Posted on

Imaging Paintings for Giclee Printing

I’ve heard enough from friends, colleagues, and others that I tend to go a bit overboard when it comes to my possibly overboard concern with quality when it comes to imaging and printing.

I spend a *lot* of time researching this, trying to find new ways to get better results, and to have more control over the process, especially since my move to Michigan, where high operating costs for businesses and a lack of local money have led to available services being less than stellar, if available at all. I have spent days, and days, every few weeks, just doing web searches, sending emails, making cold calls, and trying what few services I find here… and though I have managed decent results, they have come through much supervision, and the asking of many questions I would not have known to ask 10 or even 3 years ago.

Chances are, that if you do not live in a major city with a thriving art scene (if there is such a thing in this day and age), and if you are an artist trying to get prints made, you have found yourself frustrated with at least one aspect of the process.

Though printing for giclees is easy enough to find these days, in the most common of places, the imaging part is the absolute hardest to nail down, and the most important – because once you have sold the original painting, unless you have visitation rights in your seller’s agreement, you won’t get another chance.

There are a number of great digigraphic or reprographic places abroad. In Cincinnati, I had access to Queen City Reprographic, as well as the imaging person/photographer for the Cincy Art Museum, and the University of Cincinnati print lab. In Boston, I had access to Parrot Digigraphic in Billerca MA with their huge 72-inch scanner, and Ditto Editions in Salem MA with their 120MP Seitz camera setup (though they appear to have moved to New Hampshire) – but no longer living in the area, as amazing as these places were, there is the cost (and fear) involved in shipping a painting that has not yet been imaged many states away… and having to work by mail for proofing and other such services. A broken or lost painting can’t be imaged, so the fear of being forever screwed drives me to always look for local services if I can.

If you have $50,000 or more laying around for a Seitz camera setup, or $100,000 for a huge scanner (and recommended maintenance plan), you probably aren’t reading this anyway, and don’t need to. I’d think your time is worth enough that you’d rather outsource this work to a premium imaging company (as listed above) – So I am going to skip past those aspects of the process, and go directly into steps for the rest of us:

Step 1 – What Am I looking for?

This is pretty important. If you don’t know what you want, you won’t know whether you are getting it, and you’ll frustrate yourself and everyone you deal with, either in not knowing what to ask for, or finding that what you get is just not good enough.

— Actual size, in 600 DPI (preferable), 400DPI(still good), or 300DPI(acceptable). Sure, the human eye sees at 300 DPI, but consider that you might want to make prints larger than the originals, or highlight specific parts. Also consider that printers are getting better and better every day, as are papers, as are inks – and that if the result is a higher resolution, one might appreciate being able to take a magnifying glass to the print to see how great it is. Or, you may, like me: I am the sort of person who can read the microprint on a 20 dollar bill… I may not be able to see a tree at 20 years, but that is a product of being so detail-oriented. I put in details I can see that most others need a magnifier for, I even go so far as to put in details that *I* need a magnifier for, or details that I just *know* are there though I can’t see them. Most of all, the better your working file, the less fine details are lost to things like color adjustments and resampling through resizing. Don’t let people tell you that 150DPI is good because people are standing further away to view it – detail is lost, and they will not get the experience of looking at your work, but the experience of looking at a blurry version of your work. “Good enough”, is never, and will never be “Good enough”, else it need not be said at all.
— RAW or TIFF format. Jpegs don’t just lose data and detail every time they are opened and saved, the creation of a jpeg image alone is lossy. A group of eyelashes becomes a dark blotch of close colors that *look* perfect, until you campare.
— No adjustments. Every time a color adjustment is made, data is lost. A level adjustment brings you closer to the actual values, but causes colors to disappear. Every time an adjustment is made, close-to-white may become white, close-to-black may become black. Once they are gone, they are gone, and there is no processing you can do to get them back. It is always good to have an unprocessed image to start from, and always go back to that image if you need to make changes in brightness/color/levels.
— Quality. A 600DPI image makes no sense if it is 600DPI of blurry or 600DPI of blocky color.

Local Services

Calling or emailing in detail can be important here. I tried the universities first, hoping that they had for-pay services for non students, or that maybe I could put their students to work using the college’s equipment. It may work for you, but here, I couldn’t even get as far as finding out for certain what equipment they had on campus.

Then I turned to the local arts organizations, called a number of galleries, and asked what few local artists I could find what services they used. The answer was a resounding “uuuuuhhh?”.

