Spyder II Direct to Screen

[zanegun08]

The Direct to Screen saga has been a little bit of a long one for me and the company I'm with.  We had an unsuccessful attempt at installing a well known unit, that ultimately we decided to return, which was subsequently moved down the road and is working for another shop.

All that aside, I'm still apprehensive about the reliability of Direct to Screen in general, but there are so many benefits, that it outweighs the negatives.

After doing countless hours of research, talking to users on this forum, as well as others, I was on Team Douthitt, come ISS Longbeach the owner decided to go with Spyder for maybe just the reason to go against me, but the end result is great so far.  First off, if you are on the fence about Direct to Screen in general, make the jump.  It is great, and there are so many benefits in all areas that the sooner you can do it the better.  Secondly, if you are considering Douthitt, Mark is an awesome, helpful guy, so I'm sure you won't go wrong.

Spyder I believe is building these all to order, so from a few weeks after ISS when we ordered, to the install on March 25th is about how long it takes for delivery.  The Spyder II is new for them, and the main difference I believe is bi-direction printing, and touch screen control panel.

Alfredo who is at ISS personally came out with two other techs Kim, and JC who writes the software and knows the machine very well.  The install started out a little rocky with figuring out template sizes and placement on our screens, but that could've been compounded by my own preferences since I like to push limits of printing sizes so to have more options in the future.  I can print larger on the screen than we can actually do on press due to squeegees and floodbars.  But to have the option to push it an extra couple inches now, be able to print both sides of the screens at once with different images (this effectively makes tri-loc not work on the bottom image) and never have to change settings on the machine later was very important to me.  I'm still refining my templates to my liking to this day, but I have it now, to where unless any major changes are made, then our art will consistently be on the correct place without any changes from the operator.

We got the machine with two frame sizes for 23" x 31" and 25" x 36", I have since put in an order for our 20" x 20" manual frames, as well as our rapid tag frames so that everything we print will be able to be printed computer to screen.  Besides the case of super jumbo which we will still be dependent on film but that is like 1% of our work.

Stealing from another thread about how to compare CTS / DTS -

Machine Cost

The cost is lower than the I-Image STE-I / II, but is more than the Legend and some other ink based units.  Comes in way lower than the Kiwo I-Jet II, and is comparable to the Douthitt

Production speed and ease of use (in art department)

This is what frustrates me about CTS / DTS in general is that they don't show the workflow beforehand, and are almost a bit secretive about it.  As well as since the sales force or tech's aren't in the art department, I don't think they realize the importance of workflow in the art department.

With that said, having experience with both, I much prefer the RIP that comes with Spyder than the M&R.  We currently use a Harlequin RIP so nothing really changed besides getting a newer version that is extremely fast, has more options than I could ever know what to do with (not in a bad way) as well as can actually preview (or roam) the images correctly, not a fuzzy image.

Since our RIP is on a dedicated windows machine, we hit print like normal, put in our halftone settings, or leave blank if vector, and hit print like normal.  Then the RIP outputs the files to a server that we can check remotely so we can actually sort of "Roam" preview the files from our desk's which is much more efficient.  One thing that is a bit annoying to me is that the file names have to be windows compliant so if they have %, /, *, or names that don't work on windows, it will fail to create the file.  You can strip those from the file names, but then it gets rid of spaces as well, so it makes it hard to read.  A file name ends up looking like this - 5508962 Farmers Market - FB.ai (White)00

Imaging speed

The Spyder prints reasonably fast, compared to the I-Image I believe it is slower, but not by much.  The plus is we get a very opaque image, great halftones with 2% halftones coming out perfectly which is an improvement over our films.

Screen loading speed

They Spyder is super easy and quick to load.  As well as from the time you hit print, to it actually printing the file seems fast as well.  Loading is simple, put the screen in position, and click the foot pedal.  It has tri-loc blocks built in, so there is a bit of a learning curve to make sure you are loading accurately.  But if you load it right, print a screen, unload it, and then put it back on and hit print again, it will image right on top of each other perfectly.  It is important on any machine to load the screen properly!

Image quality

The Spyder seems to image awesome, the wax is super opaque, and we are getting better halftones than films.  I made an exposer test that we then printed on press, and we held a 1% dot all the way up.  Our shop as a whole has a dot gain issue, we print with too much pressure, so I have been working on a tone curve which I'm still developing to get smooth halftones all the way to solid.

