The gear version of 0816 feeder is completed

[bing luo]

https://www.youtube.com/@bingluo8417， Watch   the  videos  here.

      Please tell me if you have any suggestions for improvement after watching the video. At present, it works well and its reliability needs to be verified.

https://github.com/microsmt/Microsmt-PNP-hardware/blob/main/poorest%20man%20%20%20feeder.zip

Drawings and BOM can be downloaded here.

According to different situations, its cost is between $10-15.

      

[bing luo]

It has two advantages.

The first is the use of gears, which can accurately advance 0402 elements. 

Second, the reciprocating sliding block presses the first  element , so there is no need to worry about the component jumping out.   All industrial automatic feeders have this function. Electric feeders(as  simense)   use electromagnets  to complete this action。

[Jon Raymond]

Neat looking feeder! Where can the metal parts be purchased?

[Mike Menci]

The cover tape motor could be with 90 deg output shaft in order to free space for next feeder. 

One of these: https://www.google.com/search?q=n20+90+deg+6V+motor&rlz=1CDGOYI_enDO1044DO1044&hl=en-US&prmd=isnv&sxsrf=AJOqlzVWvdzFnhCRsddqUkMwWJFG4qjIOA:1678462608614&source=lnms&tbm=isch&sa=X&ved=2ahUKEwixj7i82NH9AhUubfEDHdvyC7YQ_AUoAXoECAIQAQ&biw=320&bih=524&dpr=3

Sent from my iPhone

On 10 Mar 2023, at 16:16, Jon Raymond <dirtye...@gmail.com> wrote:

Neat looking feeder! Where can the metal parts be purchased?

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/74786c48-c213-478d-b134-e1563908e820n%40googlegroups.com.

[jbasia]

What comes to my mind:

• the tape roller motor is quite wide - so you need quite a gap between feeder
• the metal parts should increase reliability a lot - but difficult to do for the DIY people (flat parts are easy, but 1mm 304 is difficult to bend - accurately)

[bing luo]

the tape roller motor   You can stagger like 0816 without increasing the thickness。

The metal parts  must be purchased. We can't do it ourselves. These accessories look simple. In fact, complex processing was carried out. They are first stamped(Not  bend） and then polished. The stainless steel plate that prevents the gear from turning upside down needs to be hardened, otherwise it will wear out in a few days.  

However, it is not difficult to get them, because all brands of pneumatic feeders have the same working principle, only the appearance is different.

[Jim Young]

Does this have the same mounting profile as the original 0816 feeder? I would like to try this feeder out, but I want to make sure it will fit with my existing 0816 feeder array.

[M. Mencinger]

Jim - do you mean the hight of feeder and picking up location of nozzle - if this is in the same location as 0816 feeder is ? 

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/65149cc7-6852-4c95-b379-fbe53a2a2cf6n%40googlegroups.com.

--

Mike

"--

 **************************************************************************************************************

This e-mail may be confidential and it may also be legally privileged. If you are not the addressee you may not copy, forward, disclose or use any part of it. If you have received this message in error, please delete it and all copies from your system and notify the sender immediately by return e-mail. Internet communications cannot be guaranteed to be timely, secure, error or virus-free. The sender does not accept liability for any errors or omissions. 

 SAVE THE NATURE - SAVE PAPER - THINK BEFORE YOU PRINT!  "

[bing luo]

It looks like this now, all the metal accessories are the same as the old version, only  the shell of the 3D printed part has changed.  

It uses the same controller and code as the 0816 feeder, but it is obvious that its installation requirements are different  of  0816  feeder.

[Jim Young]

I need to know if this feeder will fit on the mounting rails of my existing setup. Here's a picture of my current mounting setup.

