Cheap, plentiful, accurate nozzles (that you may already have...)

[darens...@gmail.com]

Building up a lightweight BLDC part rotator for the P2 and I was generally unhappy with using the luer lock nozzles.  They always needed correction for runout.

Had a little brainstorm for some new nozzles.  Use stainless steel 3D printer nozzles.

ID of the nozzles is 2.0mm, which is also the OD of my part rotator shaft.  

It's a close fit, but I added the nut to make sure it stays put.

The centering is basically perfect, and you can get nozzle diameters from 0.1mm to 1.0mm or higher.

The height is also basically standard so you don't need to recalibrate your depth when you change tips.

The clearance to neighboring tall components could be an issue, but if I get to that bridge, I could get that down to ~3mm by grinding off the bolt pattern.  Until then, I'm keeping it as it gives me a nice way to automate nozzle switching.

The project is just getting started, but I like where it's going already.

So, what do you think?

[Michael Anton]

I just read your hackaday.io page, and noticed you are looking at trying the L6234 BLDC motor driver.  I have an eval board that I designed for that IC, and I'd be happy to send you a bare PCB to use for your experiments.  Message me your address, and I'll get one out to you.

[Daren Schwenke]

Ha!  Your design was pretty much the one I was going to emulate. :) 

I have not gotten around to linking or attributing the research I have done yet, but your work was number one or two for sure.  I will message you.

--
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/MykMjh_EL2Y/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/adddfffc-0361-4c11-8b3f-a89b1d5c3b69%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Michael Anton]

I'm not totally surprised... :-)

  You probably came across the Berryjam site first, which has a link to my board.  I get the bulk of my traffic from his site.  Every now and then I thank him by sending him some beer money, and the first time I did that, it really surprised him.  It turns out that my design was the inspiration for his project, and here I was thanking him.  The internet certainly can be a well linked community at times, and I think that is fantastic.

On Friday, April 26, 2019 at 10:33:32 PM UTC-6, Daren Schwenke wrote:

Ha!  Your design was pretty much the one I was going to emulate. :) 

I have not gotten around to linking or attributing the research I have done yet, but your work was number one or two for sure.  I will message you.

On Sat, Apr 27, 2019, 00:26 Michael Anton <3d.m...@gmail.com wrote:

I just read your hackaday.io page, and noticed you are looking at trying the L6234 BLDC motor driver.  I have an eval board that I designed for that IC, and I'd be happy to send you a bare PCB to use for your experiments.  Message me your address, and I'll get one out to you.

On Friday, April 26, 2019 at 4:06:58 PM UTC-6, darens...@gmail.com wrote:

Building up a lightweight BLDC part rotator for the P2 and I was generally unhappy with using the luer lock nozzles.  They always needed correction for runout.

Had a little brainstorm for some new nozzles.  Use stainless steel 3D printer nozzles.

ID of the nozzles is 2.0mm, which is also the OD of my part rotator shaft.  

It's a close fit, but I added the nut to make sure it stays put.

The centering is basically perfect, and you can get nozzle diameters from 0.1mm to 1.0mm or higher.

The height is also basically standard so you don't need to recalibrate your depth when you change tips.

The clearance to neighboring tall components could be an issue, but if I get to that bridge, I could get that down to ~3mm by grinding off the bolt pattern.  Until then, I'm keeping it as it gives me a nice way to automate nozzle switching.

The project is just getting started, but I like where it's going already.

So, what do you think?

--

[darens...@gmail.com]

I also came across this one, which seems to be a little more advanced in a few ways.  

https://www.monolithicpower.com/en/mp6536.html

Lower max voltage, but slightly higer output current, a *much* smaller package, faster mosfets (which should translate to less thermal loading), and less external support components required.  

It is small enough that it could be made into a stepstick, with an STM32 on the underside.  Just mentally totalling up the real estate required for the sense resistors, caps, and footprint of the two major components plus a linear regulator.  There is enough room, probably...

It also has all three phase current taps broken out (which the L6234 has two of them sharing one), which from what I understand is kinda required if you wanted to try to do Field Oriented Control.

That is about as deep as I got into it.  

