Automatic Feeder options

[Marius Liebenberg]

Hi All

I have built two automatic feeders that I think are good choices. I have read comments on some problems that some has experienced. So I would like to ask the members that have used either to give feedback on problems or pros and cons.

I do favor Michael G's design but it is a whole lot more complex and more expensive. I did modify Michael's design to take a SG90R (continuous rotation) servo in place of the EN motor and gearbox.

The other is very simple but I read that the common roller used to collect the cover tapes can cause issues.

[Marek T.]

And?

You should show more details of your modification, and how you have solved 2 or 4mm pitch change.
This metal feeder is not Pham's design?

[Marius Liebenberg]

I am actually asking opinions on the two designs.

I did not know about the 2 or 4mm pitch problems but if you would elaborate then you are helping already.

The metal feeder I got from here https://github.com/szitabalint/SMT_FEEDER_SERVO

To this design I made changes to get rid of the rubber bands and to use springs instead.

There is not a lot to show about my modification. I just changed the motor mount to take a servo. I am also not using the little pcb in the feeder but will feed the wires to a connector that will connect to a female underneath the bar that will carry the feeders. If I go this route then I will use the Arduino Mega with the sensor shield and do some mods to the firmware to still read the switch in order to advance the servo that will pull the cover tape. I can get 24 units per shield. I will publish the new part if anyone is interested.

[Marius Liebenberg]

This is the mod for the servo

I also made changes to let the wires all go inside but I have not printed a new part yet.

[Marius Liebenberg]

Merek

Sorry I gave you wrong information. The link was for another servo driver feeder. I designed the metal feeder to be laser cut and it has a 3d printed base to give the various sizes. I did see the idea somewhere on the web or the forum but I dont remember.

[Marek T.]

I'll look over it when being at the computer. I'm preparing to print some feeders on days and every new project is interesting for me. I was impressed with Darren's design but seems it died (project not Darren I hope mean). Now I'm focused on well documented Michael's project.
Parts like 0805, 1206, sot23 etc usually are packed in tapes with 4mm pitch (tape advancement 4mm required). Parts like some 0603 and probably every 0402 need 2mm advancement (slots for parts are 2mm distanced).

[Marek T.]

Sorry my opinion is the idea is sh*. This motor placed as it is forces very large thickness of the feeder. It limits a lot the number of feeders you can mount on the plate. Total thickness of 8mm tape feeder's should be max 16-18mm including motor.
That's why I liked Darren's design, look how he mounted the motor (also not servo).

[Marius Liebenberg]

For both designs you can adjust for the 2mm pitch. Michael's design might require some code changes but then I am going to make changes anyway. The metal feeder is pushed by the head so the distance can be programmed same as what is required to make a double (8mm) feed for bigger parts.

[Anthony Webb]

I have been printing and using michaels design and it has been flawless for me. Curious how the continuous rotation servo improves on it? It is the first design I have been able to see which fully takes into account end to end the year integration into open pnp, all the way through peeling the tape and presenting the tape in a reliable manner.

There are other feeders on the horizon that I am anxious to try, but for now i like michaels.

Sent from my iPhone

> On Oct 28, 2018, at 8:40 AM, Marek T. <marek.tw...@gmail.com> wrote:
>
> I'll look over it when being at the computer. I'm preparing to print some feeders on days and every new project is interesting for me. I was impressed with Darren's design but seems it died (project not Darren I hope mean). Now I'm focused on well documented Michael's project.
> Parts like 0805, 1206, sot23 etc usually are packed in tapes with 4mm pitch (tape advancement 4mm required). Parts like some 0603 and probably every 0402 need 2mm advancement (slots for parts are 2mm distanced).
>

> --
> 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/c946fc95-17e8-4cef-844f-84a86034a5e8%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Marius Liebenberg]

The motor is mounted at various positions so you can stagger them. Should not make a difference if the stand up or sideways.The feeder stays the same size which at this time is 14mm.

The N20 motor is more than three time the price of the little servo. When you have 64 units to build the price does make a difference.

[Marius Liebenberg]

It only improves on cost in the end.

[Marius Liebenberg]

The motor is mounted at various positions so you can stagger them. Should not make a difference if the stand up or sideways.The feeder stays the same size which at this time is 14mm.

The N20 motor is more than three time the price of the little servo. When you have 64 units to build the price does make a difference.

