Off-the-shelf LED Enlarger Lightsource – part 3: contrast, exposure and coverage comparisons

In part 1 I considered ways to use a standard LED light bulb in place of the PH140 and ended up with the Utilitech Pro from Lowes.

Part 2 saw me cobble together a vaguely alarming cardboard-and-hot-glue contraption to hold the bulb in the right place.

Finally I made some comparison exposures to test for differences. I expected to see longer exposure times as the PH140, an overdriven 75W incandescent, pumps out 1150 lumens compared to the Utilitech’s 60-watt equivalent 800 lumens. I hoped for equally good coverage. I knew I’d see a change in contrast but honestly had no idea what to expect there.

Coverage

Though it certainly looks good by eye, the only way to know for sure is to make an exposure of an empty negative holder using each light source and compare them. I aimed for a mid-to-light gray with no contrast filters, which turned out to be about twice as long for the LED, all other settings the same. I set up for 6×6 medium format as my 80mm is not necessarily designed to cover larger frames.

This is how it appears on paper, with both tests scanned together for a direct comparison:

ph140-vs-led-coverage

Not bad on either one; just a tiny amount of visible variation, if anything the LED looks a little more even! Ignore the blotch in the top left of the LED patch; that’s a thumbprint from slightly sloppy handling.

Next, some post-scan curves alchemy, slamming the white and black points toward each other so that the exposed patches are represented by the full available range of tones. If there was such a thing as grade 50 paper, it might look like this:

ph140-vs-led-coverage-boosted

Blimey! So, the LED has a hot spot in the middle and falls off quite evenly toward the edges. Some of that might be down to the longer path taken by the light exposing the corners here; this was a 4×4 inch patch, the enlarger head nearly bottomed out, so I’d expect some falloff in the corners under the circumstances. The PH140′s light spread is weird and awful looking, with very sudden and severe drop off into the corners, though the center isn’t quite so hot as with the LED! LED for the win, folks.

Again, keep in mind that’s a highly exaggerated contrast level and that the first pair of patches are really what it looks like. In real prints even the PH140′s oddball coverage, which is just discernible in the unmolested scan, does not make any noticeable impact. So the LED? It’ll do just fine, thank you. I should probably be edge burning my prints anyway, which would only serve to cover up the slight loss of light in the corners.

Exposure and Contrast – Grade 2 filter

Next up, a comparison of the two light sources through a grade 2 Kodak Polycontrast filter. This will give me a feel for how exposure time and contrast selection would differ. Again, both prints were scanned together, so the tones on one compare directly to the other.

ph140-vs-led-exposure-contrast

The differences are fairly clear: the LED needs longer print times and has considerably lowered contrast. The longer time is about what I expected. Looks like highlights start coming in about 2/3 of a stop slower than with the PH140, somewhere between 8 and 12 seconds for the LED vs 6 seconds for the PH140. Since I’m about to start using Adorama VC RC and the new Ilford FB Classic papers, both of which are faster than the Ilford MGIV, I could actually use that slight increase in times.

The massive change in contrast through the #2 filter is rather unexpected. Now, since I’m moving toward split grade printing using blue and green filters, this may not matter too much, but it’s an interesting data point. I haven’t formally tested unfiltered exposure, but did make a couple of contact sheets that way and they looked to have normal contrast.

My next move is to subject this setup to some real-world abuse by making prints. I’ll find out in a hurry if there are any significant shortcomings.

Off-the-shelf LED Enlarger Lightsource – part 2: a custom lampholder

In part 1, I explored options for a cheap LED-based enlarger lightsource and settled on a 60 watt replacement, off-the-shelf bulb which looked promising.

However, there was no way I could use the existing lampholder for this. I needed to come up with something to hold it firmly in a position which would work for the purpose.

As is often the case with such prototype, cobbled together designs, I chose to work in the easiest possible media: cardboard, hot glue, acrylic paint and duct tape!

Be aware that this is a project involving line voltage electrical components and is built on the assumption that nobody will EVER be stupid enough to try running something like a 60 watt incandescent in it. If you aren’t comfortable or competent working with things which will connect to typical household 110 or 220 volt outlets, don’t even bother; it’ll hurt you, or worse, if you don’t respect it.

Anyway the build process is best described in pictures.

Position of the standard PH140 bulb. Clearly this is not too critical; my example is a bit off-center. My LED bulb has a wider frosted area facing down, so my main concern is proper fit and the correct height above the condenser.