I looked at a couple others chips as well, along with some gate drivers which people have used directly as the output stage due to their ridiculously high current handling capability.

But... I'm lazy and don't want to think about things like shoot through or high side driving.  :)

I do want to leave the future open though and implement the minimal current sensing and such so in the event someone want's to do the work... it could do FOC.

But, I got 2 L6234's sitting in my parts bin, so here we are..  :)

[Michael Anton]

The downside, is that you would likely have a very difficult time running that part at high current on a small board.  It is difficult to get the 3.5W of dissipation at 5A out of a package that tiny, and especially more difficult if the PCB is small.  Though for the motor you are planning to use, it likely wouldn't be a problem, since it probably isn't a high current motor anyhow.

[darens...@gmail.com]

Yeah... there would be serious thermal issues in a package that small at anywhere near the rated current.  The Rds_on value for those chips is better though...

Nothing a glued on heatsink wouldn't fix though, and going 'up' is free.  :)

I kept a stash of old Pentium II heatsinks (the ~6x2in brick shaped ones with the flat backplane), which I now run over the table saw whenever I need the odd shaped high dissipation heat sink.  All my Pi's got one...  The old Pentiums were beasts when it came to thermal sinking requirements.

[Mike Menci]

if you rely want this nozzles - here a better option 

https://www.aliexpress.com/item/3D-Printers-Parts-MK7-MK8-Nozzle-0-4mm-0-3mm-0-2mm-0-5mm-Copper-Extruder/32960833723.html?spm=2114.search0104.6.12.20747e9b4LKyOR&s=p

A bit hard to change - no quick change by PnP possible...

Mike

[darens...@gmail.com]

Yeah the brass ones I had were actually better for runout than the stainless ones, and have a better angle on the tip.

I screwed up this shaft trying to correct for pressing it in crooked but it's still pretty good.

https://youtu.be/BVPuP0r4SCA

[bert shivaan]

So far I like this idea, but how will you spring compensate in Z?

--
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/a96259f6-c875-489b-81cd-347c6e4b9f87%40googlegroups.com.

[Daren Schwenke]

I could get about 0.5mm from a spring on the shaft of the motor, but I was just planning on using the Z accuracy I get with my motion platform honestly.

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/MykMjh_EL2Y/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/CA%2BKNHNwvUnyM254R7rBibZh%3DdeB_BgwMdN_7BSzyQLnFsqeAVA%40mail.gmail.com.

[bert shivaan]

So I agree with machine accuracy, but then how to overcome tolerances on packages? I just did a random check on some datasheets I have handy, the spec's went from .1mm
tol on the overall height (small QFN) to almost .4mm for a SOT23-5. add to that some tol on paste thickness. I would be very afraid I would not seat some parts and squeeze the paste from under others.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CAKGtx5TsjxzaJ9qkL919c7KRW%3D3gGo-A%3DEG4MBn81cKdjYNxyQ%40mail.gmail.com.

[Daren Schwenke]

So how much vertical motion do I need, and how much force..  

I may be able to just make a little room inside the motor and use the stator/magnetic attraction as my spring..

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CA%2BKNHNwRXj03v0GhTA4LtHjmbqPmKn5FWpC6q1VrfYFhtoH1gA%40mail.gmail.com.

[bert shivaan]

Well I am not authority by any means, but I am able to measure my nozzles. On average they move 1.7mm (they seem to vary between 1.6-1.8)

As for the force, it seems to vary based on nozzle size. My smallest (.5mm id) is about 150grams and my biggest (3mm id) is about 350grams.

The force was different for different size nozzles.

I want to note these are DIMA nozzles. Used on a Dymaxon. So it would be great to get some measurements from others with different brand nozzles.

[darens...@gmail.com]

Measured the magnetic spring idea forces and travel.

It generates a pretty linear 0-80g of force over a travel of 1mm.  

Then it stays just around 80g+-10g until it starts to drop off about 4mm later.

I'm measuring this in the direction it can move right now, which is opposite the way it would need it to move from the way I have it mounted.  

Or I could use the other end of the shaft I suppose... 

Even 40g feels like a whole lot of force for placing a part...