[Marek T.]

Marius I've told about feeder from the picture you've attached. 14mm is perfect.
Yes I understand why you replaced servo with N20 and agree with you. Test it and tell how it works.
Michael's design looks great but I agree it's a bit too complex maybe. My oppinion is that as something more simple as more solid and stable, of course "simple" is relative.

[Michael G.]

Hi,
the Firmware for my feeder is prepared to be used with 2mm pitch tapes. Just append f2 to feed 2mm per advance cycle. E.g use M600 N7 F2 to advance feeder 7 by 2mm. F16 would advance 16mm in one cycle and send Ok after advance finished, to signal openpnp to go on and pick the part. Maybe you need to tweak the angles to adjust the pick position. The gcode documentation for more details https://docs.mgrl.de/maschine:pickandplace:feeder:0816feeder:mcodes#m620set_feeder_config

[Marek T.]

Do you mean that always when we need other advancement than 2mm we just need to multiply the command?

[Marius Liebenberg]

Hi Michael

That is good news. I have not studied the code yet but just looked at how to add code to handle my servo in place of the N20.

I still have to get there though.

[Michael Gröne]

Yes, see https://docs.mgrl.de/maschine:pickandplace:feeder:0816feeder:mcodes#m600advance

Just append f2 to the command. You can even set the default feeder pitch with M620 N20 F2 and store to eeprom. M600 N20 will do 2mm by default then. Please see you will likely need to adjust the angle for the 2mm pick position with M620 N20 B66. 66 is the angle https://docs.mgrl.de/maschine:pickandplace:feeder:0816feeder:mcodes#m620set_feeder_config

I tested that functionality, but didn't use until today. Let me know if you run into problems

--
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/POR-fCcKI2Q/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/fad1c79a-1101-43b6-bb63-f6af055617b3%40googlegroups.com.

[Marek T.]

@Anthony Webb,

As you have printed many of Michael's printers and I can find you as their experienced users :-), could you tell me something about:

- holding accuracy. How it is with repeatibility of the holding position, it's basic problem at small 0402 parts?

- there is no shutter in these feeders, isn't it? Is there no problem with light parts jumping out from the sockets? (again, the problem usually occurs with 0402)

- how many thousand moves (parts) you made with your oldest feeder? I'm trying to get known about the durability...

- couldn't find anywhere information what the material is recommended. ABS, PLA...?

of course the questions also adressed to Michael :-)

[Anthony Webb]

Hi Marek,

We dont have much need for 0402 in any of our boards.  We mostly use 0603 but can report that it has been flawless.  Perhaps I will give 0402 a try and report back.  We have printed ours out of PLA on a prusa MK3.  I'm all for a more simplistic design, but I've yet to see one as effective.  

In a perfect world I would prefer that the tape holes have positive engagement all the time (ie a sprocket), as opposed to pulling a pin in and out of the holes.  That said, I have yet to have a failure to engage with the hole, so it seems to be getting the job done just fine.

Regards,

Anthony

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/b1e1a207-d359-4ad3-96a0-5c6a67a5b74c%40googlegroups.com.

[Marek T.]

Hi Anthony

Thx for information.

Let me know how about 0402 when you find some time to test it pls.

I would also prefer to have the sprocket for advancement instead of the lever - but I will try how it works on days when get my printer.

[Anthony Webb]

@marek, another interesting one to watch would be Darrens Buddah feeder.  I do think if his feeder and Michaels got together and had a baby it would be an interesting concept.  Last I heard from Darren there was still some work to be done with the Buddah, but I am keep on printing one for sure.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/a82ff638-e675-4452-bf1b-fd6e9788f1b8%40googlegroups.com.

[Marek T.]

Hi Marius,

I looked wrong on your picture and idea of the side motor, mounted the same as Michael's N20.

So you have just used two the same SG90 motors in your modification?

I've tried to find N20 in my country and it's really not easy to find it, SG90 much simplier and cheaper.

[Marek T.]

Yes Anthony, this was very nice project that I've observed too.

I have posted in Buddha thread to ask Daren what's the status of it.

[Marius Liebenberg]

Hi Marek

The one servo is an SG90 and the other is an SG90R. The "R" is for rotational. It does not have to return to 0

deg and the speed and direction is determined by the pulse position. These servos are exactly the same length as the N20.

I cannot get the N20 motor over here either.