Position of the standard PH140 bulb. Clearly this is not too critical; my example is a bit off-center. My LED bulb has a wider frosted area facing down, so my main concern is proper fit and the correct height above the condenser.

The box, a fairly thick-walled 6x6x7 inch affair, will provide ample space for the design I aim to implement here. It mounts atop the enlarger lamphouse using a hole I have already marked up and cut. There's sufficient clearance to allow the enlarger head to move the full length of the column.

The box, a fairly thick-walled 6x6x7 inch affair, will provide ample space for the design I aim to implement here. It mounts atop the enlarger lamphouse using a hole I have already marked up and cut. There’s sufficient clearance to allow the enlarger head to move the full length of the column.

I needed a platform on which the lamp socket could be mounted. Careful measurement of the original lamp holder, PH140 bulb, and the new bulb, suggested a level where this needed to be positioned. It's very thick cardboard cut from the box my enlarger arrived in, initially held in place with nails pushed through the outside of the box. The large opening will have a thinner piece of mat or illustration board, cut with vents and a hole for the lamp socket to mount up. This will allow a bit of fine positioning before securing in place.

I needed a platform on which the lamp socket could be mounted. Careful measurement of the original lamp holder, PH140 bulb, and the new bulb, suggested a level where this needed to be positioned. It’s very thick cardboard cut from the box my enlarger arrived in, initially held in place with nails pushed through the outside of the box. The large opening will have a thinner piece of mat or illustration board, cut with vents and a hole for the lamp socket to mount up. This will allow a bit of fine positioning before securing in place.

The mat board socket holder, cut with small vents to allow the little heat from the LED bulb's heatsink to escape upward.

The mat board socket holder, cut with small vents to allow the little heat from the LED bulb’s heatsink to escape upward.

Socket mount panel with the lamp socket in place.

Socket mount panel with the lamp socket in place.

Interior panels in place, and light socket mounted. The panels are held by a couple of nails in each side, and hot glued in place. The light socket panel is hot glued to its platform. The upper panel allows heat to pass upward to escape though the hole on top, but blocks any stray light finding its way up from the bulb.

Interior panels in place, and light socket mounted. The panels are held by a couple of nails in each side, and hot glued in place. The light socket panel is hot glued to its platform. The upper panel allows heat to pass upward to escape though the hole on top, but blocks any stray light finding its way up from the bulb.

To be sure of trapping any light leaks, I paint the interior flat black.

To be sure of trapping any light leaks, I paint the interior flat black.

Q: What do you get if the Borg assimilate a redneck? A: a cube covered in black duct tape. Shown here with the LED bulb screwed in and powered up.

Q: What do you get if the Borg assimilate a redneck? A: a cube covered in black duct tape. Shown here with the LED bulb screwed in and powered up.

With the new lamp holder in place, the LED bulb location is close to the original bulb. Height is correct, and it's probably better centered than the PH140 ever was!

With the new lamp holder in place, the LED bulb location is close to the original bulb. Height is correct, and it’s probably better centered than the PH140 ever was!

Mounted up on the enlarger.

Mounted up on the enlarger.

So there you have it. In the final part of this series, I’ll test the thing to make sure it works at all.

Toying Around

I don’t often go for the toy camera aesthetic and I have a beef with those who portray film as being random, fuzzy and unpredictable because they think film photography and toy camera photography are one and the same.

But in the right hands, a toy camera can produce beauty. I’m not those hands, but sometimes it can be fun to go out into the world and shoot with something so basic that almost the only option is to decide on where to point the thing, and even that can be a challenge. A liberating act, a cry of freedom from the tyranny of perfect exposure and razor sharpness.

And once in a while, when the stars converge and the moon is in the seventh house and Jupiter aligns with Mars, some photons in the form of an image which is both good as a composition and technically pleasing will dance upon those silver halides, without being totaled by a random light leak. Magic will happen. I can almost understand why toy cameras are so compelling to some, because when they hit, they really hit, and they’re the absolute antithesis of the pixel-peeping, moar-megapixels, give-me-sharpness-or-give-me-death attitudes so prevalent in photography (particularly of the digital sort) today.

Anyway I took out my old Ansco Goodwin #2 box camera for a spin one morning, to see if I could make something for the FADU “disposable camera” shoot out and APUG’s Jan/Feb challenge. We’ll neatly sidestep the fact that of the 8 frames I shot, 3 were essentially unexposed, 3 more were so badly composed that they might as well have been blank, and only 2 had promise, with one in particular standing out. Such is the way of it when you make photos with toys. I’m grateful I got one really nice one out of the deal, and I had a bit of fun in the process which is never a bad thing.