When it comes to local artists, especially in mid-western cities, I’ve found that a lot of artists don’t even sell prints; Hell some don’t even actually produce artworks, or do anything more than sporting a goatee and adopting a highly-liberal point of view. Most functioning artists however are in the business of catering to galleries and coffee shops, who, since these places desire artists who can produce 20 or more artworks in less than a year, stick to artworks they can create in a single night… or several a night – which aren’t exactly the sorts of works that require printing. These sorts of works go to people who want an original, of certain colors or size, for $80 or less, to hang over their couch or in a closet…. or just to people who want to support budding artists… they tend not to do well as prints, especially considering that a person could just pick an original out of a bin.

Out of over 100 printing/imaging/reprographic places I contacted, only 10 said they were able to do this sort of work in house.

Of those, most were sort of scary – either for not having a grasp of the scale, or having a “sure, we’ll give anything a try” attitude about it. Not a one of them had a Seitz setup, or a large flatbed scanner (though one claimed to have a large enough flatbed). Some offered to use a 10 (or less) MP camera, some actually thought as far as to offer to take multiple photographs and piece them together (The latter, though ill-equipped I think might have actually known what they were doing, and *could* possibly have come up with passable results).

I went with the one who claimed to have a big scanner. The results were decent, save for some scan lines which I removed in photoshop, and the price was low (probably because of the scan lines). The color and brightness were the absolute best for adjustments – raw and untinkered with, with a broad range of values to be adjusted, just the way I like it… no data lost. When I walked into the back of their shop one day to see my painting being pressed against an 11×17 inch Epson scanner, I felt a bit betrayed, but continued to use them until that employee moved on (being mindful to remove my paintings from the canvas stretchers so they could scan without potential damage). After he left, the workers were less capable of getting me the results I wanted, and the new inconsistent pricing came straight from the “USS Make Shit Up”, so I started into the adventure that is….

Imaging At Home

Part I Scanning:

Scanning Smaller Works (under 11×17 inches):

I have a scanner that I bought through Sam’s club (I actually bought my membership specifically because the cost of the scanner+membership was $5 below buying the scanner elsewhere); The scanner is a Mustek A3 scanner (11×17 inches). They sell for about $199 including shipping most places you find them. I bought mine for $142+$7 S&H through Sam’s. The colors pretty much dead on, straight off the scan, and the detail is superb. It also plays nicely with photoshop and has the best import GUI I’ve had short of an Epson professional scanner ($thousands for a tinier scanner).

Until I broke it (I’ll talk about that later), I’ve always used this for any work smaller than 11×17; I have however found (in a moment of need) that Fedex/Kinkos has a flatbed 11×17, and will scan for under $2 a scan at any resolution you want. I tend to go for 600DPI for all my pieces, big or small. I know the human eye sees at 300 DPI… but I detail my paintings and other works so deeply, that printing larger than the original is sometimes desirable (plus: my tendency for quality overkill).

I think Kinkos $2 a scan is a pretty good deal. The scans come out quite well (but may vary at different places and with different employees). It is still always good to have your own scanner at home, in case your local Fedex is not staffed with adept people, or for scanning at 3AM.

Scanning Larger Works (above 11×17 inches)

As for scanning larger pieces with the A3 scanner, the biggest problem is the scanner’s “handy” beveled recess (the dip in the tray that holds papers in place). If your piece is bigger than 11×17 inches, it will sit away from the glass, and you’ll not get a good scan – it’ll be hella blurry, with colors shifting greatly from one edge to another, and utterly useless.

Is it on Canvas?

If you are so inclined, if your painting is on canvas, you can pull the staples from the back (or sides for cheaper ready-made canvas), and scan the piece in sections:

