Looking for manual stencil printer

[Mike]

I'm on the lookout for a better stencil printer than the crude frame I currently use.
Needs to be able to take unframed stencils with no fixing holes, including small ones (few inches), and a PCB area of at least 250x300mm.
Something like this
http://www.stencilsunlimited.com/sd-360u-smt-stencil-printer-p-441.html?osCsid=8bee3abbbb0d15233b05519a51f44a52

But two grand seems ridiculously expensive for a simple device like this.

[Gopher]

Phono has one he links to in another thread for €1200 or so.
Failing that one of our clients was talking about making one that he could then sell as well, somewhat cheaper than either of these options. I don't know how far that project has progressed tho.

[phonoplug]

My one is the littler brother of the one at the top of this thread. Unfortunately it won't do the width you need though, if you really do need 250mm. I think you can just about do 220mm width. Otherwise its a pretty good printer, though I did modify it have a piece of 12mm tooling plate for the PCB to sit on rather than using the PCB supports. It is removable though so you can still use the supports for double sided PCBs.

[Gopher]

I have had a bit of look for something that kind of size myself a while back, but came up with nothing. (worthy of its price tag at least)
The Technoprints certainly look the most convincing...

RANT:

Of course all there is no such thing as a PCB with free stencil, those that offer one as such typically charge more and the bigger the board gets the lower the proportional cost of a proper stencil becomes. On top of that the bigger the board in question and/or the higher the level of technology the less suitable the free stencil becomes, taking PCB-Pool as an example a "Free Stencil" has the following specs:
 

• Stainless-Steel
• 150 µm
• Size to be 10mm (all around) larger than the PCB
• No pad modification
• No finishing
• The paste film data wil be generated 1:1 from your PCB Layout data.

So it is basically
1) Too small -10mm is not that much to grip onto with some kind of mechanism, let alone giving somewhere for the paste to go as it flows.
2) Apertures almost certainly too large for smaller parts that would otherwise appreciate a 5-10% shrink
3)Too thick for anything remotely fine pitch - A board with 0.5mm pitch QFN's would want 100µm
4)Will put down too much paste under QFN, DPAK or something like a TO-252 - A stencil manufacturer would liaise with you, offering you their expertise to get the best results through aperture reduction, hatching etc.
5)Expensive -unless you need really quick turnaround or your board is small its easily cheaper to buy a real stencil elsewhere especially if you factor in the time you might spend finding shorts or dry joints from over-pasting, under-tension or indeed the fact these boards IME are simply not as good as production ones. (I've seen crap tinning, illegible silkscreen, misaligned silkscreen, peeling solder resist and crappy via's)

IMHO they are suitable for small PCB's using very std technology - say 0805 and 1.27mm SOIC's

Now yes, proper stencil frames and printers can be even more expensive (not hugely at the low end) but you could well come to find it's an investment you might never regret. I have seen several establishments in fact, using proper framing systems to paste what I assume were prototypes and simply placing the PCB on the table - presumably the printer was busy or the time needed to set it up not worthwhile.

I apologies for yet another off slightly topic rant, I only see these Free stencils things because customers sometimes provide them, we Never use them for in-house stuff (that may mean I'm spoilt) and where a proper stencil is too expensive we typically use plenty of flux, silver solder, tweezers and a soldering iron. The client who plans to make one of these printers himself gets remarkable results just using a syringe-  but then just about anything is possible given the right amount of patience & dexterity.

Perhaps this is one of those rare occasions when we could take inspiration from the USA, they seem to have nicely equipped workshops where enthusiasts, hobbyists, Geeks, & developers etc can come and use proper tools to do things they would not be able to do in their garage. There does seem to be a fairly significant concentration of potential users of such a workshop in the South-East, if a few of you got together you could buy some really quite nice 2nd user kit, share the expense and charge anyone else who wasn't part of the initial splurge a small fee to use it. You could even have little open days where you work on open source hardware projects and the like just like they seem to. It wouldn't surprise me if Farnell or similar could be persuaded to sponsor it (or at least events it could enable) in some manner. With a group of people you could even try to get a dispenser, enabling you to print paste into boards with no stencil needed.