[bing luo]

It can be installed on the 2020 aluminum profile marked by the red arrow, but unfortunately, the 2020 aluminum profile marked by the yellow arrow will hinder it

[bing luo]

It can be installed on the 2020 aluminum profile marked by the red arrow, but unfortunately, the 2020 aluminum profile marked by the yellow arrow will hinder it

在2023年6月27日星期二 UTC+8 23:40:24<infinite...@gmail.com> 写道：

[Mike Menci]

Yes beck beam needs to be removed and front lowered for apx. 26 mm (1")  to have picking height the same with 0816 feeder see here: 

Mike

[Jim Young]

I can remove the back support beam, as I'm not attaching the existing 0816 feeder to it. I think I will still be okay with the pick height being 26mm higher. I have enough clearance for that. Is the pick location basically the same as the original 0816? It's hard to tell from the illustrations.

[Mike Menci]

It is abt 28 mm closer see attached top view

[Mike Menci]

Not  28mm !!     sorry   - diagonal view 

  abt 10mm  in top view 

[Jim Young]

Why are there different bodies for 4mm and 2mm part spacing? Is it because the internal metal parts are different?

[Mike Menci]

The sprocket has 2mm advancing  and metal cover is different, shutter is different,.... you need to follow 8x2 and 8x4 stl and metal parts guide when assembling parts to basic 3D printed body (stl files) - yes there are differences... 

Mike  

[Jim Young]

Any recommended settings for printing the body? It looks like it will need lots of supports.

[bing luo]

[bing luo]

在2023年6月28日星期三 UTC+8 05:19:37<infinite...@gmail.com> 写道：

[Jim Young]

I ordered the kit for this feeder on eBay on 6/27, but I don't see any activity on the order. How long does it normally take to ship?

[Jonathan Oxer]

I ordered 2 of the kits which were shipped 9 days after the order was placed, and then the actual shipping time to Australia was about 8 days. Total 17 days elapsed from placing the order to receiving it.

Jon

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/eab0af5f-661a-43b6-b009-17438c67a774n%40googlegroups.com.

[Thorsten.R]

I bought 20 pieces at Ebay and they were delivered quickly to DE.  Less than 10 days.

Thorsten

[Jim Young]

I received notification that my parts will ship tomorrow. I'm going to build this one feeder and if it works out well I'll be ordering more. 

[Jonathan Oxer]

Just got my first PMF running on a Kayo-A4, using a 20x20 section attached to the top of the Yamaha CL feeder manifold with 3D printed brackets.

What I *really* want is the RapidFeeder that Deltaprintr has had in development, but there's been no word on that recently so I don't know when/if they'll be released.

The pick location of the PMF sits a bit higher than the CL feeders when it's on top of the CL manifold, but there's enough Z travel on the Kayo that it's not a problem. Video of a test feed and pick:

https://www.dropbox.com/s/67hscstkgkwmpod/2023-07-03%2015.46.25.mp4?dl=0

Jon

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/5d00e3af-b174-4bb2-94b9-2c1b97595d6fn%40googlegroups.com.

[Jonathan Oxer]

Just run a test job consisting of a few caps in a row. The tape peeler motor seems a bit slow and the machine often pauses while it waits for the controller to signify that the feeder is ready, but on this particular run the feeder just barely managed to keep up with the machine. If you watch carefully you can see that the tape peeler finishes almost simultaneously with the part being placed and the machine wanting to feed again. If the tape peeler is even a few milliseconds slower it's not ready when the machine wants the next part, so the MGRL controller returns an error and OpenPnP waits and then tries again so there's a big pause on each part.

If the peeler was faster, or it was taking parts from different feeders, or the MGRL firmware was modified to always respond that it's ok, it would work more smoothly.

Part of the problem is likely that I'm running the feeder at 5V instead of 6V, so the motor could be running slightly too slow.

Reported rate was 1001cph using the single nozzle.

