PPM Quad Feeders

[Mark Harris]

First set of photos up: http://www.markhphoto.com.au/PPM-Quad-Auto-Feed-8mm-Feeder/

Want to see it bigger? Click one, then edit the URL to change the /A to /O (O for Original). Pictures are 24mpix from a high end macro lens, so you can pixel peep it ;) I shot them off hand so depth of field isn't enormous.

This is an auto feed type feeder, when the beam is broken as the pickup head goes to get the part, the feeder will progress the tape. Looks like it has an Intel 8082 processor and the circuitry to support it. The encoder wheel is 5 slot. Both the encoder wheel and the beam break use just simple IR diodes. I'm pretty sure the pins for input are just power - I'm told by the owner of the feeder that there is no feedback to the machine for these.

The reel holder bolts onto the back of this module, but I don't have it - I'll follow up this email with the machine driven type feeder, that does have the reel holder on it. The peeled cover tape goes onto the black spool that is covered with sticky tape, and is driven by the O ring belt which can slip quite easily so it can deal with the ever increasing diameter of the spooled cover tape.

I have no idea what the red and black buttons are for.

The sprocket for the tape is brass, and I can't see any witness marks to tell how it was manufactured (if it was me doing it, i'd probably just chem etch or waterjet them.)

The tape feeds through between the cover tape spool and the motor, up to the sprocket and then all the way around and back out just under where it entered.

[Mark Harris]

Here's the machine controlled feeder: http://www.markhphoto.com.au/PPM-Quad-8mm-Feeder/

Very much so more primitive than the other feeder.

The output sprocket is plastic, I'd guess ABS based on softness/feel.. its too soft for nylon and too flexible for HDPE. There's no glass fill on any of the plastics.

The tape cover reel is broken on this one, but is the same as the auto feeder in appearance.

The tape path comes from the back and goes out the front. There is a bit of sprung steel holding the tape down behind the sprocket, the position of this is adjustable.

The encoded is again IR diodes, but only on the drive sprocket for this one - the auto feeder has encoders on both the motor and the sprocket. The plastic sprocket on this feeder is identical in appearance to the brass one in the other feeder.

If anyone would like further photos or clarification of details on the feeder, let me know.

[Graeme Bridge]

Excellent pics Mark,

Any chance of dumping some where i can download them? dropbox?

[Jason von Nieda]

Awesome work Mark! Thanks for taking the time to do this and post the images.

Looks like my "worm drive with opto on sprocket" idea is not so unique... :) That second feeder there is very much along the lines of what I want to do with my own design. Very simple and much of it can be 3D printed.

Jason

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/9afb973b-8eef-4d02-9535-afc9ef6a9621%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Graeme Bridge]

I was thinking the same thing Jason

Maybe a group hangout once the nozzle is machined and hash out some ideas on the feeder. I just bought an assortment of feeder parts today form eBay so will have those end of the week and see what i won lol

also found this seller who lists feeder parts

http://goodlucksmt.com/products-detail.php?ProId=219

[Graeme Bridge]

Mark, any chance you could give us some dimensions of the feeders and gears?

[Jason von Nieda]

Hi Graeme,

I'm definitely open to a hangout, but for my part I've already settled on a basic design. I just need to find the time to prototype it. I laid out my thoughts in the "Peter's Feeders" thread: https://groups.google.com/d/msgid/openpnp/bfc1033b-2f2a-4af9-9d44-de5568348c6e%40googlegroups.com?utm_medium=email&utm_source=footer

This is something I've been musing on for quite some time and I've come up with and rejected a number of ideas. This one is the one I've settled on as something that I think can be done cheaply and reliably without a lot of custom machining.

Jason

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/453dfad3-c14a-4330-b478-70d7b9d74dbe%40googlegroups.com.

[Graeme Bridge]

Jason,

Id quite like to have a go at cloning the quad feeder, i might even see if i can find one some place. It looks to be a really nice compact design

Graeme

[Ray Kholodovsky]

You know, I just ordered a 3040 cnc so that certainly puts me in a better position to prototype such a thing.  But I question the necessity of this entire mechanism.  In earlier threads we settled on simpler designs with substantially fewer gears and perhaps a separate tiny dc gear motor for taking up the empty film.  

I really enjoyed the disassembly hangout and CAD, so I'm totally up for doing that again.  Just don't know how feasible the result will be.  

[Graeme Bridge]

I like to get my hands on something to see how it works, I'm also curious as to why so many gears 

[Anthony Webb]

I really appreciate everyones efforts here, I cant think of anything that will close the loop on a "real pnp experience" like having great feeders, up looking camera support, and tool changes.  With those items there isnt any reason our DIY machine cant compete very well with many of the "commercial" solutions on the market. 

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/15cd431e-77dd-4dcc-841a-84795a5c654f%40googlegroups.com.

[Graeme Bridge]

There must be a reason there are that many gears, manufacturers don't put things in for no reason

[Mark Harris]

The reason for so many gears is because these motors run FAST, and the dual head pnp needs to cycle a part into position in probably 200-500ms, which needs quite a bit of torque.

For those who are looking at prototyping - Solarbotics (www.solarbotics.com), who I stole the feeders from have a whole bunch of Epilog and Trotec laser cutters which at this time of year are not in full usage. They will do custom jobs and likely faster/better/cheaper than 90% of online laser cutting places. Their lasers are upto 80W CO2 so no metals - an 80W Trotec CO2 is about the same as a 150W RECI chinese CO2 in power, but a far far finer spot size (and therefore kerf).