This was it:

Roanoke RiverThe film was developed in Ansco 130 (yes, my print developer!) at 1 + 43 dilution (compared to normal 1 + 1 for prints), semi stand for 30 minutes with a single inversion at the 10 and 20 minute intervals; this is a topic for a future post but is proving to be a method worth pursuing. This is a scan from the contact sheet, showing the full 6×9 frame with overscan. I plan on making an enlarged print from this negative, cropped in a little and with the horizon leveled.

Off-the-shelf LED Enlarger Lightsource – part 1: investigating possibilities

Never content with the maxim “if it ain’t broke, don’t fix it” I’ve been toying with the idea of an LED-based cold light source for my Beseler 23CII condenser enlarger. This is by no means a new concept; there are commercially available white light and variable contrast LED enlarger lampheads on the market already and quite a few examples of home-brewed versions of the same exist.

The dream system for me, as a black and white printer, would be something with royal blue, green and red LED sources driven by a controller/timer system to make a variable/split contrast + safe light source. But that’s a long-term project involving high power LEDs, driver modules and a microcontroller unit. Right now, I’m content to work with my contrast filters, or hard/soft filters for split grade printing, so I only need a suitable white light source.

Why even bother though? The PH140 bulb isn’t that hard to find, yet, I could easily stockpile a couple of decades supply of the little buggers. Oh, but there’s a great reason to ditch the PH140: negative pop. Most of the 75W the bulb slurps down comes out as heat and some of that IR energy gets focused down through the condenser lenses and directed into the negative stage along with the visible light. The negative heats up, expands and shifts, causing the focus point to change. If I quickly focus a cold negative then keep watching through the grain magnifier, it soon drifts back out of focus. This is a pain in the ass.

Yes, I could get a glass negative carrier, or a heat absorbing insert, or a cold light head, but all have drawbacks of their own, or are expensive and hard to find, or don’t fully solve the problem, or all of the above.

Luckily, the push to energy efficient lighting coupled with the shortcomings of CFL bulbs has forced manufacturers to develop LED based bulbs and get the price point down to an affordable level. The 60W and below LED equivalents are pretty good these days, certainly better in a good many ways than the CFLs I have lighting our apartment now.

But are any of those suitable for my plan? There are a few requirements which need to be met before I can even start:

  • Sufficient light output. For sure, I can cope with less light than the PH140 cuts loose. I already have to use neutral density filters often enough, so I could easily deal with a bit less light output. The 60W equivalent LED bulbs put out about 800-900 lumens, compared to the 1150LM rating of the PH140. That’s more than good enough.
  • Light makeup. Graded paper is blue sensitive, and variable contrast can “see” blue and green wavelengths. As it turns out, typical warm or soft white light LED bulbs have a wide-ish blue peak and green output which correspond nicely with the wavelengths the paper can “see”. Combined with suitable blue and green filters I should be able to get a healthy range of contrast using split grade printing on VC papers. The LED bulbs do have a strong peak in the blue which isn’t found with an incandescent, which might impact contrast, but I’m going to take a “try it and see” approach here. Results beat theory, after all.
  • Even, diffuse light without any “clutter”. A condenser enlarger needs a very evenly lit and clean source. Even a normal frosted household bulb isn’t frosted enough to meet that need. Some LED bulbs just won’t cut it.
  • Must turn on and off without delay. It has to behave exactly like an incandescent, if not better, otherwise timing exposures could be troublesome.
  • Will fit into the lamphouse far enough to work. The 23CII condenser head has a fairly narrow lamphouse. A standard 60W bulb will fit, so should the 60W equivalent LEDs. None of them are outlandishly wide.
  • Reasonably priced. If it costs $50 to buy one of these, I’ll pass and put the money toward my super-duper VC LED project instead. Fortunately, there are good candidates in the $20 and under range.

For the most part, 60W replacement LED bulbs meet these needs. The even, diffuse light requirement is potentially troublesome though. Some LED bulbs are triumphs of form over function and while they might illuminate a room perfectly well and even look good doing so, they’re never, ever going to work in an enlarger. Some promising candidates proved unworthy on closer inspection, like the Cree bulbs which have a dark spot on the very tip, right where I need the most even illumination; and the newer plain white diffuser Philips bulbs which reportedly have a noticeable on/off delay which would screw up timing.