— Varnish can really help things, or really mess things up. If you have already varnished the piece, especially if it is a glossy varnish. Think of varnish as a lens. Not only does it increase the distance between the scanner glass and the actual artwork, but if your varnish has a brushed texture, you may get better scans in one direction than another, because such “polarizes” the piece. If I *have* to varnish before scanning, I brush the varnish in one constant direction, and scan in a way that the scanner arm is moving *with* the “grain” of the varnish. If I ever *have* to varnish, it is typically because I find that certain spots of paint are more flat or glossy than others, and I do so to even out the sheen painting-wide. A flat varnish is preferable, satin is a runner-up, and the thinnest coat possible.
— Keep in mind the most important thing when scanning in pieces is making sure all your scanned sections are square with the others. If they are off, when re-assembling in Photoshop, it is difficult getting that 0.02 degree tilt from one piece to another and making them all line up. If you *have* to rotate pieces in photoshop, don’t think rotating with the mouse will do it: use the numeric rotation (if you have gone to image:rotate, you’ll see a text box in the toolbar the top of your screen where you can enter numeric values). Be sure to zoom in and look it over from one corner to the next, to make sure it looks right (full view lies).
— Don’t rely on the photomerge tool, especially if your pieces are highly detailed. Photomerge is plenty good enough for assembling photographs for novelty’s sake, not for art.
— You may need to adjust levels from piece to piece to get them to match up well in color and lighting.
— Make sure each scanned section has at least an inch (or two) of overlap with the next so you can get rid of the edges, which tend to be shifted in color slightly due to the bevel.
— Turn off all lights in the room before you hit ‘scan’ (you can of course turn them on again in between scannings). Outside light can bleed in at the edges of your work, or through your work and make the pieces inconsistently colored/valued.
— A higher bit rate (8 bit color, 16 bit color, 32 bit color, etc), means more colors on the pallet, meaning smoother transitions between colors and values. Higher bit rate is good (though it greatly increases file size).
— Always save one assembled version that has no color/brightness/levels modificatons, and save it in TIFF (lossless format). You might need this if you decide the prints are too light or too dark.
— Always make adjustments from the original lossless and unmodified source. Every time you make color adjustments, data is lost, and your work drifts farther and farther from the original. Also, every time you open and save a Jpeg, it decays (loses resolution).

Is it on a Wood Panel?

If your painting is on a wood panel (or similar surface), this gets tougher (you’ll still want to read much of the above about varnish and whatnot). There is no amount of strength you can apply to put the board flat on the scanner, and you’ll break the scanner long before that. Even with an un-beveled scanner (like some Epson models), you need to be mindful that the board is probably not perfectly straight – press accordingly (but try not to break your scanner in the process).

I modified my 11×17 inch scanner so I could scan paintings on wood panels. I did so by

(0) **read this entire thing before you decide whether it is a good idea**
(1) carefully removing the glass. It isn’t easy. It is held on by some *very* sticky stuff. I used an old iron on the glass side to make the glue “melty” enough so I could pry the glass loose. I thought I was going to keep the glass, so I put cloth between the iron and the glass to keep from scratching it.
(2) carefully removing the calibration strip from the glass (it looks like a bar code, and lines up with the scanner arm’s starting position)
(3) built a wooden box to hold the glass nice and level.
(4) taped down the calibration strip (face down, tape on the blank side, calibration strip on the top of the glass)
(5) repositioned the box with the glass until the calibration strip was being read properly

it isn’t as much rocket science as it would seem, except for the last part (so read this in full before taking your scanner apart):
— The scanner arm has a bit of spring to it so it always meets the glass, so getting the glass “close enough” to touch the arm, is good enough… no microns of distance/precision to worry about.
— The calibration strip needs to line up pretty perfectly, but if you make it so the glass/box is just loose with the scanner body sitting below, you can move it around until you get a good scan – then secure the box in that position with tape/nails/screws.. whatever.
— The glass *has* to be the same thickness, optical quality (normal glass has a greenish tinge that can throw off the color calibration), and *tempered*. In most cases, the only way you’ll get a 1/8th inch sheet of glass tempered is for it to be chemically tempered. Why not use the sheet that came with it? Well, *that* glass is made to be secured by sticky stuff, and is typically just microscopically wider than the scan arm that needs to move against it. So, making a support for the glass that won’t get in the way of the arm is difficult to impossible.

.. That last part is what got me. Though I *have* found places that will make the glass, it is around $100 to $200 #realartistsdonthavemuchmoney, so I am making due with the original sheet… which teeters delicately on .01 millimeters of metal bracket edge, which the scan arm occasionally gets hung up on.

The FrankenScanner (above). It works… but that is all that can be said about it. The glass, which I secured with two-sided tape, after a year, is pulling away from the remaining plastic bit. So, now anything with any weight causes the scan arm to freeze. I’ll need a new scanner… but I’ll probably rebuild this one as originally planned (diagram below) once I get the chemically tempered glass I need.

