DIY 5x4 enlarger light source

[AlanC]

I am looking for advice about changing the light source on a 5x4 enlarger that I made 30 years ago. I fitted it with a fluorescent 38watt 2D light, specifically because this would give off no heat and would allow me to use a glassless negative carrier without risk of the negaaive popping.
This light source has three problems which I have put up with over the years, but I am now looking to change the light source, and will appreciate any advice to this end.
The problems:

1 When switched on the light takes about 40 seconds to reach full strength. I can just about put up with this but it prevents me using a timer, and in effect, means the light must be left on permanently throughout the printing session.

2 With my 6x9 Magnifax enlarger, with an incandescent bulb, if you make prints from the same negative across the range of Muliicontrast filter grades -and keep the same time throughout - contrast increases as you you go from 0 to 5, but a certain mid-tone remains constant; i.e. it is the shadows and highlights that change. But with my DIY 5x4 enlarger with fluorescent light, this doesn't happen. A certain highlight remains the same (if print time is kept the same) and everything from upper mid-tones right down to shadows change. I can just about put up with this but would really prefer not to have to switch from one system to the other.

3 The big problem is the way contrast changes across the range of filter grades. With the Magnifax, and other enlargers, the change in contrast is about the same when going from 00 to 0, as when going from 0 to 1/2, or 1/2 to 1, etc. right up to the higher grades. So you can look at a print , decide it need a bit more contrast, and move to the next half grade up. But I can't do this with my diy 5x4 enlarger. From 00 to 0 there is hardly any increase in contrast. From 0 to 1/2 there is a bit more, but not much. From 1/2 to 1 there is a bit more, and so on up to grade 2, after which contrast change gradually gets less. In other words there is a big jump (more than a normal 1/2 grade) between 11/2 and 2, and from 2 to 21/2, and gradually less of a jump as you go to the higher or lower contrast grades. As will be imagined this presents real problems when printing.

Will changing the colour of the light improve matters? I notice 2D fluorescent lights are now available in "extra warm white", "standard white" and "cool white". ( I have no idea which mine is )
Or will it be a better option to change the light source completely? To LEDs perhaps?

Alan

 

[darkroommike]

The classic florescent light sources including the Omega "flying saucer" have both problems. I played with one at work and quickly concluded that those issues were why the lamphouse was replaced at work with a standard condenser head. Slow starts and bad color spectrum, not so much and issue with the old, slow graded papers but real bad with fast VC papers. Aristo solved the slow start issue with a "pre-heater" and a custom tube fixed the later issue. Both fixes may be beyond the ability of the DIY. I also wonder about home brew LED solutions, white LEDs are actually UV LED's with a white phosphor coating. You would need to know both the spectral characteristics of the LED and test for an afterglow. If you decide to try another lamp in your current head look for a full-spectrum bulb with a high CRI (color rendering index) and replace your ballast with a new electronic fast start ballast. I am not an electrician and won't go into details about replacing the starter and ballast and any rewiring that may be required.

 

[AlanC]

Mike, thanks for that. I have not heard of full spectrum high CRI bulbs, but have just done a google search and they are, it seems, available. I have an electrician friend who will no doubt advise me about how many I could put in my enlarger head, and how to wire them up. Sounds like they may be worth a try if they have the same "colour" as an incandescent bulb.

Alan

 

[bdial]

If I were going to build a light source from scratch I'd use arrays of LEDs. Probably using ether blue and green ones or addressable RGB LED's with a controller to vary the green and blue channels.

Some folks in here have done it and posted info on how they did it (using white ones, I think).

 

[tedr1]

The diffuser lighting system used in some enlargers gives the best of both worlds, tungsten lamps that turn on fast and have plenty of blue and green light for variable contrast papers, and, cool running because the lamp redirects the heat in the opposite direction to the light. This is done using what is called a dichroic lamp, typically they are small and very bright, and commonly used in overhead projectors among other things. My LPL enlarger uses an 82V 250W lamp this requires a transformer to give the correct voltage, however they are also made in other voltages here is a 120V device on ebay http://www.ebay.com/itm/Emerald-Dic...916858?hash=item3f699e39fa:g:u0kAAOSwmtJXbCt2. A small fan may be required to carry away the heat from the rear of the lamp outside of the lamphouse.

The diffusion is achieved by making a hollow white box, the lamp shines in on one side and the bottom is where light shines down onto the negative, this is known as a light mixing chamber and they can be made using white cardboard or purchased ready made.

With some ingenuity a dichroic tungsten diffusion lamp head could be made using DIY parts and skills.

