How to make shirts?

July 11, 2005 9:01 PM

I've been actively trying to create shirts of the quality seen here. While I'm less concerned with the technical aspects of creating a vintage/fatigued looked -- how does one go through the initial stages of design? (What program is best to design in, how should I approach the silkscreener, fabric acquisition...)

Where would be the best place to get the luxurious cotton used in the t-shirts made by Grail, Rebel Yell, Paper Cloth & Denim... I'm really frusterated as I cannot find anything to match the lightweight cotton fabric used among these. Do such companies travel to Italy or is a trip to Manhattan sufficient?

What would be the best format to bring to a silkscreener? Right now I have a virtual replication of the t-shirt in Photoshop to scale, the design is somewhat complicated and I'd like to ease any technical problems before they start. I can easily redo my designs in Illustrator if that's easier to work with.

I'm convinced there's some sort of standard process for this designer casual wear, as everyone and their brother seems to have a company. I'm just out of the loop. I've talked to people in the apparell industry but my contacts are unfortunately within large multinationals geared toward totally different markets.

If someone in the know is reading this and is aghast and tells me "I have no idea what I'm in for", please feel free to say it.

posted by geoff. to clothing, beauty, & fashion (6 comments total)

As to t-shirt selection...

Nothing beats American Apparel. They've got what you're looking for, and sweatshop-free to boot! Most t-shirters are using them now (although Threadless.com switched to 50/50 fruit of the looms for the guys, telling me that too many guys complained about the fit of AA shirts) and they have that slim, fitted cut you appear to be craving. Careful -- they do run smaller than you might be used to. As for the silkscreening...

Most I know do it themselves -- hell of a lot less than paying someone else to do it and far more control over the process. It'll take you a few tries to get it right, but eventually you'll be a pro. It's not rocket science, either. Of course, this depends on how many colors we're talking about. Google around for silkscreening -- there are loads of sites about just such a thing. You can start out with a very basic kit, or you can dive in and get the bigger set-up. Completely up to you.

Those I know who've been successful at t-shirting have a variety of designs to go right away, as well as a good outlet for those designs. I'm in LA, so we're talking Fred Segel, Kitson, Barney's, for the high end stuff. If you're targeting that market (and it seems you might be, since you're referencing extremely high-end t-shirters), then getting your t-shirts to that market will take some scrapping. Everyone I've known seems to have done the consignment thing, and it's worked out well for them. I'd start local first, testing out the marketplace. Another reason to do the printing yourself -- you can print more of the most popular, as opposed to needing to anticipate demand weeks in advance when working with a silkscreener.

Don't ignore the lower end market, though -- much less margin, but much greater potential for volume. (duh, incessant)

Another idea -- buy bulk vintage tees from thrift stores. Silkscreen over the images already there, or turn inside out, screen, and cut out the tags. Totally cheap, recycling materials, and you wouldn't believe how much people flip for this stuff. My friend cleared at least a thousand tees in six months at 50 bucks a pop at the trendiest stores in LA. People are dumb.
posted by incessant at 11:47 PM on July 11, 2005

American Apparel Ts, for all their hype (since 2002 or so), are good but definitely not as good as the shirts that Rebel Yell and Paper denim source. Those companies scour the globe for their shirts and the shirts cost a bit more (prolly $5 or $7 each instead of the AA wholesale $2 or $3). I don't know exactly how to find 'em, but I think Thailand, Malaysia, and Vietnam are the countries who have the textiles that you're looking for. Just saying, definitely get some shirts from AA to check out, and maybe find a way to distress them yourself -- 1 to 20 bleach and sand in a bucket much -- but they're not thin as the others, esp in men's styles.

Another complication for you will be the "all over" effect that's shown in your link, and in many recent shirts from Rebel Yell, and other hipper t-shirt companies. Many standard screen printers will not print on the sleeves of the shirts and stuff, at least not without some prodding. This argues for doing it yourself. Besides, you're sure to have some happy accidents that will turn out to be your favorite stuff.
posted by zpousman at 7:41 AM on July 12, 2005

$85 freakin' bucks for a fake vintage worn-out t-shirt? I'm in the wrong business, and I left the right business over a decade early. No, scratch that. I don't want to go back to the unmitigated hell and crapflooding that is the sportswear industry. idontidontidont. Anyway. :)

You have no idea what kind of specifically fresh hell you're in for. :) But go for it anyway.