Frankenscanner 2 (Bride of Frankenscanner),waiting on a sheet of tempered glass. The nice thing about this way of doing it, is that the glass and box can be as big as I want them to be – lending to a better support area for large paintings. I’ll probably make some brass bars on hinges up top for reference/alignment… something I can swing down to match up with lines I’ve drawn on the back of the piece, for better squaring (not included in diagram).

(yay!) Out-of-the-Box For Wood or Canvas: The Scanjet 4670
(boo!) Discontinued

For scanning large items, especially if on a wood panel, you might be better off finding an HP Scanjet 4670. Unfortunately they are discontinued, because marketing gurus and management are typically idiots, and don’t know *who* to market a good thing to in order to make it a good thing.

I found out about this 4670 scanner, which can be laid with its nice flush surface down on a large object, through my friend and colleague Brigid Ashwood. I just ordered one days ago, but from what she has told me, it does a pretty damned good job and is a godsend (for all of the reasons outlined above). You can typically only find them used – try Ebay. This week, I’ve seen three in box sell for $103 (plus $36 shipping).

I plan to buy another of these as a backup BTW… so please don’t buy all of them.


Long have I waited for digital cameras to be good enough for this. I waited for cameras to get up to 7 MP, then found out that was not good enough, then 10MP… still not good enough… then 14MP… Now Nikon and Sony have 24 MP cameras out – which are good enough for 18×12 inches at 300 DPI… which really, is still not good enough.

I realized this lately, when I got an incredible photographer, both in skill and in talent, to photograph my pieces for print. The images are amazingly crisp and perfect in color, but at 12 MP, they will never be so crisp that they can be printed at 36×24, especially with all of the detail I put into my works.

It takes 25MP to equal 35MM film, 100MP to equal medium format film, 500MP to equal large format film.

If you are waiting to be able to do this in one single shot, you might want to just give up on digital for now, and try to find yourself a good medium-format film camera, or a photographer who has one. Just be sure you also have the means to scan the negatives (or a photo lab who will scan them *for* you).

For the rest of us, taking multiple shots in order to make one image.

It *can* be done, no matter how few MegaPixels you have – just how many individual pieces/shots you need depends on how great the resolution is.  Another important part is realizing that moving the camera is bad, and why (outlined in the list of caveats below).