[Mike]

I think I've only ever used the PCBpool free stencils once for a fairly 'easy' PCB, but there are other services like Eurocircuits and PCBTrain who do variable-size stencils, so being able to use smaller unframed ones can work out a lot cheaper.
As regards pad sizes, you can always pre-shrink the data you send them.

[Gopher]

You could indeed, I was being a little unfair, PCB-Pool is almost certainly the worst possible example.

I was designing in my head how my ideal printer of this type might look, then I found someone had made it already... (well its damn close to what I was envisioning)

http://www.mechatronic-systems.com/products/screenprinters/S1_E.html

I can't quite see how you put the tension in other than by pulling bloody hard but then there is just the one image.

[Mike]

You could indeed, I was being a little unfair, PCB-Pool is almost certainly the worst possible example.

I was designing in my head how my ideal printer of this type might look, then I found someone had made it already... (well its damn close to what I was envisioning)

http://www.mechatronic-systems.com/products/screenprinters/S1_E.html

I can't quite see how you put the tension in other than by pulling bloody hard but then there is just the one image.

Looks like it needs locating holes rather than clamping a plain stencil - user manual is here :

http://www.somersetsolders.com/downloads/Stencil_printer_S1_e.pdf

[Gopher]

So OK for when you buy an "Element" stencil separately or can otherwise specify the holes but less so for the so-called free ones where they ask you nothing. (unless you can somehow sneak them into your Gerber data)

I was thinking of something like that but with a plate that bites the foil -looking at the picture you could probably just about manage that with a little adaptation or maybe some really sticky tape.

I was asking a stencil rep today what pople did with these stencils and he said mostly they stick them down with tape - so it's not just me  , they themselves do do a tiny frame but its for reworking BGA's and the like. Part of the problem must be very few buyers of these things, if you do a lot of prototype runs you may well have picked your SM machine to have paste capability. (£1500-£5000 option on many of the smaller machines)

[Mike]

Another reason why being able to use stencils  with no holes is to combine more than one design on a single stencil order to save cost.
 I was wondering if you could use something like a metal rod or multiple ballbearings in a tapered cavity, such that as you tension it, it clamps the stencil tighter

[Gopher]

Multiple PCB's: With our Tecan foils we can send them back to have extra apertures, which could be a whole new PCB should we so choose. Also with that stencil type and source its a fixed price not an aperture based one so you can have them add as many boards as you want, which could be a lot on a 23"er.

Mechanism: A cleat of sorts, I like it. Trick with such mechanisms being to get the initial grip. I think it should be a frame that can be mounted into any printer rather than a printer with specialist frame permanently fixed.
Tensioning technique on a Tecan frame is quite cunning, foil passes over flat of a semi-circular bar, lock down, rotate bar, done. Now who has the lamp so we can summon a Genie?

[Mike]

Just had a price on that S1 printer:  £1550.
From memory I think that's a similar price to the Eurocircuits one - EC site is down ATM.

Still seems a lot of money for what is basically a hinged frame.

I was wondering about the feasibility of making something based on thick FR4 instead of metal, easily and accurately machined by your local friendly PCB house with low tooling costs.
Loosely hinged with frame-to-base registration with (tapered?) pins, or maybe a full-width  piano hinge with minimal sloppiness.
Magnetic PCB holders (pinned and glued stacks of FR4 with neodymium magnets inside) sticking to a thin steel sheet base glued onto FR4, MDF or ply base board.
For  alignment, between the steel bottom sheet and magnetic holders sits a thin layer of either thin FR4 or nonmagnetic stainless (recycled stencil) , which can be pulled from the sides to move the holders.
This takes the alignment mechanism out of the critical path for registration accuracy - there's enough friction that it won't move unless pulled fairly hard. ISTR reading about someone using a carbon fibre sheet for this purpose.
 