 

[Paul Howell]

The classic florescent light sources including the Omega "flying saucer" have both problems. I played with one at work and quickly concluded that those issues were why the lamphouse was replaced at work with a standard condenser head.

I have the Omega cold head as well, but mine come with the electric shutter, a couple of years ago it broke and I did not bother to fix it, same issues with using VC paper, very slow when using MF and 35mm negatives. I agree that building a LED head would be the best solution or maybe you could source an Omega D2 4X5 condenser and convert it to your enlarger, don't know how common D2s are on your side of the world.

 

[nworth]

All fluorescent lamps, including the cold light tubes, have warm-up problems. If I were making a new light source, I'd look at LEDs. They have a full range spectrum (you can use all the filters) and come to full intensity instantly.

 

[mshchem]

There are LED lamps that are being introduced that look like fluorescent tubes. I've seen many that are designed to replace standard US 4 foot tubes. I'm not sure what different configuration types are available.
Secondly you can buy from EBay strips of white light leds that can be shaped (somewhat ) these run off of low voltage dc.
The use of a shutter with a full spectrum fluorescent tube might be a solution.
I would find an enlarger with a dichro head that someone wants rid of.
Best Regards Mike

 

[AlanC]

Many thanks for all the advice everyone. Much appreciated.
Regarding the dicroic lamp option, I could probably make this work as I have a 5x4 mixing box from an old De Vere enlarger. I will have to discuss the transformer and fan requirements with my electrician mate.
The LED option really seems the way to go. But I have no idea about such things, and would need some quite specific advice if I chose this option. What worries me is that I have read accounts that describe mixed results with LEDs.

Maybe my best option would be to forget DIY and look for a good used 5x4 enlarger.

Alan

 

[paul_c5x4]

Regarding the dicroic lamp option, I could probably make this work as I have a 5x4 mixing box from an old De Vere enlarger. I will have to discuss the transformer and fan requirements with my electrician mate.

An EFP (12V 100W) Dichro lamp would run off a standard low voltage lighting transformer and provide a fairly low cost solution to your problem. Another advantage of using a 12V supply is that you could utilise a cheap computer fan to provide cooling to the lamp housing.

 

[Kilgallb]

I placed an LED recessed fixture on top of the mixing chamber of my Beseler 4x5. It works well. I did have to slightly adjust my split grade printing because the lamp was equivalent to a 100W lamp, not the former 250W halogen. I also got the grade00 exposure to increase by mating the filter with a ND filter.

See the link to a recent thread.

(there was a url link here which no longer exists)

 

[AlanC]

Paul and Kilgallb, thanks for your comments. I had a word with my electrician friend today and he sugested I get a suitable halogen bulb, and he will then sort me out with a transformer and fan

Alan

 

[MartinP]

Options light include a diffuser-screen with one of the high-CRI fluorescents used for video lighting. That would imply timing via a shutter, of course.

On LargeFormatPhotography.info there was a search for new light-sources, with all sorts of incredibly complex LED solutions, then someone tried a very high-quality lightpad and instant success. The particular brand worked better than random others which were tried by other forum members. Details can be found here, and on threads linked from within that one. IIRC the start-up time was consistent and it could be run via a timer.

 

[AlanC]

Martin, thank you. I have followed your link and it makes interesting reading, though it is a pity the thread got sidetracked into a discussion/argument about enlarger lens focal length, and didn't go into the important issue of the lightpad's effect on contrast with multicontrast filters. I looked the price up anyway. The Artograph lightpad 920, recommended for a 5x7 enlarger, costs £369.77 off Amazon. I could buy an enlarger for that....!

Alan

 

[Jim Noel]

Contrast filters are designed to work with an incandescent light source. Any change from that will alter their effect. I know of no LED which has the same color spectrum as Incandescent. After all your troubles, I believe you wil be happiest if you go back to the original light source, you should at least give it a try. The heat s realy no problem unless your exposure times are unusually long. I keep mine in the 20-30 second range and have no problem with heat when I use this head.

 

[AlanC]

Jim, there would be no big difficulty in trying what you suggest. I already have spare incandescent bulbs for my Meopta Magnifax enlarger. So all I need is a bulb holder and a bit of wire...And as the lamphouse is made of wood I could easily ventilate it, to cool it.
Thanks Jim, will give this serious consideration. My one reservation is that I may get popping negatives. I could make a glass carrier to solve this problem, but this would involve fairly major alteration to the head of the enlarger. A bit of a mess about, but not impossible to do.

Alan

 

[M Carter]

Contrast filters are designed to work with an incandescent light source. Any change from that will alter their effect. I know of no LED which has the same color spectrum as Incandescent. After all your troubles, I believe you wil be happiest if you go back to the original light source, you should at least give it a try. The heat s realy no problem unless your exposure times are unusually long. I keep mine in the 20-30 second range and have no problem with heat when I use this head.