I only ordered 2 of the kits to see what they're like, but I'm going to order more now that I can see they're functioning well.

https://www.youtube.com/watch?v=dnfv0cQlZVk

Jon

[Mike Menci]

Hi Jon, 

I have purchased as well one of the feeders and I see that in compare with 0618 mgrl feeder the worm has very fine pitch ( I think its module 0.3 or 0.4) it should be at least double Module 0.8 I thinks, this would mean that  collecting cover tape would be 80-100% faster with the same motor & firmware... 

What do you think ?

Mike

[Jonathan Oxer]

Hi Mike,

Definitely, there's scope to make the tape puller much faster. Increasing the pitch may also increase friction on the 3D-printed gear though: or perhaps it would decrease it? I don't know!

The motor is certainly strong enough though so I would change to a lower gearbox ratio and leave the worm as-is. I have a few of those motors around so if I have one with a suitable ratio I'll experiment.

Cheers

Jon

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/1ce4c260-1a34-45a5-96a4-88ee27dca5fan%40googlegroups.com.

[Thorsten.R]

Hi Jon,

See my video with SebG3D Puller without worm fine pitch , so can also super print itself.

Video

This certainly runs much faster.

Thorsten

[Jonathan Oxer]

Great tip! I printed a SebG3D puller a few years ago to try on the original MGRL0816, but never installed it.

The puller mounting method is different on the PFM: did you edit the design?

Jon

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/b57533a5-15d1-4625-9415-650741211775n%40googlegroups.com.

[Mike Menci]

I modified those files a bit for needle bearings  and torsion spring - see 

here; https://grabcad.com/library/cover-tape-drive-0816-my-way-1

[Thorsten.R]

Hi Jon,

Yes, a small flange drawn on it to screw the side. Otherwise original with ball bearing.

But since I'm a bad 3D drawer, you'll have to do it yourself :)

[bing luo]

Jon

      Your machine is very good and fast, so you may find it too slow. Currently, I am using a worm with a modulus of 0.5 and an N30 reduction motor with a speed of 200 rps. For machines like PNP3 or   lower cost 3D printing machines, its speed matches.  

[bing luo]

Hi   Mike

       The modulus of the gear and worm in my drawing is 0.5. You can test it with a modulus of 0.8 to see if the torque is sufficient. If there is no problem，The speed will increase a lot , I can customize a worm with a modulus of 0.8. Perhaps a worm with a modulus of 0.8 can be printed using a 3D printer. I am unable to print a worm with a 0.5 modulus using a 3D printer, and I don't know why.

[bing luo]

Hi,  Thorsten

     Your feeder looks great. If you have time, you can try printing my latest drawings by removing the front tension spring and adding a compression spring at the rear to reduce the resistance of the sprocket, which will be more stable.

[Mike Menci]

Torsen 

Your worm has double start  - I just updated this stl file in my GrabCad

Mike

[Jim Young]

Will this worm gear design work with the 60rpm motor used in the original 0816 design?

[M. Mencinger]

We have to test this. 

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/ec7b8901-be42-45dc-a74d-3be0328bc364n%40googlegroups.com.

--

Mike

"--

 **************************************************************************************************************

This e-mail may be confidential and it may also be legally privileged. If you are not the addressee you may not copy, forward, disclose or use any part of it. If you have received this message in error, please delete it and all copies from your system and notify the sender immediately by return e-mail. Internet communications cannot be guaranteed to be timely, secure, error or virus-free. The sender does not accept liability for any errors or omissions. 

 SAVE THE NATURE - SAVE PAPER - THINK BEFORE YOU PRINT!  "

[bing luo]

  
Thorsten            

         The N30 I am using has a speed of 200 rpm and its high speed is to solve the problem of gears rotating too slowly. If I use a speed of 150 rpm, it will not keep up with the speed of PNP.  But when the speed is high, the torque is low, so the load on the N30 is high.    By replacing it with a large gear like yours, the motor can be replaced with a 100 rpm one,  increasing the torque while also reducing the current. Printing  turbine can also reduce costs.  

