BambooFeeders getting stuck

[Nick Werhle]

Hey everyone,

I have 25x bamboofeeder AS feeders connected to OpenPNP and I am having an issue I cant figure out. Most of the time, the feeders move forward 4mm to the next part in the reel no problem, but sometimes it only moves the reel forward 2mm, and comes up with a "cant find sprocket holes" error. 

The issue would not happen if it was moving evenly each time. I am not sure if this is an issue with my physical feeders or the OpenPNP software. I have it set to 4mm on each feeder.

Does anyone else use Bamboofeeders and know how to make them feed more reliabily? If not is there a better choice of feeders I should be purchasing? 

Thanks,

Nick

[Kevin Bralten]

The bamboofeeder AS set are far from what I'd describe as "reliable" but they're also literally 1/10 the price of the cheapest "reliable" option I know of. So we'll aim for good.

There's usually three issues and you can figure out which it is
 - The feeder doesn't move back far enough to catch the rear hole, there's a command to change the servo angle for this (unlikely with 2mm advance)

 - The feeder catches the rear hole but skips forward (unlikely with 2mm advance), you need to use a stronger spring, print the "advance arm with dogs" with a finer profile to get sharper teeth or just file the teeth sharp.
 - The feeder catches the rear hole and stays engaged but can't move the tape all the way forward (likely), you either need to extend the delay for the servo advance (there's some command you can send to adjust this) or you need to loosen the tension screw on the tape guide (usually the thing).

If you're really really invested in this, I've heard swapping to name brand metal-geared servos is a huge upgrade for them. But just those tweaks (mostly attention to the feed dogs and tension screw) made mine reliable enough for like 100 feeds without error. They still sometimes skip or more commonly fail to peel.

--
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 visit https://groups.google.com/d/msgid/openpnp/903accbb-5472-43db-89fc-3d4fe266b728n%40googlegroups.com.

[Jarosław Karwik]

Next ones - if you think about price are these:

https://www.microsmt.com.cn/blogs/microsmt-pnpv3-bing-feeder-system/microsmt-pnpv3-bing-feeder-system

Very good ones and very reasonable price.

[Nick Werhle]

I am looking to use this machine as a tool not mess around with feeders all day… maybe I will upgrade to those feeders you mentioned. It looks like I would need a different PCB bar to mount them to, is there any closed loop feeders compatible with the pandaplacer without changing rails and things?

[Koichi Kurahashi]

I’m very interested in this product. 

I’m currently using an Autofeeder mounted on a PSE-3000 rail with electrodes on the top surface. Would it be possible to use this product with some modifications?

2026年8月21日金曜日 23:04:35 UTC+9 jarosla...@gmail.com:

[Jarosław Karwik]

Your are unlikely to setup different feeder system without at least slight modifications.

In case of microsmt feeders you need to attach 20x20 aluminium profile - which is usually quite simple.

Please note that these feeders need a lot of printed parts and assembly - they are sold as kits, not ready made parts. You cna buy controller and cabling for them as weel.

I wanted ready made feeders - so I sold my microsmt kits (after testing one feeeder) and went for Afarco - ( https://afarco.com/products/af-feeder )

These require 40x20 proflie for their base plates. I am now about to test them. These are also reasonably priced.

[Koichi Kurahashi]

Thanks for the information. 

The Afarco feeders look very nice. I’m completely satisfied with the performance of the feeders I’m currently using, but I’m looking for options that could make setup a little less cumbersome.

2026年8月22日土曜日 14:05:35 UTC+9 jarosla...@gmail.com:

[Brian Piccioni]

I am hardly an expert but, for what it is worth I do a sort of natural selection on my Bamboofeeders. When one is problematic I replace it and later take it apart so maybe the issue is sometimes the smoothness of the feed parts. They seem to get more reliable the more you use them. I found the servo 'horn' and starting angle of the servo is important. I found (or designed I don't remember) a more reliable servo horn. Little things like the tightness of screws matter. 

Of course I only assemble a few boards at a time so your mileage may vary. 

I must admit that the feeder issues have encouraged me to consider other options even though the PandaPlacer was a great introduction to OpenPnP

[bing l]

Its operating principle is identical to that of the 0816feeder, and the controller is also compatible. It is only a miniaturized version of the 0816feeder with one peel‑motor（0.7$） removed, and it does not fix the inherent flaws of the 0816feeder. For this reason, I think it should be named 0816 Min feeder to prevent new OpenPnP users from being misled into selecting it.    https://docs.mgrl.de/maschine:pickandplace:feeder:0816feeder .     

This is a solution we abandoned many years ago. Any newly‑designed sprocket feeder boasts far superior stability compared to it.  Take Lumen as an example — it’s an open‑source product available at https://www.opulo.io/products/feeders. If you think it’s pricey, you can build it yourself.      Ours is good as well: https://www.microsmt.com.cn/blogs/microsmt-pnpv3-bing-feeder-system/microsmt-pnpv3-bing-feeder-system

We are continuing to make improvements. Whinis is developing a version with RS‑485 bus communication. FailSafe is working on a Wi‑Fi‑capable version using the ESP32‑C3. It seems they are nearing delivery. 

[bing l]

An all‑aluminum feeder is certainly the best choice if you have a sufficient budget, as it suffers virtually no wear.

[bing l]

 but I’m looking for options that could make setup a little less cumbersome.  

Could you explain what "cumbersome" means?      Whinis is also developing a similar 485 communication feeder. Currently, the focus is on making the feeder more flexible.

[Koichi Kurahashi]

By "cumbersome" I mostly meant two things:

1. Loading the cover tape — the take-up reel is pretty far from the peel point, so I either peel back the cover tape in advance and remove (or set aside) the parts in that section, or splice on a leader tape before threading. I get that's just how the mechanism works, but if there were a feed-button mode that winds up only the cover tape, splicing alone would be enough to handle it — that would make things a lot easier.
2. Every so often a part spills out and gets caught in the gears. Clearing it isn't hard, just eats up time. I get that you can't fully prevent that, but when it happens right at the very end of populating a big board... it's brutal, haha.

Anyway, not really complaining — just wanted to spell out what I meant. Would be nice if setup could be a bit smoother on both fronts.

2026年8月23日日曜日 23:21:01 UTC+9 tianl...@gmail.com: