this. Incidentally, I prefer to have my darkroom as bright as possible, and if you use the correct wavelengths of light, this works perfectly well.
The logarithmic approach is not a luxury, but a necessity, as exposure occurs in exponential steps. Only with the logarithmic scale can one tell whether the value is a true zero or merely close to zero. In digital photography, the logarithmic scale makes no sense, as the sensors operate strictly linearly, and the spectrometer mentioned above appears to have been specifically built and programmed for this type of photography.
Best regards
Renate
