Reversal RA-4 experiment thread.

[hrst]

Reversal RA-4 processing is discussed a little in many threads, but there's quite little definite experimental information available.

I've done my own little research for two years, every now and then when I have something else to do and want to escape from it . I've spent a few nights experimenting. I've thought of making some kind of Reversal RA-4 manual after I've enough results, but as I'm not sure if that day will ever come, I'll start now by giving all these little results I have now, and ask you also to publish your results!

Reversal RA-4 is much fun! I genuinely recommend experimenting a little. You may be surprised how easy it is to get nice positive images, though not flawless.

So, here I go.

Reversal RA-4 processing involves a first developer, basically a BW developer, then wash, light reversal (with room lighting), normal RA-4 color dev and blix and wash.

This will produce usable positive prints from slides, having some inevitable problems, because the RA-4 papers are not designed for this kind of use but for printing negatives, having lower contrast index.

What we want, is a tone scale of about 3.0...3.3 D, or 10 to 11 stops. This is about the tone scale of slides, and has to include the toe and shoulder of the paper to get proper white and black. Some compression of toe and shoulder is inevitable and present in Ilfochrome printing also. So, in practice, we might want to sacrifice some shadow&highlight detail to get more contrast in these areas, so, maybe a tone scale of 2.7D or 9 stops would be quite good for a paper.

I've found that the tone scale of RA-4 paper when reversal processed is about 10-11 stops! It is definitely not too "high in contrast"; but the real problem is the curve shape.



I'll go later to these step wedge tests, but this first example is to show the tone scale problem.

So, I've found that the characteristic problems in reversal RA-4 processing are as follows:
1) Too high contrast in shadows
2) Too low contrast in highlights.
3) Cyan highlights and upper tone scale -- not particularly red-cyan crossover, because shadows are not red and cyan highlights cannot be fixed by filtration.
4) Green-magenta crossover, giving green shadows and magenta highlights.
5) Uneven extra texture in even density.

All of them give a unique look, and while they maybe cannot be solved completely, there may be and there are some tricks how they can be coped with.

I have tried the following:

Paper
- Kodak Supra Endura and Fuji Crystal Archive

First developer modifications:
- Trying different BW paper developers as first dev
- Trying different dilutions
- Adding hydroquinone to make it more active
- Adding sodium thiosulphate (hypo)

Reversal exposure
- Amount of exposure, how much is really needed.

Color developer modifications:
- Adding sulfite (Photo Engineer's tip) to reduce contrast
- Adjusting pH to change color balance

And now, to the results.

The first thing is that many of the problems are not easy to deal with; there are random variations. For example, when I look my tests, the uneven texture problem randomly appears and disappears. Sometimes it's more evident. I've found that adding hydroquinone in FD clearly decreases it but still, I have some examples without added HQ having no texture at all. Similarly, I used more than 10 hours just to try to get rid of cyan highlights by means of filtration and CD pH adjustments, just to find that it isn't crossover problem but the layer just don't develop completely. It's funny, because it's the topmost layer. Then, later, I discovered that well, suddenly I get proper whites just with enough exposure, using quite dilute developer compared to earlier experiments.

Currently I'm confused about a few things and have many directions to go to. But first, some results. I'll post more specific technical information soon.


(Sensia 100.) Some of the first experiments. Print number 25 after many modifications.
Quite good, but shows the cyan forehead problem.
First dev: Agfa Neutol 1+3 (normal dilution is 1+7) with 20% of a medium-sized hypo crystal in 500 ml (didn't have scales at that time), 3min30sec
CD: Normal
Filtration: C17 M0 Y140 (on Fujimoto, normal neg filtration about C0 M50 Y100)


(Sensia 100) Newer experimentation with much less work.
First dev: Dektol 1+2,5 (about normal dilution), no modifications, 2min.
CD: Normal
Filtration: Might have been about C0M50Y180.

So, as you can see, using different first developers affect greatly to the filtration!


(Velvia 50.) Experimentation with added hydroquinone.
First dev: Neutol 1+3 with huge loads of HQ (I have the amount down somewhere but it is around 30-40 g/liter!!!)
CD: Normal
Filtration: C3 M0 Y146


(Sensia 100.)
Same first dev as in the previous example. Quite good. A little cyan problem and the whites could be a little more white, but if exposure is added, they go too low-contrast.