在2023年7月3日星期一 UTC+8 19:45:01<Thorsten.R> 写道：

[bing luo]

The reason why I didn't use the N20 motor is because I burned several N20 200 rpm motors，Its torque is not high enough , which were pulling very hard. I am printing your drawings and testing them with an N20 60rpm and an N20 30rpm motor. 

[Mike Menci]

It seams that worm has 3 x winding see pictures and file enclosed ( from master folder)

[Thorsten.R]

Hi,

I used the supplied N30 at 200 rpm, you can select that in the puller scad model which motor 100/150/200 you use.

Thorsten

[Thorsten.R]

With these pullers, the current consumption is < 40mA, while the original pulls much more and sometimes jumps to 100mA.

[tianl...@gmail.com]

       That's because the traction force of the front stretching spring is too strong, which is the main reason for the increase in current. If you place it behind, its current will not exceed 40 mA.

        Under this premise, keeping the servo rotation angle as small as possible will further reduce the current and vibration.   

         The larger the angle of servo rotation, the greater the vibration.

---Original---

From: "'Thorsten.R' via OpenPnP"<ope...@googlegroups.com>

Date: Tue, Jul 4, 2023 20:07 PM

To: "OpenPnP"<ope...@googlegroups.com>;

Subject: Re: [OpenPnP] Re: The gear version of 0816 feeder is completed

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/ad2d224c-89f7-4f10-a00f-87742b78d2d1n%40googlegroups.com.

[Thorsten.R]

I no longer have the feeders on the front.
It is due to the puller without ball bearings and the Worm only one-sided guided, which can never run smoothly. This will always stand slightly oblique due to the pressure.
The Worm belongs above also held, in my opinion.
But everyone can build it as he wants, my variant is just a suggestion :)

[Thorsten.R]

Sorry. The spring in front I no longer have it.........

[Jim Young]

I received my PMF kit this week and started assembling it. I've come to a blocking point where I can't figure out what to do with the spring that is attached to the metal cover. The instructions don't mention it, none of the illustrations show it and it with it attached it is impossible to place the cover on the plastic body. Is this come artifact from a different assembly? Should I just remove it?

[Jonathan Oxer]

There's no clearance for it in the 3D printed body, so I assumed it was left in place from the way these are assembled onto a metal chassis.

I removed them, but it's surprisingly hard!

Jon

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/dee5fa1e-4844-4e51-ad83-e035f098b651n%40googlegroups.com.

[Jim Young]

Thanks Jon,

A good pair of side cutters made quick business of it. On to the next problem!

[Blowtorch]

I have ordered some of these parts but not yet heard anything.  I thought I would print some housings in the meantime.  Has anyone here optimised the 3D printed body yet?  It prints out OK as is, with supports, but removing the supports is a painful and bloody business.  It also takes long to print.  It strikes me that:

the amount of plastic could be reduced, for example the very front tape guide is 2.7mm, the top of the tape guide is 3mm thick, and is not structural, so both could be made a lot small, using less plastic and therefore faster to print.   

To the left and right of the servo, on the bottom, there are plates that cover wiring.  I would rather print that separately and glue it on to retain rigidity, but avoid printing a bunch of support material.

Ditto the input tape guide.

how do you put in the screws that hold the servo?  Other plastic blocks access by screwdriver.

Bear in mind I have not built one of these yet...

[bing luo]

I have released a very complete teaching video  today .   after watching it, you can correctly assemble the poor man feeder.   

https://youtu.be/mWer4ytN0Tg

Of course, you can also modify it according to your own ideas, and you can also search for cheaper accessory sellers,   don't need to buy it from me.

By the way, don't buy it  on eBay  because eBay   charges a high transaction fee and I have to raise the price. If you need it, please purchase it on my website or Aliepress.

[Thorsten.R]

I print the case without any support so in 1 hour with 0.2mm layer height. :)
The picture is still a predecessor housing without the new pressure spring.