Straight off default settings we got 1% - 80% and then it turned solid, this was on auto where I know they crank the pressure.  I curved back the top portion, and now we are holding halftones from 1% - 90% and then it turns solid.  I still want to develop my curve further, to compensate for the dot gain issue on press, I'm just concerned with flatting out the mid range and loosing smooth tones so if anyone has any tips there it would be greatly appreciated.  This is done within the RIP, which is easy to make adjustments, but I'm simultaneously working on getting people to print with less pressure (slowly improving) as well as using S thread and LX mesh, which we can't do until we fix our pressure issue since we pop screens, but we can't print with less pressure due to our current mesh so it is a paradox!

We are currently using Kiwo Versatex but are also experimenting with a Murakami emulsion which both work great with the Spyder, the wax will really stick to any emulsion I'm sure, so you will have more freedoms than with a wax based vs ink based unit.


Ease of machine use

We have multiple people running the unit and I don't think the operation of the machine could be much easier.  As long as when they import the file they select from the correct template for the screen size, (screen size will be in file name for everything besides standard 23" x 31") and make sure that the file they want to print is at the top of the queue, that is pretty much all you have to do.  Make sure you pick the right mesh, file (based of number which is searchable and will bring up all the layers), template, and that it's at the top of the queue, then hit print. 

They made it pretty simple, although I still think it would work best if we could just have one template that would work on all frame sizes, that then if the image is taller than the frame size it would automatically error out.  The reason this can't be done is because to get the right position on the frame, you have to offset from y.  The issue is it offsets from the top, instead of the bottom so it wouldn't tri-lock correctly if you went off that axis.  Maybe in version two of the software they can address this.  You got that JC?


INK/WAX

Admittedly the wax is going to be more expensive than inkjet, although I think we get better quality, as well as that we have way less head clog issues.  Like any printer, the more it runs, the less it clogs.  The blocks of wax are $40 from exile, which when we are doing ~150 - 200 screens a day I think we will go through a block a day, of course it varies based off the size of the print, density, etc.  This one is hard to just throw out a number, but next week I plan on resetting the counters, and estimating wax cost, but hoping it is around ~.25 or less per screen.  The good thing is that not having to catalog the films, find them, pin them, expose them, peel tape, recatalog, this ~.25 is a small cost to pay to save that in labor, as well as room for storage of our 120,000+ films

Exposure options

Currently using a Nuarc light, but I am going to demo a starlight, and I like the vertical setup that LuckyFly posted https://www.youtube.com/watch?v=UifJ3t6ziKs.  Although I would face it towards the wall so that it isn't blasting your eyes all day!  Need to get that operator some sunglasses!

Customer Service.

This is yet to be known, ask me in a year.  I know that Douthit has great support, but with Exile everything has been great so far with Alfredo, time will tell on this, but from what I have seen so far it has been great.  JC even emailed me over the weekend so I could set up my templates to my specifications after he just got off the plane and back to his place, so that was above and beyond the call of duty.

Recap

So far, everything with the Spyder has been going great.  I am involved in Art, Sales, Operations, and many things, so already I have seen that I'm called down much less because the machine is running great, intuitive, and has less learning curve from the beginning.  Exile has done a good job, and once we get our new drums and can go completely CTS we may get to the point of needing two since it is going to increase work flow in so many places, art, screens, exposing, setups, reclaim.

If you are on the fence of going CTS, do it.  Secondly, if you are between any of the CTS units, anyone is better than none.  I've heard good things about all three, I'm on Team Spyder, but it is still early on since this is a new unit, and based on our past experience a bit apprehensive.  If it keeps working like it does, which Alfredo promises it will, then we will be very happy.

If you have any questions feel free to ask and I'll do my best to respond.

http://www.exiletech.com/products/spyder.html

Photos of the Machine and Software




What an output file looks like


I also attached an uncompressed tif as well, the files range from 50k - 2mb depending on complexity.  Most are in the 100-500kb range.

[LuckyFlyinROUSH]

Hey man! I like the write up, very solid. Can you PM me the price of that unit? I am just curious.

I do however like the vertical stance of this unit, I just feared that it wouldn't work with trilock as well as the I-image. Let us know how it turns out. Also preferred dye rather than wax, but that's a personal preference!

