Again, another feeder design

[mrandt]

The tape then simply goes against a piece of steel wool - just like a brushed motor - making contact with another assembly at any rotation

I see how that works now, thanks. And you hacked together your own "slip ring" - nice

Can I ask you one more favor? Could you post two pictures of the Juki nozzle (connector side) - one with nozzle pushed in (spring compressed) and one in "normal" state? I am curious how these babies work.

when you fully compress the spring than there should be enough pressure to trigger the switch.

Sure, I could - but the point is that triggering the original LitePlacer switch needs to much force IMHO - so I think Karl got a nice solution here.

By reading in OpenPNP group I also found out that Juki Nozzle has another advantage: The Nozzles are held in place (and sealed) by a ball bearing mechanism - similar to the pneumatic couplings you find on air compressors and air-powered tools.

Check out the post here:
https://groups.google.com/forum/#!topic ... 73I8OIXxjo

I believe that is the same nozzle holder Karl has - it is also available from eBay but at a steeper price:
http://www.ebay.com/itm/Customized-SMT- ... SwAYtWOGI0

The nozzle holder seems ideal to build an auto tool changer like Karl already has. Push down the holder onto some kind of U-shaped plate and nozzle will be released - mount new nozzle the other way round. If I understand correctly, Karl achieves this by sliding the holder onto a sloped / angled plastic:
https://vimeo.com/144454866

Great stuff - more reliable than magnetic in my opinion.

If William (from OpenPNP group) is not going to setup a group order, maybe we should do that ourselves or suggest Juha buys a bunch of nozzle holder and remarkets them - would make sense if demand was more than 5 holders.

I would probably attach that holder directly to a NEMA8 hollow shaft motor like this one:
http://www.robotdigg.com/product/43/Nem ... pper+Motor

Attach that motor to the Z-makerslide and you will have a very precise nozzle - potentially elimnating the need to calibrate machine for "needle wobble", enabling auto tool change and reducing push-down force and stress on components.

In turn, I would also need to find a solution for the limit switch, probably along the lines of Karl's contact or using some sort of optical, capacitive or inductive coupling.

Just to be clear: I see that setup as an alternative to the original LitePlacer head. I know it costs more but it also has benefits. So maybe it should come as an option for LitePlacer 2.0

[mrandt]

The tape then simply goes against a piece of steel wool - just like a brushed motor - making contact with another assembly at any rotation. This assembly is pulled to +V through a resistor, as the nozzle (which is grounded) hits something it shorts the connection to ground.

Hey Karl,

I had an idea how you could replace the steel wool contact with something cleaner and more reliable:

Instead of special slip ring contacts, just use a steel ball bearing - not as "ghetto style" but still cheap and effective

There are special bearings which use a conductive graphite based grease as lubricant to enhance conductivity. I have also read that some people opened bearings and exchanged the standard grease for some special conductive grease or mixed carbon dust or graphite into it.

But good news is: Many ball bearings will conduct just fine as they are. I just measured a few 608 bearings I had lying around the workshop and all had very little resistance between inner and outer ring. As you are only transporting very little current, it should work like a charm.

Take a bearing one with an inner bore large enough to fit over a piece of the plastic tubing which carries your vaccum and connect the inside ring to your wire. Fix the outside ring in some plastic mount and connect it to V+ with a resistor - voilà.

I just rembered reading about that somewhere a while ago and finally found the source again:
http://stephanschulz.ca/sts/howto/slipring.html

Cheers
Malte

[JamesMarten]

Many thanks to the creator for putting such an excellent diy p&p out there - having a vision system on board from the off was really pushing the boat out, i'm impressed ^.^

Right, down to business, really keen to see the reel advancer system implemented, for my tinkering, in my move from through the hole components to smt I am likely to get through loading alot of passives & having a machine function in a 'set and forget' manner would make a hell of a difference to sucking the pain out of build sessions.

However, I would also like to make a suggestion for an intermediary option which bridges the halfway gap in convenience at little extra effort - How about having a manual advance mechanism whereby reels are mounted on the back of the machine & fed into retainer grooves in the bed; each time the machine uses up all the components in the accessible area of one strip it stops & asks for that particular strip (and others getting close to the end of their loadout) to be advanced manually (by dragging the strip through the groove until its fully advanced (trimming off excess at the end of the bed manually, or by use of a guillotine blade installed on the exit - as needed)
Notes: Tbh, I tend to try and design my projects to use the same components wherever possible, so most of the time the same reels would be mounted to the machine.
This little hack would merely (?) require a tiny, simple software hack and some manual engineering by those so inclined to implement the feature

[Knas]

I had an idea how you could replace the steel wool contact with something cleaner and more reliable:

Instead of special slip ring contacts, just use a steel ball bearing - not as "ghetto style" but still cheap and effective

I just rembered reading about that somewhere a while ago and finally found the source again:
http://stephanschulz.ca/sts/howto/slipring.html

Oh wow, this is actually a great idea, thanks!! Yeah the steel wool was what i had at hand so

[mrandt]

Hello Karl,

how is your feeder performing? Have you changed any aspects of the design?

I would appreciate if you shared your CAD files so that others - including myself - could give it a try

Thanks and best regards
Malte

[Knas]

So i've been working like crazy on these feeders and i'm ready to start testing for serious tomorrow, if it all works out ill try and get the cad files out next week. So far i have designs for 8/12/16mm tape, paper or plastic, 1 or 2 hole spacing and working on 3 hole spacing rn. There are a lot of things that aren't perfect with this design but i think it's a good start, attaching some teaser photos.

[mrandt]

Karl,

these feeders look awesome - impressive work!

Your teasers immediately triggered my "me too reflex"; can't wait for your CAD files to start building feeders

Once you publish those, I will try to 3D print some of the parts - should be possible at least for the case; maybe not the plates but the middle part. As discussed before, the mechanis probably need to be CNC milled for stiffness and durability. Have you made these from aluminium or steel?

What I would also like to try is make a feeder that has your mechanics for tape advancement but also includes the spool holder and tape cover winder; so I could easily change parts by changing feeders just like on a commercial PnP. I am thinking of using a back plane cut from sheet aluminium; similar to what Protovoltaics designed. But first things first; for a start I would be more than happy if I could build something that works as nicely as your feeders.

By the way - how are you routing the cover tape now? I can't see it in your pictures.

Best regards
Malte

[jarekk]

What is the mechanism for advancing tape ?

[JuKu]

What is the mechanism for advancing tape ?

See the video linked in the very first post of this thread.

[dave]

Do you think I can make them mostly in aluminum on cnc milling machine? Or dose it rely on some flexible plastic parts that need 3D printing.

I dont have a 3d printer

If you want to sell me 20 or so tell me the price. I will spend a small fortune on commercial feeders in the new year otherwise.