[Jim Young]

The one screw for attaching the servo is rather difficult to get to. Using a driver with a very narrow  shaft, it is just barely possible to screw  down.

I looked at the Aliexpress shop and shipping charges for 23 units is $130, a third of the product cost.

I'm not sure how one prints this without support material. I use the Organic support option in Prusa Slicer, which makes  supports much easier to remove.

[M. Mencinger]

Jim 

Send Luo  PM here - he can review shipping depending on where you want it to ship to. 

See the video - second screw from servo truth hole where the spring comes later.. 

Lay it down flat - only few supports are needed - I printed it out without supports. 

Mike

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/5ce4f900-b8c1-47b2-99c0-407a494d348en%40googlegroups.com.

[Mike Menci]

See Video min.: 8.45 sec : https://www.youtube.com/watch?v=mWer4ytN0Tg 

Mike

[bing luo]

You don't need to purchase on AliExpress because the shipping cost includes many remote countries. My website can send to Europe and North America, and their shipping cost is very low.

The servo screw you mentioned that cannot be installed. I have enlarged the hole today so that the screwdriver can pass through.  

You can use a hot wire to expand the hole on the shell that you have already printed

[Thorsten.R]

I have removed these huge overhangs, since not useful then you only need to support at the guide.

[bing luo]

I have reset the shipping cost for Aliexpress,  take a look again.  

在2023年7月13日星期四 UTC+8 22:16:39<infinite...@gmail.com> 写道：

[Jim Young]

Thorsten, can you post STL files for this mod?

[Thorsten.R]

However, I have not yet drawn in the new spring at the back.

[Thorsten.R]

Although you don't necessarily need this spring. A light rubber band in the front works as well if everything is smooth.?

[bing luo]

No, that compression spring is very important. It replaces the function of the spring removed from Yamaha cover. If you use rubber bands or tension springs in front of the cover, it is difficult to adjust to the appropriate pressure.  

[Thorsten.R]

Hi Luo,
how you mean :)

[bing luo]

          What I mean is that the compression spring at  the red arrow can make the pressure of the cover most suitable. Don't move it to the front.

It  replaces the function of the spring at the green arrow .   

   

[Thorsten.R]

But I know if everything goes smoothly, the arm only needs to be held down very slightly, that's how it is in the original with the small spring, which also exerts very little pressure.

But at the back it is of course more elegant with the spring.

In your videos all the metal parts wobble, there is simply too much play in the holes and that's why the SMT chips also jump during the feed. My experience.

[Thorsten.R]

Mike noticed the same thing.

[bing luo]

Yes, my  3D printer is too old, it causes too much clearance between the shaft and shell. 

[Blowtorch]

Thorsten what size nozzle & speed are you printing at?  

[Thorsten.R]

Hi,

0,4mm nozzle and 250-300mm/s

[Blowtorch]

That is fast!

[Thorsten.R]

Yes, Bambu Lab P1P 🏎️ 😎

[Jim Young]

I installed the compression spring, indicated in red, and it seems like it is too strong. It makes it difficult to raise the cover to thread a tape through (I'm assuming the cover needs to be raised to thread a tape).

[Thorsten.R]

You don't have to lift the cover, if it's printed well you can insert the tape until the first tooth.

After that, it's automatically withdrawn, that's how it is with me.

[Mike Menci]

The spring is needed to hold down & beck from cover tape pulling and to press down the advancing tape onto sprocket teeth during advancing of tape. 

Mike

[Thorsten.R]

exactly

[Blowtorch]

My idea - happy to glue on covers after printing.

[Mike Menci]

Lau made a choice with Vertical spring at the end - There are other options as well  see enclosed
1) Spring at rotation point 

2) Eccentric Lock at the beck 

3)..... 