TO BE CONTINUED

PS. And, share your knowledge!!

 

[ricksplace]

Wow. Your prints are far better than I have ever seen. What are you using for exposure to white light? I found in my fooling around (far from experimentation) that the white light source, and the exposure made a difference. I have been meaning to try a re-exposure under the enlarger at the same exposure/fstop without a negative or any filtration. Have you tried this? I have been getting really crappy results with room lights.

 

[polyglot]

Have you tried filtering the reversal light?

 

[hrst]

I currently believe that the reversal step should be taken to completion, but I cannot say for sure if some benefits could be achieved if the reversal step would be controlled somehow. I don't see any reason now why it should. At least it wouldn't be easy. Just expose to bright room lighting at least for a minute or two, or take closer to your light source (fluorescent or tungsten), 30 seconds is probably sufficient at 20 cm (8") distance. But, it really needs quite a lot of light compared to normal exposure. 10-20 seconds of normal room light was not enough as I tested it.

 

[Photo Engineer]

Well, firstoff, there is quite a long discussion of this process at www.photo.net with myself and Bujor B among the contributors. There are examples of quite a few types of process presented including shots taken in-camera.

The first developer does indeed influence the results more than any other step. I found that Dektol 1:2 or 1:3 to be best. I even mixed a no-bromide version and experimented with bromide level. Sulfite level in the color developer is also a significant contributor to reduction of contrast. And, the curves do not match as you see. There is crossover if it is not done properly. You get cyanish highlights which means that the cyan layer in that region was not developed properly in the first developer. Bromide seems to control this to some extent.

PE

 

[hrst]

It makes sense that bromide could retard the development of the red-sensitive layer too much, but doesn't it usually affect more Dmin than Dmax? The key to getting proper whites is to get proper Dmax in every layer in the first BW developer; there must not be any halides left. I would understand the problem better if it was the undermost layer. :confused:

I have some quite mixed results about giving longer development time or using stronger developer. Usually they reduce the cyan problem. The best results with the cyan problem so far are the ones using huge HQ levels. It was about 40 g/l, which is near it's solubility limit . I'll find more test strips and my notes and post more results as soon as I find them. This one doesn't show any cyan, just the green-magenta crossover: http://www.students.tut.fi/~alhonena/siwa/siwachrome-HQ.jpg . It's Agfa Neutol 1+3 with 11 g/l HQ added if I remember correctly.

If I currently had to suggest something, I would say some compound of my results, like:
FD: Dektol 1+3 with added hydroquinone 10 g/l and hypo 0,3 g/l
CD: Added sodium sulfite 0,8 g/l.

But again, I might be completely lost .

 

[Photo Engineer]

In your first step wedge, not enough cyan silver is developed in the first developer, and so too much dye is formed in the color developer giving a cyanish colored toe. This is probably due to the bromide inhibiting the cyan layer developing to completion, which should take place evenly in all 3 layers to make a neutral. Some silver halide solvent in the developer might help along with lower bromide.

IDK. I got a number of very nice cross processed prints, but never got a good portrait. All of my best prints were landscapes. If you see those by others, you will see that the same is probably true. It may be an unsolvable problem.

PE

 

[hrst]

Yeah, I found that adding 0,6 g/l hypo changes the cyan problem to yellow problem . That's why I'm going to do further testing with 0,3 g/l hypo. And thiocyanate could also be tested, I bought it for this purpose.




I like these interesting color palettes. As you say, it works with landscapes quite well. It's probably because it's not always easy to get really green foliage in photographs like in paintings; trees are sometimes quite dark; but the green shadows of Reversal RA-4 print helps this! And again, magenta midtones give a nice warming effect, and finally, highlights in reflections and sky get a nice cyanish look. This is probably the best possible set of crossover problems for a landscape but certainly not for portrait, where you get magenta-brown skin and cyan forehead.

 

[hrst]

So, when I'm adding ridiculous amounts of hydroquinone, I'm probably just fighting against bromide present in developer. No wonder it helps. Maybe it would be time to scratch-mix to save some valuable HQ ...