— Don’t let *just* MegaPixels dictate your decision though. Recently I found an amazing 14.5 MP camera with a great lens, and true SLR format for $149. The downside is that it saves all images as Jpegs, which are instantly lossy with bits of detail and nearly matching colors blurred together in chunks.
— Make sure your camera saves in either TIFF or RAW format. This is more important than MegaPixels. I switched back from the 14MP to the Canon 10MP because of the Jpeg thing.
— Don’t use a point and shoot. Don’t use a macro lens. *All* lenses round out at the sides (meaning a square will be “bowed” at the edges), and the best you’ll possibly get is a “true vision” or 50MM lens.
— Don’t use digital zoom. That just uses computer logic to try to translate an image more crisply from smaller pixels. A computer cannot know to make a row of eyelashes from a black blotch – all it does is to crisp up the edges. Make sure your zooming in or out is all optical.
— Use a Tripod! This should be a given, but I am saying it. Camera shudder is your enemy.
— Use the timer, or a remote, or a cable release. If you click the button by hand, your camera will move and shake, no matter how good your tripod is.
— No on-camera flash! Seriously… flash for any photo subject is best as a last resort, and terrible for capturing a painting. This isn’t a sporting event or a party, you don’t need to be mobile. You’ll get hot spots, color shifts, and maybe glare out your image altogether if you use a flash. Use stationary lighting. It isn’t like you need actual 200 watt or 500 watt lights anymore, there are low-wattage equivalents everywhere. Stationary lighting allows you to see your shot exactly as it will shoot.
— Lighting: You need a natural daylight spectrum. *but* you want it to be controlled and constant, since you’ll be trying to put multiple shots together, and you want them to be identical in lighting, distance, and angle. You can buy a 200-watt equivalent (40 watt actual) daylight spectrum bulb for $6 to $9 most anywhere online (try google shopping). You’ll need several of them.
— Lighting stands/reflectors are highly recommended. It’ll save you the frustration of gathering lamps from the household and propping them up on books and moved furniture. You can use extra photo tripods or tall lamp posts and utility light reflectors, or pie plates if you need to improvise.
— Put one light on each side of the camera, at 45 degree angles, and slightly higher than your painting. If the lights are in front of you, keep them far enough away from the camera that the light isn’t bleeding through from the sides of your shot. Also make sure there are no hot spots/glows/reflections on the portion of the painting you are shooting. Move the lights around until the *portion* of the painting you are shooting has just the right amount of light. Indirect lighting is absolute best for avoiding hot spots, so don’t point the lights *at* the painting. Try moving them higher, lower, more outward, more in – maybe even experiment with bouncing them off white surfaces (walls, ceilings, tacked up sheets) to soften them up and spread them out. Bouncing the light is of course best. If you have reflectors (umbrellas) made for this, even better.
— If you have a camera with light metering and related controls, and can set your white balance manually, do it. A *lot* of the fine-tuning can be done through just adjusting these levels.
— Set your camera’s ISO to its lowest setting for best image quality.
— Set your aperture to between f/5.6 and f/11 – this allows you the best sharpness, while still allowing some room for minor adjustments.
— Shoot each portion without moving the camera. If you move the camera, you will never have the lights and painting the same distance and angle from shot to shot, and you’ll lose all consistency. Instead, move the painting a foot or two to the left/right/up/down until you have all parts photographed.
— If you are good with building things – for this, I’d highly recommend making something where the painting can be mounted securely to a flat and straight board, and where that board can easily be shifted up/down/left/right. You can use drawer slides, or brackets, or just screws (screwing board A to board B, unscrewing it and re-screwing it for each move), or you can go more advances and use gears and cranks. I am using the low-tech boards and screws method (but on the lookout for old window cranks and worm gears so I can do the latter). Paintings on boards, or on canvas, when unframed, tend to be slightly bent or bowed.. so just hanging it in different places on the wall might not do.
— Make sure (as with scanning) your photos have a bit of overlap – not just so you can crop off the edges and keep the center (which will be slightly rounded), but so you can more easily line up one piece with another.
— do not pack up your lights and painting and camera. Leave it where it is for now. You may need to come back to it later.
— Again, do NOT use photomerge.
— Put each piece in its own layer, move them into place, crop away some of the edges of each (but leave yourself some overlap), make minor color adjustments to each piece if needed. Once all pieces are properly aligned, use your eraser on the edges of each overlying piece to ensure an extra smooth transition. Once you are certain everything is going to meet up perfectly, merge them all together. If there are any parts you aren’t altogether happy with, go back to your camera setup and re-photograph that piece or section.

I’ve found that at 12 megapixels, I can get a 36×24 inch painting with 15 photos (5 photos wide, 3 photos tall). If yours needs to be printed smaller, you can of course get away with less photos.

If however your painting is less that 11×17 inches – this is silly. Take it to Kinkos and scan it, or buy an A3 scanner.

Some Last Little Tips:

— Calibrate your monitor. There are free apps for this such as Calibrize. I have mine calibrated for perfect colors and perfect web brightness/contrast, and then have my *mind* calibrated for perfect print. I *know* how much lighter the images need to look in order to get them to print as they are seen on my screen, for each and every printer I use, and I know this through trial and error.
— Don’t use “Image:Adjust:Brightness/Contrast”, use “Image:Adjust:Levels”. Move the triangle on the left over a few points so it is nearest to the first spike (black values), slide the middle slider right a bit to make the mid tones richer, do not slide the white point (far right), ever if you can avoid it unless you want to forever wash out your near-whites. Don’t take this as law though: Experiment, and always do what is best for that individual piece. Most importantly, do not save over your original file, or you’ll be sorry when you need to adjust these levels for a new printer or printing medium.


1) Scan small stuff at Kinkos, or buy a Mustek A3 scanner.

2) For large works, if you can get one, grab the scanJet 4670 from Ebay or elsewhere – or go the photography route and (recommended) build a rig for photographing paintings in pieces. Also, don’t forget that you can (carefully) remove canvas paintings from their stretchers if you must scan on a standard scanner.

3) If you can, pay the man (time is money) – If you are not hell-bent on full control, and have the money, and can find a good local service: Use them. This stuff is a huge headache, and you can use all the hours you’d spend on research, practice, fine tuning, and execution on making works that would be worth whatever you are trying to get out of paying. Reasonable imaging for a painting is $300 for something in the 36×24 to 48×24 ballpark.