[Mike Menci]

Your STL:    8-4mm  frame-0626 JN trim v3 (1).stl

To glue the cover would not be a problem - but lower wall is not thick to hold all together and front part would break before you start putting in the tape. From my perspective this would not work long therm and the cost of material for 3D print is very low in compere to reprinting all and experimenting where there is no need...... 

[Jim Young]

Okay, I understand the need for the spring. But, still the compression spring supplied in my kit is way stronger than it needs to be, IMHO. Looking at the assembly video, after the spring is installed the cover can still be easily raised by hand. Not so with the spring I received in my kit. 

[Mike Menci]

I follow Lou closely and I know this change was few days or max a week earlier - this was likely after you have received your PMF parts.

What you could do is shorten the received front spring and extend it accordingly. 

I am using mostly from Bauhaus this  200 pcs Spring Assortment Kit Set Bauhaus 21995444
Mike

[Thorsten.R]

or Amazone

https://www.amazon.de/dp/B009T1OTBM?psc=1&ref=ppx_yo2ov_dt_b_product_details

[Mike Menci]

Correct - I think this will do the job: 

[Thorsten.R]

Yes, the pressure on them is very low. He keeps his arm straight down, but that's enough.

[Jim Young]

Okay, that is good to know. I also have an assortment of springs left over from building a set of original design 0816 feeder and I used one from that set that appeared more appropriate in strength.

Another question, why is there so much slop between the trigger lever and the shaft it rotates on? I'm tempted to put a washer over the shaft to prevent the lever from accidently coming off.

[Thorsten.R]

This is due to Luo's 3D printer, because he drew everything a little larger. There is up to 1mm of play, so everything wobbles. You have to adjust his drawing so that you have 0.1-0.2mm of play, this is normal on any 3D printer, the more cleanly the tape runs.

[Jim Young]

I see. I also just now discovered that the N30 motor housing is way too tight of a fit. I'm going to have to make some adjustments  there. It would be nice if the 3D printing was parameterized.

[Thorsten.R]

You should take Luo's 3D drawing as a good template, but don't print it directly, there's room for improvement.

[Mike Menci]

From my point of view the "  the trigger lever and the shaft it rotates on" can have a larger gap (0.5mm or so) to function better (less friction !) - but a washer and screw are needed to prevent lever from falling down :-) 

[Mike Menci]

Thorsten - here the compare in size of springs attached - for info...  softer - shorter and about the same OD.

(I can not test this - I have no version with spring pocket printed out)

[Thorsten.R]

HI Mike, the ones on the left are definitely steel wire(black) and much harder ?

[M. Mencinger]

Correct ! 

Mike

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/b91aff43-3f3d-4e3b-9037-8c136745ae3fn%40googlegroups.com.

[Thorsten.R]

I just print the original with the pressure spring at the back to test it.  1:27 min but it's late here.

[Thorsten.R]

Some Bambu advertising :)

Video

[bing luo]

This is a 0.8 * 6 * 25mm compression spring, and I think its strength is appropriate because the pressure it applies to the sprocket does not increase the servo current. And the tape can be easily pulled forward by hand. If you use a spring that is too soft, the cover may loosen the tape during   working.

If you want it to become soft, use scissors to cut off 5mm and make it a 0.8 * 6 * 20 spring,  Then stretch it to 25mm, but I don't think it's good.

[bing luo]

The spring you received is the same as the one I used   (0.8*6*25mm  spring steel )  . When you press it in, it won't be as hard as it looks,.

[bing luo]

You can elongate the 0.8 * 6 * 30 extension spring and  cut  25mm to make a compression spring.   

[Jim Young]

Testing out my first build of the PMF. I ended up using the stronger spring that came with the kit on the rear of the cover. A weaker spring was not enough to keep the cover from moving out of place during a feed. So far the feeds are very consistent, with the part always in the same location and not jumping out the tape. 

[Mike Menci]

Nice - were you able to place two feeders next to each other ?
I get this with 15.5 minimum distance: see enclosed. 

What do you get ?