 

[hrst]

Update.

Oh yeah!!!11 :munch:

I'll post some results, omitting many tests.

I highly recommend this process, it's really possible to get usable&interesting results. I've mainly solved problems 3 and 5 (cyan highlights and texture). Contrast also seems a bit better, but still very high. Green-magenta crossover in fact can be quite interesting in some pictures, and I think it cannot be solved completely.

Okay, what I did. Yes, I made my own first developer, a non-bromide "Dektol". I found out, however, that a little bromide is needed, and a little amount of hypo is good.

So, this is my current formula:

FD (add in this order, stirring at every step until dissolved):
Start with: Water 500 ml
Metol 3 g
Sodium sulphite 45 g
Hydroquinone 12 g
Sodium carbonate (monohydrate 80 g) (or 68,4 g anhydrous)
(I replaced 5,1 grams for bicarbonate because I haven't enough carbonate )
Potassium bromide 0.20 g (the original "dektol" type formula asks for 2 g!!)
Sodium thiosulphate pentahydrate 0.20 g
Water to 1 liter

Dilute 1:1 before using. Develop 1 min 30 sec at 23C.

CD
Normal RA-4 dev with added sodium sulfite 0,50 g/l (0,80 g/l could be even better).


Adding more hypo increases contrast and clips highlights and shifts highlights towards yellow, and vice versa. This current value seems to give nice whites, not yellowish nor bluish, while allowing to adjust yellow filtering to get quite neutral shadows at blue-yellow axis.


No hypo, tried to balance colors. (I could select between bluish highlights or yellowish shadows)


Hypo 0,20 g/l, tried to balance colors. (Now I could select between bluish shadows or yellowish highlights). And too high contrast.

These results led me to take the hypo concentration between 0,20 g/l and 0,00 g/l: 0,10 g/l.

Decreasing bromide may give even whiter whites, but may quickly lead to very bad DMAX. I added 1/10 of the "dektol" formula's bromide and expected to see some change, but DMAX went straight to black. Well, that was the objective .


Effect of no bromide at all - unexposed sheet - should be black. The first one without hypo and the yellow one with 0,20 g/l hypo (double compared to the final formula)

Stand development (2 min 15 sec total; 45 seconds agitation first and then 1 min 30 sec completely standing) helps manage the contrast a little bit, boosting the shadows a little and restraining highlights.

Dead Link Removed - normal
Dead Link Removed - stand

And then, some more pictures. These are all quickly made, without bracketing exposure or filtering!


AHH, whites!

I like the green crossover, this is originally quite a dull image, I used long ago expired Agfa slide that has gone yellowish.

Ahh! The flame is not clipped.


EXPERIMENT!! It's FUN, and eventually you start getting results!

 

[Tobster]

Wow indeed! I'm very impressed by your results, keep up the good work!

I did som reversal RA-4 experiments a couple of months ago. It worked, but the results were not that great.

After I have seen your pictures I will sure try again, as I have a fair amount of slides that i would like to print.

 

[hrst]

Some quick tips:

- It's all about the first developer. Other steps don't affect too much. Test with unexposed sheet and a much exposed sheet, you have to get proper black and white.

- If you encounter cyan / not white highlights, don't waste time trying to correct it with filtration. It won't work no matter how hard you try and you just mess the shadows and midtones. You have to lenghten development time (doesn't work very well), add more developing agents to developer (fighting against bromide and thus not so good solution, but works) or to scratch-mix first dev like I described before (it's really the easiest way, I highly recommend it!!)

- If you get the cyan problem, you will find that you can get better whites by adding much exposure. But it's not much of use if you mess your midtones, so, adding exposure is not a solution to the highlight problems. You have to chemically modify the first developer.

- Two-step wash without running water after FD is enough if you change the waters after every few sheets, but if you have running water, it's easy and safe to use it so I recommend running water.

- 30 second reverse exposure is enough but only if you take the paper very near to your lamp.

- Fix exposure first, then filtration, because overexposed sheet will appear magenta and underexposed green.

- Use much dodge&burn because of high contrast. When you burn, you may adjust magenta filtration lower for burning to avoid magenta crossover.