I’d recommend against proofing ($200 or more additional) – do the color adjustments at home, make crops of your image (saves money) with different color adjustments, and get those printed. I have several versions of the most crucial 8×10 section of a painting printed in my first try. Be sure to label each try, so you know which was the one that looked bright on your screen, which was the one that looked dark, etc… Hold these printed bits up to the original and compare.

Once you have this down, use the same printing company for as long as you possibly can… Once your prints are perfect, buy all 20,30, or 50 limited editions of that print that you plan to offer – and store them in a flat file or a large flat box. Why? Because a printing company’s printing profiles can change, or maybe the day you order print #27 of 50, some nose-picker replaces the worker who used to print your works perfectly. Maybe they change inks, or canvas type… either way, that Limited run you thought you could run one at a time (the benefit of digital offset world), is screwed… and by the time you make adjustments to account for the variation, you are sitting on three or four prints you can’t (in good conscience) sell.

I’d go into recommended printing companies,things about color profiles and other such things here, but that would make my summary/conclusion long. That will have to wait for another post. Stay tuned.

Posted on 11 Comments

Custom Framing on an Artist’s Budget

One of the more painful and costly steps along the way to an art show or convention for us, and other artists we know, is most always framing.

It seems that artists and artisans anymore, can be some of the most heavily exploited tradesmen when it comes to tools and supplies …and frames are certainly no exception.

Art supplies tend to be priced for Warhol-famed artists, design agencies with multi-million dollar clients, or students who simply have to have these things, and will have to take out loans and grants accordingly; Frames are often priced for artists of Ryden renown, or remarkably wealthy art buyers – factor in huge percentages for art sales through galeries, and it is no wonder that the gallery owner lifestyle of nice cars and sipping champagne contrast so beautifully with those of the artists supported: which I suppose for most of us… well, I don’t know how you get by, but lets say “robbing liquor stores for brush money” as a rather broad example.

And since for any show or convention it is good – if not mandatory, to have hang-able art. Many of us artists need to get creative if we wish to account for a 40% cut on sale price, and frames, without demanding a “Just who does he think he is?” sort of price.

One route, of course, is to gallery-wrap, or museum wrap paintings and giclees in a way that makes for a nice display without the need of a frame. Add a touch of color or a reflected image to the overlap, and such things do really tend to look very nice… though modern by default. Gallery wrapping and museum wrapping is available as an option for ready canvas through most any art supply provider, and having giclees stretched these ways is an option with most decent giclee printer. Doing this one one’s own is moderately difficult for gallery wrap, and strenuous for museum wrap – but by no means impossible.

If however your works are already stretched and mounted, with staples down the sides, or if they are painted/drawn/engraved on boards or paper and absolutely need to be framed – there are still a few creative, inexpensive, and simple ways to go.

One of which – if your works will fit standard frame sizes, is to go on “50% off!” weeks (we like to call them “regular price weeks”) to Hobby Lobby or Michael’s and buy their pre-made frames. Another solution of course is to go to places such as Target and Walmart and browse, looking for discounted frames that can be redecorated and customized to look nice… perhaps even to become works of art in their own.

One problem I have found with many of the ready-made frames (and even many custom frames), something I particularly dislike, is that I find many of these frames are either plastic, or low-grade wood, either made to look like better wood by way of plastic/acrylic veneers or overlay, ornate clay designs painted or coated over, or by other means. The most standard being those which are low-grade wood with the clay and paint overlay, which tends to chip and break in shipping… if it isn’t already broken in this way at time of purchase. The ones at non-art stores tend to be plastic molding with a plastic wood-grain sheet ironed onto them… again… trying to avoid that.

So, yesterday, after browsing many ready made frames at all the aforementioned stores, not finding anything I liked at a reasonable price – and finding mostly overpriced junk, I decided to try to make my own.

The first step was a trip to the hardware store for lumber, where I purchased four 6’x4″x1″ pieces of wood.. having struck out on all avenues in looking for a source for the wood strips framing galleries use. I’ll have to try that some other day.

So… wood…

I’d like to say I went for the fine cherry, a nice fire maple, or something moderately exotic – but what I ended up going for was a compromise.

At Home Depot, I found their “exotic wood” be be very exotically priced. I am used to prices being high on hardware at woodcraft – but Woodcraft’s prices certainly beat (by light years) the wood prices at Home Depot. I will have to remember: Woodcraft for wood, Home Depot for hardware.. but, of course it was late at night when we went… So Home Depot was pretty much where we were going for the lumber.