Furthermore, I have a very solid CNC router that can CNC mill aluminium and all plastics - there are pics on the same page as the feeders. At work I have access to very large CNC mills, lathes, waterjets (water jet cut sprockets?) but access to these is not as easy as doing things at home.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/24310fc6-7492-48dc-9377-0937fc453f01%40googlegroups.com.

[Mark Harris]

Where are you located Graeme, I might be able to get more of the obsolete design of these feeders off Solarbotics - they might sell them cheap (I'm just borrowing these). You could email Dave about buying some, da...@solarbotics.com - he's the head of their engineering team :)

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/15cd431e-77dd-4dcc-841a-84795a5c654f%40googlegroups.com.

[Mark Harris]

Which dimensions are you looking for specifically? I'd prefer not to reverse engineer the whole thing but I can take photos of the feeder with calipers on it :)

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/24420617-b96e-4f7b-9340-d96a8c564893%40googlegroups.com.

[Graeme Bridge]

Mark, that was my thought the torque being the important factor here.

[Graeme Bridge]

Im in the UK Mark, i have a couple of cheap feeders I'm watching on eBay.com ones missing parts for $20 the other is complete for $60

[Anthony Webb]

Perhaps I am wrong, but I think that n20 motor might have all the same gear reduction happening, but it is more compact and just part of the motor design, as opposed to doing it with external gears?  I'd really like to verify that the n20 with a simple advancement gear WONT work before going through the added complexity/cost that some of these feeders have.

[Jason von Nieda]

Anthony,

I agree completely. I haven't seen any compelling argument yet that the N20 won't work perfectly. That's where I will be focusing my efforts.

Jason

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/0d6bccca-4e4a-4920-8fa0-a7265f9111a4%40googlegroups.com.

[Mark Harris]

Keep in mind, this gear chain probably cost PPM around 1-2 cents per feeder, injection moulded ABS/Nylon gears are uber cheap. I wouldn't be opposed to doing something like this with urethane cast gears - a set of gears could be made every 10 minutes per mould and cost 20-30cents. I don't know that it needs to be as complex a chain as the PPM ones ;) Plus, it looks complex and like its worth paying lots of money for :P

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CA%2BQw0jyoDkEibYiApR-19oTZehF7X7Hi0HLDkGYg518vhrCXuA%40mail.gmail.com.

[Graeme Bridge]

Mark i wouldn't be surprised if we could get pretty close to the final ratio with off the shelf gears available on eBay etc

[Mark Harris]

I wouldn't be surprised either, but ebay gears add up fast in cost.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/3192365a-1946-42e8-b2af-6104f864f844%40googlegroups.com.

[Michael Anton]

I agree as well.  I think the N20 motors could work very well, and they are available in many different gear ratios, so you can balance the required torque vs feeder speed.  A few small gears from Aliexpress, or ebay, to handle the other issues, and you are done.  Gears from Aliexpress are pretty cheap, especially once you get to the 100+ quantity level, which is certainly where we would be at with feeders.  Overall, I suspect the gears will cost far less than the motor, and that sounds reasonable to me.  I'd like to keep parts cost below say $50, so that we could afford lots of them.

Mike

[FredG]

Maybe the body of the feeder can be cast in urethane. Can you give us a link to urethane for casting?

[Mark Harris]

I'd probably stick with milled aluminium for the body - it should be able to get designed with a large endmill in mind as there should be no small internal radius/narrow pocket features other than the tape track (and even that can probably be designed in such a way a 6-12mm endmill can cut it). This means that a body should be able to get mashed out in a few minutes which keeps costs way down - and aluminium is cheap as well as very sturdy.

Smooth-on (http://www.smooth-on.com/) are a pretty good way to get started with casting - they have kits for silicones to make moulds and urethanes to go into it. They have a very diverse range of mixes from things that are super high temp capable (relatively speaking, for plastics) to very hard, flexible, or replicating ABS properties - its also available pretty much all over the world. Once you've gotten into it and figured out what you are doing, your local industrial/plastics supplier can probably put you on to something cheaper.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/78aeaebf-faa5-4008-8451-0a5d87dd2e1b%40googlegroups.com.

[Graeme Bridge]

Hey Mark, I snagged a feeder last night from the USA for $20 lol its got no pcb but I'm fine as i don't want that anyway.

I agree milled aluminium, part like that should be no more than say $15-$20 machined although I'm curious about printing it as a uber cheap option 

[Graeme Bridge]

Feeder has arrived will organise a hangout to tear it down

[rayk...@gmail.com]

Cool, let's find a time. I think weekend works best for everyone. 

Ray

You received this message because you are subscribed to a topic in the Google Groups "OpenPnP" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/openpnp/Md61mbZPMco/unsubscribe.
To unsubscribe from this group and all its topics, send an email to openpnp+u...@googlegroups.com.

To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/be7cc99f-7d22-4bc8-9ed5-20f2c64e5bcf%40googlegroups.com.

[Jason von Nieda]

Hi folks,

I am out of town until Tuesday, but please go ahead without me. Will look forward to watching the video :)

Jason

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/B50E4546-FA87-4244-A31F-3FEB14F740FC%40gmail.com.

[Graeme Bridge]

No rush Jason i need to figure things out here. probably won't be this weekend but perhaps next

[Ray Kholodovsky]

In an ironic twist I'm busy hand assembling PCB's anyways :)