- If you add sulfite to color developer, you can still use the same developer in normal RA-4 as a low-contrast developer.

 

[j03]

Hi.

Thanks so much for all these details. But there's a few things I'm still fuzzy on.

I'd like to make some largeish pinhole photos using RA4 paper and the reversal process.

I gather I'll need to filter the daylight coming through the pinhole to warm it up.. I was planning on using a piece of developed c-41 film for the filter.
Does that make sense?

I was planning on tray processing with fotospeed mono.. will that make a difference? Can I use a monobath with the reversal process?

Found this overview of the RA-4 reversal process on another thread, might be a useful addition here:
http://www.gr82bart.com/APUG/Traditional printing of transparencies without Ilfochrome.pdf

Thanks,
j03

 

[Anon Ymous]

...I'd like to make some largeish pinhole photos using RA4 paper and the reversal process...

Have a look here then.

 

[Athiril]

Thanks for the thread,

"1) Too high contrast in shadows
2) Too low contrast in highlights."

Sounds useful to me...

Will have to get into this when I get an enlarger when I get my own place

 

[nickandre]

Hrst when you're making those step wedges is it a logarithmic exposure or linear? Is it 10 seconds, 20 seconds, 30 seconds, 40 seconds or 10, 20, 40, 80?

 

[hrst]

Logarithmic. 1 stop for every step.

 

[bill williams]

I hope that this experimenting will ultimately produce a technique that will allow me to have consistent results. But not yet. So, meanwhile I have been printing transparencies while staying within the RA4 process completely. I do this by enlarging the transparency onto RA4 paper through a mask and processing the print. I then contact print the enlargement once again onto RA4 paper and process it. I have had better results doing this than using the b&w developer step. Doesn't take very long either.

 

[nickandre]

I hope that this experimenting will ultimately produce a technique that will allow me to have consistent results. But not yet. So, meanwhile I have been printing transparencies while staying within the RA4 process completely. I do this by enlarging the transparency onto RA4 paper through a mask and processing the print. I then contact print the enlargement once again onto RA4 paper and process it. I have had better results doing this than using the b&w developer step. Doesn't take very long either.

So you use the same filtration for both steps? or do you change? and any problems with backprint?

 

[Ektagraphic]

WOW! There is a light at the end of the tunnel for slide shooters

 

[bill williams]

I use Mitsubishi paper from Photo Warehouse, it hasn't any back printing. Of course Kodak and Fuji do. This is an interneg procedure but with a paper negative, so I don't have to fool with C41, and therefore I have a quick process. I balance the filtration for the paper first with a standard negative. Then place some leader from a C41 negative and a diffuser in the light path. I flash the paper, then remove the diffuser and leader and put the slide in the carrier, and expose the paper. I've used full strength developer and developer diluted 1:1. Then after blixing, washing, drying I have the internegative. Contact print it, and process it. I'm still experimenting, but I've had some results that were just as promising as the process using the b&w developer. In fact, I've had some very good results. As I said though, I'm still experimenting. Hoping either I or someone else will make something work well.

 

[Ektagraphic]

Not to go off from the threads topica but how do you like the Mitsubishi paper for regular printing?

 

[bill williams]

I'm very satisfied with it. I may be very mistaken, but I think I read somewhere it is or was the most used paper for the 1 hour labs. That of course only says it was one of the most competitively priced papers. I have Kodak, Fuji, and Photo Warehouse paper on hand. When I want to print something important, I usually use Kodak. Sometimes Fuji. For everything else, such as snapshots, proofs, etc. I use the Photo Warehouse paper as it is easier on my budget. I might add, all three papers have slightly different looks. I think the Mitsubishi paper fits between the look of the Kodak and the Fuji, and therefore is good for a lot of subjects.

 

[Photo Engineer]

Mitsubishi = Konishiroku = Sakura.

PE

 

[bill williams]

After balancing the paper, you can dial in an extra 60 yellow and 40 magenta instead of placing a strip of C41 leader in the path. I've tried to find my notes, but its been several months since I last printed any color or experimented with the RA4 interneg. I committed to a couple of print exchanges and a negative exchange and so I've put all my time into those for now. After reading the recent post to this thread though, I want to go back and try the RA4 reversal again.