As to the starlight! When you have a screen on it and exposing you can stare at it and not have any issues. That is me in the video haha. Also we usually hit the button and have enough time to walk past to move a screen to the dip tank, etc before it turns on. I suggest you play with it as this vertical setup is absolutely killer! Or if your around KC you are welcome to come play with it.

[Dottonedan]

Wow!  What a great post. Very good info.


There are two areas that I will clarify as it pertains to my experience.


*The fuzzy preview (speaking as it pertains to I-Image) is in the Color Print (RIP) program. It's a low rez proxy and it's intent is for a quick preview of print location and not image quality. (e.g.) making sure it's going Top center. I'ts kind of a confirmation that a left chest is going in the correct position. It's also where you can confirm you've got the right number of colors coming out. We do have a quality image preview and that is in the program that you are actually going to print to the machine with. That is the Print Production program.  There is an option for previewing all colors together or just one color at a time. Most prefer to preview just the one color. The option to preview all colors together can take (2-5 seconds) to refresh depending on number of colors in the file, versus instant 1 color. Most don't need or want a full color preview here, but it's available. The color previewing (confirming the seps are coming to the machine correctly. Typically or should be, done in the art department before it is ripped. In our Print Production program, you can preview actual halftone dots and various size previews. This is a good checkpoint for photoshop files to confirm you don't have any unwanted dots or % hanging out in the area outside the art where nothing but shirt color should be.

[size=0px]We are currently using Kiwo Versatex but are also experimenting with a Murakami emulsion which both work great with the Spyder, the wax will really stick to any emulsion I'm sure, so you will have more freedoms than with a ink based vs wax based unit.[/size]

[size=0px]


I don't have experience with the other ink based machines, but I do have with the I-Image and emulsions. Our concern is never (will it stick or bond) but how fast or slow it will expose. That's an emulsion solids subject and not related to the ink it'self. We have no issues that I've ever heard of, with our inks not working with any type of emulsion. When you refer to sticking, I take it that you might more so, be referring to proper exposure ability. I can't see any body's ink [/size]not sticking or bonding. I don't know anything about how wax washes out, but our dye based ink runs down and dissolves at first touch with water and then easily blowout the non exposed emulsion. Washing out of the emulsion is never an issue and opacity even at one head, is never an issue. For us, opacity is controlled by print passes and either uni or bi direction.

Just yesterday, my customer and I did a test of 70lpi, on a 305 mesh, coated 1:1 round, holding down to 2% and up to 99% at 70 lpi (in the screen). Then we printed it on the shirt and held the same 2% and down to 95% on press due to gain, with black ink on white G2000. I know that part excited you, as I get excited about the details of adjusting the curves and making it work on press as well.  For that, here is a tip below, that will help you.

* When adjusting your gain, use a smooth gradation bar, as well as a section bar. Section, meaning 1%,2%,3%,4%% on up to 100%... but the smooth gradation bar that goes from 100% solid down to 0% with (no breaks or sections) is the one that will revile any bumps in the curve. Then, RIP that it, so you have your original to compare to... and make another curve with a slight adjustment in the shadow tones and RIP it again. Then preview both, back and fourth, and you can begin to see where you need to make more adjustments if any. It will show any slight bumps in the gradation. Once you like what you see, print both on the same screen, let the press operators print a few strike offs to get ink flowing then on the 3rd or 4th, check the print (from the press) and see where you might need to make more cut backs.   What you are looking for is a smooth transition. Curve (points) can be off by a single % and be recognizable in a smooth gradation. That's why it's important to test or preview smooth gradations and not just sections of %.

I agree, that the vertical use of the Starlite is pretty cool.

(edit)  One more thing, The "clogging" you refer to, for us, is not an issue with wet ink. I don't know about other Co's.  If referring to inks clogging and staying in heads and having to replace heads. Thats not us. We have different heads, (not Epson) and our inks don't dry up in heads tow here you have to buy a new head. If we get a "clog" or a block, it's air coming back into the lines for some reason such as turning the machine off over a weekend or holiday and not having any Negative pressure regulation, or, not having the valve totally closed. All you do for that, is do an auto clean or two, and you're up and running. Worse case is that you would need to purge the air out and that can take 1-3 min depending on how many heads you have.

Much appreciate the good feedback and appreciate that it's unbiased, and just good honest feedback for others to research.

Thanks
Dan

[jvanick]

I'll add that if you're really geeky about gain, buy yourself a reflective densitometer on ebay.... you can get one around $300 or so shipped.