Pine being cheap and sub-quality wood, I looked to avoid that route – but in the good wood section I found what is called “select pine” – which is not quite gold-grade heartwood pine, but certainly not shabby construction-grade pine. It is nice and heavy, very dense actually, with a really fine grain and very tiny knots. I was rather surprised by how sturdy and dense it was, and at $4.19 a  6 foot piece – I decided to give it a try (and having tried it, I am actually quite happy with it).

[Edit] Having returned, and taking a closer look at Home Depot’s stock, I also found some really nice Maple there for about $1.40 a linear foot – still more expensive than the pine, about twice the price but certainly not bad. I also found cherry and red oak shelf capping for around $1 a foot, which I plan to use in future projects in place of the moulding below.


I also grabbed some decorative wood strips from Home Depot (sort of like above, but not the same design)… rounded and decorative on one side, flat on another. My plan is to use these to make the frames a touch more ornate. The price wasn’t bad.. $2.16.. Oh? Not “per piece” .. that’s a linear foot? Okay… I bought one six foot section. Also in my arsenal is some brass chain at Hobby Lobby (clearance priced)… about 120 inches of it. Home Depot had wood appliques – but their prices on those were ridiculous – I am supposing they were trying to get as much as possible out of cabinet makers. The things are incredibly cheap by comparison at Michaels… I’ll have to go there tomorrow, as they didn’t have them at Hobby Lobby and I wasn’t about to pay the Home Depot price for them.

Wood Applique
Wood Applique

So… first step: Cutting the wood

So… first step… measuring the wood.

These artworks are 5×7 inches. So, I want the inside of my frame to be about 1/8 of in inch smaller than that. Typically I’ll sit there and figure up the best mathematical formulae to get the perfect result – but diving right in, I simply measured 1/8 of an inch in, and used my 45-degree plastic triangle to mark where the first 45 degree cut would intersect that drawn line. I measured 6 + 3/4 (7 inches minus two times 1/8 of an inch) of an inch along that line, marked that point, and used the triangle to draw a line where the next cut would go.

From there I carefully and precisely made the first cut on the table saw, using my 45-degree angle-guide. Then I carefully and precisely made my second cut. From there it was just a matter of sliding forward my 45-degree guide, placing one edge against the (stopped) table saw blade, and lining up the straight guide to the end of that piece. After this I could just cut, flip the wood over, cut, and flip the wood over, making perfectly identical trapezoidal wooden pieces… my 7-inch sides.

Table Saw: The same 45 degree jig and same position as the very first cut. The difference is the piece is flipped and moved to the guide in order to reflect the opposite angle, and the achieve the right length. This can also be done with a miter box, but the table saw really makes it go much faster.
Table Saw: The same 45 degree jig and same position as the very first cut. The difference is the piece is flipped and moved to the guide in order to reflect the opposite angle, and the achieve the right length. This can also be done with a miter box, but the table saw really makes it go much faster.
Table Saw Guide - My clamp keeps the wood from popping/leaning up, or sliding along the guide.
Table Saw Guide - My clamp keeps the wood from popping/leaning up, or sliding along the guide.

After cutting the first two, just in case my angle guide was not entirely accurate, I matched up corners on the cut pieces, and checked them with my triangle, to make sure they were squaring up. They weren’t. So I made an adjustment of half a degree – and used these pieces over and over again until the angle was perfect.

One of the cut pieces. Remember - Trapezoid, not parallelogram. No adjustments need to be made to the guide for this, just flip the wood. Also note my sleeves... not recommended for anyone without a death wish. Loose sleeves, long sleeves, are not the best attire for workshops.
One of the cut pieces. Remember - Trapezoid, not parallelogram. No adjustments need to be made to the guide for this, just flip the wood. Also note my sleeves... not recommended for anyone without a death wish. Loose sleeves, long sleeves, are not the best attire for workshops.

I made 18 of these long sides, saving the first two ‘test pieces’ for use in making the shorter sides, since they were shorter after all the test angling.

Then I made the 5-inch sides, in the same manner as above, until I 18 of those as well.

After this, I needed to make it so that the artwork could rest inside the frame, meaning that 1/8 inch recess (rabbet) planned for needed to be carved into the inside dimensions along the back of the frame-to-be. This is a good time to have a router (and a router table is all the better).

