search

TinWhiskers / fpd-design

Design document for FirePick Delta
1 stars 0 forks source link

CNC PnP Nozzle Redesign, or retrofit Juki nozzle onto FPD #50

Closed FlyingLotus1983 closed 8 years ago

FlyingLotus1983 commented 8 years ago

Need to quickly make a call whether to:

  1. Redesign the current CNC nozzle to have a large, colored ring for up-looking vision purposes
  2. Source or fabricate a stepper motor shaft mount that allows off-the-shelf Juki nozzles to be mounted.

Either of these choices, ideally, would have the ability to perform auto-nozzle changes.

vonnieda commented 8 years ago

As a temporary (or even permanent, phase 1) solution we could use: http://www.robotdigg.com/product/517/Nozzle+w/+Adapter+to+5mm+Hollow+Shaft+Stepper

itsmefelix- commented 8 years ago

We have a finite amount of engineering resources at the moment and I wonder if bespoking a solution for this part is the best use of our time. I'd like to find something off the shelf if possible.

WillD77 commented 8 years ago

Does the robotdigg part count as OTS?

FlyingLotus1983 commented 8 years ago

Yes, that would count as off-the-shelf. However, the RobotDigg shaft adapter uses a set-screw, so it's permanently mounted. If we want to be able to switch between paste application and PnP with our current head, then we'll need an o-ring setup at the least. We've done this type of thing before and would be extremely easy to make an 8mm to 5mm shaft with an o-ring on the nozzle end... that's perfectly within our engineering bandwidth right now.

The only other thing that worries me about the Juki nozzles is the spring-loaded end. While great for placing SMT parts gently, it wasn't what I had mind for Z probing. My idea was to use a fixed (non-spring) nozzle and use the back-EMF sensing. If we can figure out Z probing some other way, then I'm otherwise fine with the Juki nozzle but @itsmefelix- we are going to need a good reliable source of not only the nozzles, but also the shaft adapters. This means that we need to be able to order nozzles with an exact orifice size and other characteristics, for example, 0.8mm, if we get something else or a mixture of different types, that's going to get the pressure sensing all out of whack, since vacuum/pressure measured is influenced by orifice size. Have you already looked into finding a particular reputable vendor and asking them if they're able to meet our needs?

WillD77 commented 8 years ago

Switching between paste application and PnP? Paste application doesn't require the nema motor. Why would that not work?

vonnieda commented 8 years ago

@FlyingLotus1983 The spring loaded end should easily provide enough push back to sense with back EMF. The springs are pretty strong. If we can't sense that we'll need to reconsider back EMF anyway, because that much pressure to trigger it will cause mechanical flexing in the machine.

@WillD77 we need to be able to quickly remove the nozzle to switch to paste application. If the nozzle is mounted with a set screw this is less convenient for the user.

itsmefelix- commented 8 years ago

I know another issue is the length of the adapter and the actual nozzle, which when combined will be much lower to the bed plate than our current configuration, but maybe we can try other off the shelf nozzles as well.

WillD77 commented 8 years ago

It's only 7mm longer.

WillD77 commented 8 years ago

Adapter has been shown to work as proof of concept. Will test with O-rings as soon as we receive them. We might need to use a set screw to hold the adapter to the NEMA motor.

FlyingLotus1983 commented 8 years ago

The prototype will have a set screw. I'm closing this issue for now, but I've opened an issue here: https://github.com/TinWhiskers/fpd-design/issues/65 for upgrading this design to a quick-change using o-rings, ball bearings, springs, or something to that effect.