It will really open your eyes to just how much gain excessive squeegee pressure, bad curves, thickness of emulsion really causes.

Sometimes I wonder if the cts vendors should just include them, as the results end up so much better if you can linearize the entire process from art to print, your proofs can look the same as what you print without your artists "knowing" the particular press or press op.

[Dottonedan]

I'll add that if you're really geeky about gain, buy yourself a reflective densitometer on ebay.... you can get one around $300 or so shipped.

It will really open your eyes to just how much gain excessive squeegee pressure, bad curves, thickness of emulsion really causes.

Sometimes I wonder if the cts vendors should just include them, as the results end up so much better if you can linearize the entire process from art to print, your proofs can look the same as what you print without your artists "knowing" the particular press or press op.

Ooh  good one. The need for a densitometor is way WAY under valued/under promoted in our industry for film and DTS alike. Look at all the years we've had film and never went as far as really testing our film output to confirm what % we put where. It's one of those things most consider not that beneficial....till you have one.

[jvanick]

Especially once you start to learn that 40-50 percent dot gain is not unusual....  Heck, you can cause 20% dot gain just by having too much squeegee pressure.

[zanegun08]

I'll add that if you're really geeky about gain, buy yourself a reflective densitometer on ebay.... you can get one around $300 or so shipped.

It will really open your eyes to just how much gain excessive squeegee pressure, bad curves, thickness of emulsion really causes.

Sometimes I wonder if the cts vendors should just include them, as the results end up so much better if you can linearize the entire process from art to print, your proofs can look the same as what you print without your artists "knowing" the particular press or press op.

Alfredo from Exile brought a densomiter during the install to help with the tone curve while he was here as well as linearize the machine.

Do you have any links to a good densomiter that can read the image on the shirt, I would love to develop my tone curve more, but as of now I'm just guessing based off sight.  I would like to get more scientific about it though and really get it perfect for us.

Ooh  good one. The need for a densitometor is way WAY under valued/under promoted in our industry for film and DTS alike. Look at all the years we've had film and never went as far as really testing our film output to confirm what % we put where. It's one of those things most consider not that beneficial....till you have one.

I attached what I used for a dot gain test, I recall seeing one posted once, but I couldn't seem to find one so I made my own that I output at different LPI and dot shape for comparison on the same screen.

* When adjusting your gain, use a smooth gradation bar, as well as a section bar. Section, meaning 1%,2%,3%,4%% on up to 100%... but the smooth gradation bar that goes from 100% solid down to 0% with (no breaks or sections) is the one that will revile any bumps in the curve. Then, RIP that it, so you have your original to compare to... and make another curve with a slight adjustment in the shadow tones and RIP it again. Then preview both, back and fourth, and you can begin to see where you need to make more adjustments if any. It will show any slight bumps in the gradation. Once you like what you see, print both on the same screen, let the press operators print a few strike offs to get ink flowing then on the 3rd or 4th, check the print (from the press) and see where you might need to make more cut backs.   What you are looking for is a smooth transition. Curve (points) can be off by a single % and be recognizable in a smooth gradation. That's why it's important to test or preview smooth gradations and not just sections of %.

I believe you are talking about the bottom bar in my test, is this correct so you can see the full range, vs squares of percentages?

I'll post my tone curve as I have it right now, although it still isn't exactly where I want it as I want to hold dots in the higher percentages since we have lots of gain in that area, while not so much in the low amount.

At what point is realistic to stop seeing halftones?  Should visually 98% look like 100% black, or should we see halftones in the print there as well.  Where is Coudray when we need him?  To me on screen at 95% it pretty much looks pure black, or at least like pantone black.  What is a realistic goal for me to compensate too.

[Dottonedan]

Yes, that's the one. A nice smooth gradation bar.


Using a densitometor (on your screen) will not get you real accurate reads nor your real target. If you are reading what you are putting on your emulsion, then you have to zero out the densitomitor to your emulsion color and density. That gives you various reads and is inconsistent since you don't print onto film or paper. If you did a test printing from your machine onto paper resting on top of a screen, it's not very beneficial. Sure, you can get adjust and get closer or even accurate to whats on the paper, but that doesn't help you for what you get in the end on press.


You want to read your shirt print (that has the dot gain on it) and go back and adjust your curves (for that). Thats a reflective and a trans-missive densitomiter you need.