*If you do not have a router, which I am sure a lot of you may not, I would recommend starting with decorative beading (the sort of trim that, in a house, runs along a floor or a ceiling… available at most lumber stores). From there you can just build a simple square and square-edged frame for your work, with the inside of this being the exact size of your artwork (n0 need for a rabbet), but use the measuring and cutting steps above on the trim. The trim can be assembled on top of the box frame, glued, pressed, and clamped for a very nice look and sturdy frame. Those are the basics… I’ll go over that particular method in more detail, but it will have to wait for another post*

Router and Router table - I set up two guides for this one. One to set how far back towards the bit the wood piece can go, one to keep it down to the table (while at the same time to keep my fingers away from the bit)
Router and Router table - I set up two guides for this one. One to set how far back towards the bit the wood piece can go, one to keep it down to the table (while at the same time to keep my fingers away from the bit)

I used a router bit which has a curved into flat bevel, called an “ogee”. The flat edges would hold the artwork well, and the rounded surface would allow for the least amount of surface contact on the art side of the artwork.

common router bits and their profiles
common router bits and their profiles
Routed edge - better than a rabeted edge, because the curve keeps more of the arts surface away from the frame itself, which prevents a rubbing on the art surface.
Routed edge - better than a rabeted edge, because the curve keeps more of the art's surface away from the frame itself, which prevents a rubbing on the art surface.

I also used this same decorative “Ogee” bevel on the inside front as well. As a bonus, being the same on the inside dimension front to back, I could choose which side looked the best (according to the smoothness of the cut, the look of the grain, dents and dings, etc.) .

Having chosen which side would face front, I then took to beveling the Outer Edge on front side, giving it a smoothed and decorative face on the front (inside dimension and outside), but leaving the back flat and flush. For several of my frames, I also made the back outer edge beveled as well, giving the frames a nice rounded edge.

Three routed edges - two on the inside near the artwork, one on the front and outside of the frame. The fourth edge is left plain so it can be flush with the wall.
Three routed edges - two on the inside near the artwork (against my thumb), one on the front and outside of the frame. The fourth edge is left plain so it can be flush with the wall.

Once I had all my pieces done, I quickly sanded off all frayed edges and burs from my cuts, so the frayed ends would not get in the way of the next step: gluing.

A stack of frame pieces, ready to debur, glue, and press.
A stack of frame pieces, ready to debur, glue, and press.

Glue, actually holds more securely than nails or screws, if done properly, and of course keeps unsightly nails and screws from your frames. Such also negates the risk of splitting your pieces with such fasteners during the construction process.. which would be a terrible waste of all the work done up to now. There are special fasteners for framing (to be used later), but the glue is most important.. and will make that fastener step much easier if you choose to use fasteners.

I used a four-point box clamp for this. It makes jobs like framing, assembling stretcher bars for canvases, making boxes/cabinets, all go much more smoothly with these sort of clamps. If you don’t have them, I’d recommend using two (preferable four) bar clamps or pole clamps for this step.

4-way clamp around the frame, glue at each edge. The pressure pushed glue deep into the end grain, giving the frames a very strong and sturdy bond.
4-way clamp around the frame, glue at each edge. The pressure pushed glue deep into the end grain, giving the frames a very strong and sturdy bond.

Put glue along the ends of each piece, in thin streams, but try to cover most of the ends, and try as best you can to get the very edges of each end. Clamp them together, make sure all pieces line up square and even. Un-clamp and re-clamp if necessary… Such is much better than letting it dry uneven and having to redo this step or sand things down.

Let the frame sit this way for at least a half hour (preferably longer) before moving it. If it has only been sitting a half hour, remove move it carefully and then leave it be. You will however get better results if you leave each frame clamped for an hour or more. I plan to use 3/8 corrugated fasteners on each corner – they aren’t completely necessary, but they give a bit of added peace of mind to neurotic people such as me.

corrugated fastener
corrugated fastener
Not-so-finished product. Staining and other decoration yet to be done, as well as the addition of the hanging hardware. The artwork is not secure yet btw; It is just there for the photo.
Not-so-finished product. Staining and other decoration yet to be done, as well as the addition of the hanging hardware. The artwork is not secure yet btw; It is just there for the photo.

Next steps – added (wood) decoration, staining, added decoration (metal), placing the artwork, hardware (for hanging).

(Next installment)