In the end, one bulb stood out after some online research, and even better, it was the cheapest one I found, and available at my local store: the Utilitech Pro 60W at Lowes. I went ahead and grabbed one. I also picked up a little 7.5W nightlight bulb, to use for pre-flashing paper, a subject for another post later on.

The contenders: 75W PH-140, Utilitech Pro 60W equivalent LED, and the 7.5W incandescent nightlight for flashing.

The contenders: 75W PH-140, Utilitech Pro 60W equivalent LED, and the 7.5W incandescent nightlight for flashing.

When I powered it up, it was immediately apparent that I was on the right course. The bulb puts out very even light throughout the angle of coverage, and especially on the top where it matters most to me. It’s bright, too, almost as bright as the PH140. It turns on and off without delay compared to the incandescent. As heat goes, the plastic diffuser stays cool and little to no heat is directed that way. In a base-up configuration, there won’t be any noticeable level of heat going downward. Bye-bye, ghastly negative pop!

Unlit comparison. The PH140 is not as wide, both are quite evenly frosted-looking.

Unlit comparison. The PH140 is not as wide, both are quite evenly frosted-looking.

Lit comparison. Very even light from both. PH140 marginally brighter.

Lit comparison. Very even light from both. PH140 marginally brighter.

I’ll need to fabricate some sort of holder, as it won’t screw into the Beseler lamp socket (the heat sink clashes with part of the assembly) but this is likely a blessing in disguise, as the heat buildup in that space would be potentially bad for the LED bulb. Also, it would put the bulb very close to the condenser lens.

I won't be using the enlarger's standard lamp holder. Those 3 nubbinses clash with the LED bulb's heat sink and prevent it from screwing in. Even if I could, the bulb surface would be too close to the condenser.

I won’t be using the enlarger’s standard lamp holder. Those 3 nubbinses clash with the LED bulb’s heat sink and prevent it from screwing in. Even if I could, the bulb surface would be too close to the condenser and the poor thing might well roast itself alive.

To test how it might work in a real enlarging setup, I configured the enlarger for 6×9 (the absolute worst case, compounded by my 80mm lens being only marginally suitable for that format) and compared the image of my empty negative holder projected onto the baseboard for the PH140 bulb in standard configuration against the LED bulb held by hand in the enlarger.

Try *that* with an incandescent! Holding the LED bulb in place for testing.

Try *that* with an incandescent! Holding the LED bulb in place for testing.

Light spread: 6x9 holder, 80mm Componon-S lens, PH140 enlarger bulb.

Light spread: 6×9 holder, 80mm Componon-S lens, PH140 enlarger bulb.

Light spread: 6x9 holder, 80mm Componon-S lens, Utilitech Pro LED bulb handheld.

Light spread: 6×9 holder, 80mm Componon-S lens, Utilitech Pro LED bulb handheld.

The light spread looks quite similar in both cases, which is a good sign. I feel like I can proceed with fabricating a holder for the bulb, which I cover in part 2 of this trilogy of posts, so that I can run some real life tests.

Some Night Time Shooting with Ilford FP4+

A couple of weeks ago I went out on a night-time photo walk in downtown Roanoke with the Exposure.Roanoke group. It was my first time shooting FP4+ at night and with it being a traditional grain film, reciprocity failure is to be expected.

I didn’t want to bet an entire photography outing on guesswork, so did a lot of hunting around for reciprocity charts. I don’t entirely trust the official Ilford ones, to be honest. It seems a mite suspicious that they show identical charts for all their films, even the modern “core shell technology” Delta line which should have much better reciprocity characteristics.

I had a great set of reciprocity data for HP5+, found in an old Photo Techniques article by Howard Bond, but it did not include any figures for FP4+. Eventually, after much searching, I found a chart online with data for several films and which closely matched the HP5+ figures which I knew to be reliable.

reciprocity-chart

I must confess I can’t remember where I found this, and I didn’t note it down. It was on one of the forums, possibly the Rangefinder Forum or Large Format Forums. Since I’m mostly sticking to traditional grain type Ilford films (Pan F+, FP4+, HP5+) I printed those columns, sized to tape onto the back of my Luna Pro F meter. I exposed at EI100 using the 30 degree reflective averaging metering method. Development was in D76 1:1.5, using the official box speed time for this film in D76 1:1 and inverting once every 30 seconds.

The gallery below shows quick (no spotting done) film scans from the roll.

I’ll be interested to see how they print, but the scans certainly look promising.