[zanegun08]

As a quick follow up, here is my favorite thing about our CTS printer.  Got a jig built so we can load our Rapid Tag screens.  Going away from films is the best, it is fun to make a screen now, I love to do the art, hit print from illustrator, go load a screen, print it in 30 seconds, expose it in 20 seconds, and have it on press in 5-10 minutes from when I sat down at illustrator.  Not that it happens that often, but when it does need to it is nice that it is possible.



I built a template that we do size XXS / 5XL for everything using overprints, the logo and fabric content / details are on a registration layer, so if you need to build a new fabric content, change one line of text, art number, hit print and you have XXS - 5XL ready to go.  Do it once right so you never have to do it again.  We do a ton of relabeling so it was important to get something that is versatile to be able to accommodate many screen sizes.



The Spyder II has been working great, since this post, we have had very little issues, and the ones that we did have were resolved over the phone quickly and we were back running (air pressure drop)

We are so happy with the purchase, we removed a role of just filling and pulling films, streamlined art, and make better quality screens, that register much better, and have great support from Exile.

In my other thread about screen room / man-hours efficiency, on unrelated circumstances we had two guys leave (move from the city) so it solved itself for the time being and I'm now happier with the throughput vs labor.  And we are only improving our workflow.  Still haven't gotten rid of reg marks, my printers fought going from 6 marks to 2, previous people in my role were a little registration mark happy.

Anyhow, constant progression, there is a lot of talk about computer to screen, if you can swing it, do it, enough said.

[jsheridan]

It looks just like the Douthitt CTS..

What's the difference between the two?

[BorisB]

It looks just like the Douthitt CTS..

What's the difference between the two?

From my memory, might be wrong: newer version of Fuji printhead, more printmodes, bidirectional printing, improved printcontroller (whatever that means)

[zanegun08]

From my memory, might be wrong: newer version of Fuji printhead, more printmodes, bidirectional printing, improved printcontroller (whatever that means)

It is the same chassis as the Kiwo I-Jet 1, and Douthitt, but the internals are all different and moved.  Bi-Directional printing so it prints much faster, touch screen control panel, improved print controller which if is in regards to the software I like how it works, it is very simple to use, show the image in good detail, which has an option to zoom in.  Easy to call up files from numbering system, prints from tiffs which are small compressed file size, but high resolution, the rip is really fast.

It's a great machine, I've had experience with this and others and I'm really a fan of the Spyder, as well as they have great support, and this is a new machine for them, so they are a little more hungry which is a good thing.

[jsheridan]

Ok cool.
So essentially it's software package. Yeah that would make them completely different.

I used the Douthitt for a year and was wicked impressed with it. The density of the wax is amazing and we held some 85 line tonal work under testing conditions. Mark is a great guy to deal with and they were there every time I called, even remote in and fix it on the fly for ya kind of thing. 

Some of my issues were wax levels didn't match the size of the wax block, I had to break them in half or I'd get clogging of the vacuum line to the head, that made it drip out into the collection box, sometimes a huge hunk when got there in the morning after the early am auto warm up feature.
Head alignment was finicky and took a few adjustments for it to settle into place, until the next time someone bumped a screen into the head when loading a heavy screen.
Out of nowhere you'd get a clog and come back to a screen with missing info here, and then there but not over there but kinda over there so it got developed anyway but you don't realize till it's on press that the left head is missing from the tan and the right foot from the red screen.. whaa.. yeah.


 

[Colin]

Zane,

That was an awesome read through.  Looks like you guys have really stepped up your game!

Congrats!

[zanegun08]

Mark is a great guy to deal with and they were there every time I called, even remote in and fix it on the fly for ya kind of thing. 

Head alignment was finicky and took a few adjustments for it to settle into place.

Out of nowhere you'd get a clog and come back to a screen with missing info here

Every technology has it quirks, I've dealt with it on an I-Image as well as here.  About every 1 in 1000 screens the image will double image, meaning it prints a light ghost image, it can even be a certain color on a 10 color job.  The solution, switch to uni-directional, and move on.

Some times we get head clogs that when we first got the machine they would make it to press.  Luckily, we have learned to be more diligent on inspecting the screen, and raised the air pressure a bit to solve that issue.

We are happy overall, but have found our own way to deal with the quirks which every machine has.

Exile has also revised their software based off our input which was super nice of them to do at no cost.  Workflow is overlooked in some other setups in my opinion, so it is good that they were willing to take customer input to increase productivity.