While this is true, there have been a lot of advances in the CRI of LED lights. The brand I use for corporate video shoots needs a 1/8 green filter to really make it perfect - it's just a hair pink, and I usually don't even gel it unless the subject has white hair (it's usually a rim or BG light). Sucker really has some punch though.

So one could potentially gel an LED source to tungsten, though there could be spikes in the color rendering that might affect filtering, it would be easy enough to test.

They way I dial in my daylight sources (I have biax flo's, LED, HID, and HMI) is to shoot a gray card with a given light using a DSLR. I open the raw file, and white balance the card. The raw sliders for temp (the blue-orange axis) and tint (the magenta-green axis) give me a good idea of where I'm at, and I'll gel the fixture, shoot again, and fine tune, and mark the gel combo on the fixture (usually talking very mild filtration, like 1/4 CTB and 1/8 minus green). One could do the same thing with a daylight source and CTO gel, to get it dialed down to 3200k and work any tint issues out. Or start with a warmer LED to use less gels.

I know you Arduino guys are all about mixing colored LEDs and soldering up stuff and making a circle of chicken bones on the floor... my knowledge ends at gels though!

 

[Kilgallb]

Contrast filters are designed to work with an incandescent light source. Any change from that will alter their effect. I know of no LED which has the same color spectrum as Incandescent. After all your troubles, I believe you wil be happiest if you go back to the original light source, you should at least give it a try. The heat s realy no problem unless your exposure times are unusually long. I keep mine in the 20-30 second range and have no problem with heat when I use this head.

I totally agree with you. I had to add a ND Filter to balance the exposure time. That said, many LED manufacturers are working on systems with less blue light and actually come quite close to matching tungsten filament lamps in spectrum.

 

[AlanC]

Maybe we are not quite there yet with LEDs . Perhaps I should wait until ND filters and chicken bone circles aren't needed to get them to work with multicontrast filters....

Alan

 

[Kilgallb]

No chicken bones here. LEDs simply have a different spectrum with more energy in the blues. His means the numbers on the filters do not match paper grades. if you do split grade printing you get the same results as if you had a tungsten lamp, times for the yellow versus magenta filters are just different.

 

[pcyco]

hallo
i put some powerleds into my durst 138 and it works fine for me. maybe there are some differences to a "real" lamps but for my eye they are minor. the print must look fine, then its ok.
i also have a kaiser enlarger with a "normal" bulb and sometimes i work crosswise without any troubles

regards

thomas

 

[Steve Smith]

My enlarger has the LED PCB from an LED stage light as its light source. Do a search here as I documented it in a couple of forum threads.


Steve.

 

[AlanC]

Hi Steve, thanks for that. I have found the thread you mentioned and am working my way through them.

Alan

 

[MartinP]

The Artograph lightpad 920, recommended for a 5x7 enlarger, costs £369.77 off Amazon. I could buy an enlarger for that....!

They are under a hundred GBP, roughly speaking. Here (NL) they are about 115 or 120 Euros, including case. The larger sizes don't increase much in price, so it might be worth getting one to cover 8x10" (in case of future need) then masking it down. There is no heat penalty involved.

Possibly you were accidentally looking at the super-duper, high-priced range of Lightpad-Pro lightboxes from the same manufacturer? They have a couple of pricepoints within their range of products it seems.

 

[Withnail]

They are under a hundred GBP, roughly speaking. Here (NL) they are about 115 or 120 Euros, including case. The larger sizes don't increase much in price, so it might be worth getting one to cover 8x10" (in case of future need) then masking it down. There is no heat penalty involved.

Possibly you were accidentally looking at the super-duper, high-priced range of Lightpad-Pro lightboxes from the same manufacturer? They have a couple of pricepoints within their range of products it seems.

Hi guys, sorry to barge in here, but I'm hoping I've found an answer to my problem in this thread....or more specifically what MartinP said.
I've been given a Durst Laborator 138 4x5 enlarger with Rodenstock 150mm lens that is basically complete and working except for a couple of things. The light source (Durst Lacoli) has a smashed bulb and there is no transformer/power supply. I was thinking of fitting LED strips into the Lacoli box and building a simple power source for them.
But now I've just seen the above comment about the Artograph Lightpads....
Would it be possible to engineer one of the lightpads into the housing where the Lacoli light normally sits and just use Ilford under the lens filters? Or do I need a different type of filter?
I'm only an amateur printer so it doesn't have to be perfect at all.

Thanks
Andy