[Marek T.]

Understand. What's needed to be modified to use it, only the sprocket or something more. Can you share the files if I'd like to print it?

[Marius Liebenberg]

Marek

The gear and the housing changes. I will post the file for the gear tomorrow when I get to my CAD computer. Cant seem to find it on the printer PC

[Marek T.]

Thx!

[Marius Liebenberg]

I also made minor changes to the main part. Things like slots to let the servo wires run inside and the mounting hole I enlarged to take an M4 Allen Cap screw to bolt onto the 20x40 with a T nut. I also enlarged the elongated slot on the tensioner arm on the very back of the unit as mine snagged with the screw that held the spring. I made the pin that fits through the feeder arm to 2mm so that I can purchase needle rollers that fits there perfectly. The slots that they slide in was left the same for now.

[Marek T.]

Sounds good. When you'll have a time and possibility - pls put your project (or stl) onto some virtual drive. Could be good to print and see your changes. Of course, on the moment the most important for me are the changes related to the motor changing.

[Marius Liebenberg]

I will do so.

Just noticed that the base of version 1.75 has rebuild errors when I import it to Solidworks so the output STL file is also not complete. Maybe Michael can help please?

[Paul Jones]

I didn’t find the small motors to be expensive. I got 30 for $2.36 USD Each.

https://www.aliexpress.com/item/Micro-Speed-Reduction-Gear-Motor-with-Metal-Gearbox-Wheel-DC-6V-30RPM-N20/32675171204.html?spm=a2g0s.9042311.0.0.785f4c4dMn4quo

The SG90 servos I got 20 for $1.10 each

 

Paul.

 

--

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/9c4a1c91-617b-4e61-b9dd-994c026bd43f%40googlegroups.com.

[Marek T.]

Local market 2 days, China 30 days.

[Marius Liebenberg]

That is very good if you can get them. My problem is if I order anything from China it must be couriered or it will take one year if I get it at all. The courier cost will be $50 or more.

The cost of a N20 here is from R200-00 up to R279-00 and the FS90 is R75-00.

No contest.

[Marius Liebenberg]

Marek

I posted the new STL files in this thread.

https://groups.google.com/forum/#!topic/openpnp/0ZMk60x51tw%5B101-125%5D

On Sunday, October 28, 2018 at 3:51:07 PM UTC+2, Marius Liebenberg wrote:

Hi All

I have built two automatic feeders that I think are good choices. I have read comments on some problems that some has experienced. So I would like to ask the members that have used either to give feedback on problems or pros and cons.

I do favor Michael G's design but it is a whole lot more complex and more expensive. I did modify Michael's design to take a SG90R (continuous rotation) servo in place of the EN motor and gearbox.

The other is very simple but I read that the common roller used to collect the cover tapes can cause issues.

[Marek T.]

Hi Marius,

Thanks! I need only wait for my 3D printer so...

Understand it is for 8mm lane. How about the wider tape feeders? Will be needed another stl files from you or the parts making feeder wider are from Michael set?

[Marius Liebenberg]

Marek

Yes we will need modified files for the wider units. I will do them as well once I am happy my mods are all fine.

It should only be the base that changes

[Marek T.]

Clear!

I hope to print it in the next week then I'll tell you at once how I see this.

[Marek T.]

Hi Marius
Have you thought about it to add additional worm sprocket on the rear motor and then mount this motor along the feeder (in it's Y axis)? In this case every motors would be mounted on the same level and no need to change them.
I mean the sprocket looking like this one:
https://www.motionindustries.com/motion3/fsdb/images/item/MARTIN_SPROCKET_GEAR_CO_00366329.jpg

[Marius Liebenberg]

Marek I have not considered that but it is a good idea. I will look into that option. I ill have to design a worm gear like that and print it.  The two existing gears might then also change just to get a little bigger pitch to work with so it will print easier.  It's most definitely a good idea.

[Marek T.]

This way we'd avoid needing of motors position changing and get the feeders flat as they are without actual motor :-).

I must find SolidWorks and try mods at my side... Using freeCad today but it's definitely not dreaming...

btw: have you maybe had occasion to test these feeders with some 0402? My printer came today but must upgrade and tune it first untill start feeders printing.

[Marius Liebenberg]

I am not in a position to test yet. Still busy building heads etc

[Marek T.]

Understand.