[darens...@gmail.com]

What I'm measuring is just the static pull of the magnets against the stator.  I imagine that is going to change a little once the stator is energized.

I was kinda hoping here for a moment that perhaps I could increase/decrease the field strength of all phases the same to maintain the same angle, but increase/decrease the force applied.

However, I think that idea has a big hole in it in that we are both pushing and pulling on two magnets when we energize a coil, so the net change in the retention force over the normal magnetic attraction to the iron stator would probably balance out to zero.

It may still have some inductive resistance to faster motion though at higher current, which could be used?  I dunno.  Just brainstorming.

[bert shivaan]

Maybe you can simply have the entire thing on a spring mount like between it and what drives it? Does this make sense?

[darens...@gmail.com]

Rather than contaminate this thread further with conversation about the spring pressure provided by various nozzles, I made a new thread.

Nozzle part placement pressure

[darens...@gmail.com]

On Tuesday, April 30, 2019 at 10:17:05 PM UTC-4, cncmachineguy wrote:

Maybe you can simply have the entire thing on a spring mount like between it and what drives it? Does this make sense?

It would probably be difficult to keep this constrained to a single axis of movement.  It would need a sliding interface (and not a pivot) to not introduce rotation about the X or Y axis as the nozzle presses on that part.

That being said, I do have a sliding interface with the shaft already.  It doesn't have any bearings for that motion, but I'm ok with a drop of lube for the weight savings it would provide. 

[Jarosław Karwik]

Pictures as promised. Last one is 400steps/rev stepper which I intended to use for these nozzles ( it has 5mm hollow shaft, fits directly for 5mm linear bearings)

[mike....@gmail.com]

So on EACH nozzle OD is 5mm ?
Are all nozzles the same length ?

Mike

[Jarosław Karwik]

The diameter of the nozzle "lock" part is 9mm. Length differs between nozzle types.

See pictures for the original idea ( note - there is one part missing - could not find it - magnet holder on the motor shaft to keep the nozzle)

[mike....@gmail.com]

Is the white collar the seal (sealing ring)?
For automatic changing of nozzles all nozzles should have equal length.
- Maybe the part from seal to nozzle tip is the one that has the same length?
- how many different Nozzle types are there?
Mike

[Jarosław Karwik]

White collars ( 3D printed)  - they are just rings to be used for nozzle holder. One has metal part ( washer), so magnets on the motor shaft can keep them.

There are lik 10..15 different types - I have not counted them.

Original head does not care about different nozzle length - I did have such machine, it had software configuration for it. Mechanically , the "socket" of each nozzle is in the same place. If you make your own system, you can compensate length by having different collars.

[Jarosław Karwik]

Seal (rubber gasket) is in the linear bearing.

[Marek T.]

I won't change the system now, but in future who knows. It could be good to see this with own eyes...

What city is your place to live? Not Warsaw maybe?

[Jarosław Karwik]

I live on Gdańsk, but i am quite often in Warsaw. 

--

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/baff4dd1-70e3-4e78-a61c-0059620fe675%40googlegroups.com.

[Marek T.]

Great, so let me know if you will get the plan to be here and we'll try to meet.

[Maple_Dude]

I'm not sure I understand what I'm looking at. 

[Jarosław Karwik]

More pictures then :-)

[Maple_Dude]

So are these nozzles something you designed or what am I missing here? 

[Jarosław Karwik]

These nozzles are stock siemens nozzles ( for their SIplace line).

See e.g. 

https://tools.adoptsmt.com/catalogs/AdoptSMT_Catalog_Siplace_pick_and_place_web.pdf

You can buy automatic holders and heads for them - but it is very expensive ( I used to have third party machine using the system).

You have seen on my pictures some adapters I made for one of my prototypes. 

The nozzles are good quality originals, so if you have multi head machine without need to change nozzles they are excellent choice.

[Daren Schwenke]

I'm happy this thread got more use out of it than just my little idea.  Thank you!

--

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/MykMjh_EL2Y/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/601f6f57-f933-489b-9bd9-19e092333179%40googlegroups.com.

[Mike Menci]

Hi Jarosław

Are you you willing to sell some for my own tests ?

Mike