There is a general methodology to prepping art for screen printing, but it does vary somewhat widely from shop to shop. A good shop will be glad to answer your questions or provide a spec sheet on what they consider to be "screen ready", "camera ready" or "output ready" artwork, depending on their process and equipment.

Most basic screen printing is done in 'discrete' separation. (See also: Spot color.) Each color gets its own solid plate of black. For basics, it's usually butt-to-butt registration with no trapping or overprint.

For printing on darks, a white or light underprint is usually needed. The underprint is usually slightly "choked", or very marginally smaller than the sum total of all the overprinted spot colors.

To save art charges, you're going to want to separate them yourself. Each shop does it a little differently. If they're oldschool, a black plate of each color, with registration marks, and a color comp should be fine.

Discrete separation can also be more complex then that. You can do crashes (color on color), half-tones, half-tone crashes on solid color, or half tone crash blends with two or more half tones of spot color blending together. This can be a pain in the ass to replicate on screen even in Photoshop or Illustrator without knowing the reflectivity values of your substrate (cloth) the transparency/opacity of the inks, and the saturation of the pigments.

The reason why they do discrete is A) It's technically easier for the screen print process - it always has been and it always will, it's basically a stencil-based system. And B) the color is pigment based, not based on the CMYK process, which is more technically difficult for the screen print process.

CMYK (and even Hexachrome or extended-gamut process) can certainly be done on screen printing. But getting really good color and consistent quality is a black art nearly or literally exceeding that of CMYK process on paper or other flatstocks with offset litho printing.

So, before getting too crazy with the design end, talk to your local print shops. Talk to out of the area print shops. Get art specs to find out what they can handle. Get quotes based on actual inked image size, number of colors, type of print.

Printing on or over sleeves and seams (spot prints and over prints) will quickly increase the per unit cost. It increases the amount of time, labor, and handling of the shirts exponentially. It will also create more rejects due to wrinkles and misalignments in the printing, a cost either you or the printer has to factor in and eat.

However, if you go to a company like American, or a company like Murina, for your shirts or fabric, they can probably provide you with in-stock or custom "cut piece" stock. These are unassembled, pre-cut t-shirt components.

For doing stuff like that rainbow shirt across the sleeve, this will be the best way to do it. But it requires a fanatical, insanely anal level of attention to detail that is a full time, very stressful job to manage across multiple suppliers. Your cut pieces have to be accurate. Your art has to be accurate. Your printing has to be accurate. Your sewing has to be accurate. In millimeters (or less). Across dozens of steps. During which any number of things could happen. Fabric can shrink. Before or after printing. Fabric can be stretched in the printing process. Prints on fabric can be off by centimeters, which is an industry default standard, generally. I can say with a great deal of certainty - excluding printed on-roll material - that t-shirts or cut stock is hand-placed on the adhesive-coated printing pallete, and hand removed.

It's brutal, brutal work.

The best option for large numbers (Thousands or tens of thousands of shirts. Or hundreds of thousands.) is to contract the whole thing out with a specified reject rate. Industry standard is about 2%.

They'll charge you more for this kind of high end work per unit, but if it's an all-inclusive contract (Art, plates/outputs, screens, color matching, ink mixing, ink purchasing, fabric/cut-piece/shirt sourcing and purchasing, folding/bagging/tagging, neck labels, sewing, and even pre-assembled cross-dock shipping directly to distributors or stores) and it's a good company, you'll save yourself years of your life due to stress, and make a willing printer/producer happy. Because then they have full quality control, even if they sub-contract work, which they must likely will.

For smaller numbers, there's alternatives. You can set up perfectly adequate and useable but simple screen print rigs for far less than the initial order cost of a large run. Barely. (And not considering also buying sewing equipment and everything else) There's lots of resources out there about how to print and what to buy. Check out the archives of Screen Printing and Imprint magazines.

And there's another alternative. You can have "hot split" transfers printed of various designs or elements of designs. A "hot split" is not like a bubblejet transfer. It's a screen print on transfer paper, with specialty inks and processes. Be warned that I believe hot split transfers have a shelf life.