Thoughts?
 

[Gopher]

eC-stencil-mate is 1.495,- Eur according to google cache, isn't it also really quite small?
To me the worst feature of these is the fact still the fact you can't put frames in. I some cases nothing else will do, I know you outsource your bigger jobs, but if (when) your RV dies and you got an Essemtec Pantera for instance you might find you no longer need to - they're 3 times quicker.

I think your idea could work but my first thought is "my word just how crap is that thing you already have?"

maybe a friendly metalworker would be more handy, IMHO the table part and the hinge is the easy bit, its not necessarily precision engineering but someone with a welder could be handy as could a proper well stocked ironmongers for inspirational clamps and fittings. And lets not forget plenty of people don't use a proper printer at all just a desk.
The stencil holding bit however, tricker, tho given a frame can be 1" thick maybe it would be rigid enough cut out of chunky MDF and then bolting devised devious mechanism to it, actually sellotape hold quite well.

This needs pictures

[Mike]

I think your idea could work but my first thought is "my word just how crap is that thing you already have?"

The problem is although it has a row of posts to hold a stencil via holes (Like the S1), the posts do not have any individual compliance to cope with slight inaccuaracies in their position, so they end up distorting the stencil when tensioned. For now I just fix stencils with gaffa tape, but can't get any significant tension
The hinge has about 1mm of lateral play, and the position adjustment system doesn't work very well.

maybe a friendly metalworker would be more handy, IMHO the table part and the hinge is the easy bit, its not necessarily precision engineering but someone with a welder could be handy as could a proper well stocked ironmongers for inspirational clamps and fittings. And lets not forget plenty of people don't use a proper printer at all just a desk.
The stencil holding bit however, tricker, tho given a frame can be 1" thick maybe it would be rigid enough cut out of chunky MDF and then bolting devised devious mechanism to it, actually sellotape hold quite well.

Maybe, but I was thinking beyond my requirement, along the lines of a product/kit or open-source design, as nobody seems to have addressed this area, and the commercial offerings are very expensive.
If you could make a self-assembly kit of FR4 parts as basically a gerber file you send to a PCB house it could be quite an interesting proposition, and I think probably quite doable.

[Gopher]

Ahh, and it tensions with a spring, ambitous.

Budget low level engineering (I think) idea, need a pillar-drill and a tap just for holding the stencil and tension

https://skydrive.live.com/redir.aspx?cid=2c6c3ed0db53f6fd&resid=2C6C3ED0DB53F6FD!107&parid=2C6C3ED0DB53F6FD!105&authkey=!AImi8KevWikAPo8

Idea would be  2 sides would have preset tapped holes to fix two outer rails, then you sellotape your foil to the one fixed and one mobile bit and pull it tight with the knob.

table, hinge, adjustment etc are then anyone's game; and it should fit in a std printer. My printer doesn't have PCB holders or supports for instance, you hold the pcb in place with tape and PCB offcuts.

[phonoplug]

Have either of you seen the Eurocircuits one in the flesh? I have - and its got some major flaws! Firstly, your PCB must be panelised, even if its a large single impression per panel, as the panel border must have locating holes in specific locations with reference to the stencil. Secondly, it has NO lateral adjustment, no really. You can adjust in one axis, but it relies entirely on the locating pins/holes lining up exactly in the other axis. Even if that lines up ok, I don't hold up much hope for say a 0.4mm pitch device thats more than 100mm away from the edge of the locating holes.

Eurocircuits have made it solely for people to use with their PCBs and stencils. And if you aren't doing any close pitch stuff and are only buying from EC (and doing enough quantities to make it worth while buying their printer in the first place), then its going to be absolutely fine. Otherwise you could be spending quite a bit of cash on something thats really not very useful.

Also not sure whether it can work with PCBs other than 1.6mm thick. I quickly lost interest after I found the first two issues!