The major upside of this technique is that you can order a run of printed designs - I have no idea what the minimum is - and you can store them for a period of time and apply them as-needed with a small heat-press transfer machine to completed shirts or cut-pieces, with as much accuracy as you'd care to put into it.

If you go this route, ask the printer what heat-press machine they recommend and use. AFAIR you're going to want a digitally controlled pneumatic model. Hot splits are finicky on application, but gorgeous and durable once properly applied. They work well on all colors. They distress well, or you can build the 'distress' into the design.

AFAIR InstaPress is the industry standard for quality for presses. Expect to spend at least a grand on one, if not more, depending on the size of the heat element and platen. You'll also need a compressor.

If you go cut-piece then you'll either need to sew the imprinted yourself, or contract the sewing. If you can do or source the sewing yourself, you'll have better control. But a good company (like American) should be able provide highly accurate sewing, but probably at a premium.

Disclaimer, promotional semi self-link: For hotsplits and direct printing, I highly recommend Screen Art, Inc, in Huntington Beach, CA. My dad and one of my brothers work there. (Ask for Alan, my dad.) Screen Art practically invented the modern hot-split transfer - they're masters of it, and their direct screen printing is also world class. They do work for hundreds of internationally recognized clients. But they're not totally ginormous. It's only about a 8,000-15,000 sq foot shop.

My dad has been doing sportswear screen printing for over 30 years. He's a maniac about it. We used to run a print shop that had a couple of months that broke a million imprints in a month. (But barely broke even, if that. Tough business. Really tough. Why aren't we still in business? Short story: NAFTA, investor embezzlement and an overzealous IRS.)

Screen Art is one of the last of a dying breed in that area. There are very few large-run but impeccable quality artisan-level apparel screen print houses left in the Continental United States. I would recommend them even if my dad didn't work there - we used to subcontract to them all the time when demand exceeded our production. They run a fully compliant, clean shop and treat their workers exceedingly well. A rarity in the clothing industry. (American Apparel also treats their workers very well. Excellent company.)

They're also honest. If they don't want your work or can't handle it - either from being at full capacity or from not wanting to handle the specific technical requirements - they'll tell you.

Best of luck. Email me if you have any other questions.
posted by loquacious at 9:51 AM on July 12, 2005 [2 favorites has favorites]

loquacious: THANKS! you've inadvertently saved me a bunch of work -- I was supposed to track down a company that did "hot splits" (which my friend kept referring to as "transfer prints that have real ink in them, like as if they were silkscreened") ... Thanks for the massive post. I find it hugely useful and I'm sure other folks will too.
posted by fishfucker at 10:14 AM on July 12, 2005

oh, and geoff:

in my experience, selling shirts for mad money ain't about the tedious process of making the shirts, it's about the branding and marketing. If you really want to charge outrageous prices for shirts, you should look into phenomena like Just Another Rich Kid and the types of stuff that gets linked by Cool Hunting.

Anecdotal experience: My girlfriend makes similarly styled shirts and clothing -- and although they are good quality clothes that sell well and are well-loved by folks in Sacramento, she's found that SF and LA are more reluctant to take her clothes, because dude, the market is FLOODED with this shit right now. Trust me, you're not the first to have this idea. I'd imagine the main way to get it going is to a) either be able to nail down a niche market (the nerd market, for example, if you think you know it well) or b) bust out with a viral website.

good luck!
posted by fishfucker at 10:25 AM on July 12, 2005

Fishfucker: You're welcome. I hope it works out.

And yeah, the words "$100 dollar t-shirts" tend to saturate the market pretty quickly. I heard about this faux-vintage thing a couple of years ago and was just totally appalled. But then I've been shopping in thrift stores since I was a kid, and since I used to have access to enough t-shirts to wear a different one every day for years, I was mighty confused as to why anyone would want to pay 100 bucks for a worn-out shirt that cost about 7 bucks to make. But then again, $20-$30 dollar shirts confused me as well. I used to literally use them as shop rags or even bath towels in a pinch.
posted by loquacious at 10:51 AM on July 12, 2005