[atlab]

atlab, I'm unsure as to your calculation of "2.5*10/400=0.0625 mm". The steppers are 1.8 degree, or 200 steps per revolution and the stepper driver is 10-bit.

Stepper have rotor's stable points at step and halfstep positions, so 360/1.8/2=400.
Microstep is used only to make movemet smoother, not for positioning.
My driver have 1/4, 1/8, 1/10, 1/16, 1/20 microstep modes, but you can see in video that i can't positioning precisely (i used 1/16 mode).

Read this about microstep:
http://www.romanblack.com/stepper.htm

[Eclipze]

Stepper have rotor's stable points at step and halfstep positions, so 360/1.8/2=400.
Microstep is used only to make movemet smoother, not for positioning.
My driver have 1/4, 1/8, 1/10, 1/16, 1/20 microstep modes, but you can see in video that i can't positioning precisely (i used 1/16 mode).

Read this about microstep:
http://www.romanblack.com/stepper.htm

This description of the stable points for half and full step are with respect the holding torque and applied load. This holding torque is significantly lower when microstepping, however you can still move the stepper to a microstepped position. Microstepping is not just for smooth movement, but you do have to have sufficient torque for the load. If you do have insufficient torque to hold a microstepped position, then the stepper will slip to a half or full step position where higher holding torque can maintain the position. This means you need a stepper with higher torque for that application.

The only time I've read material on microstepping error is with respect to the actual held angle. The higher the microstep fraction, the higher the error (steps are not equidistant).

[Eclipze]

Left Z-Axis carriage plates. Bearing is a nice press fit. I had to machine down the height of the pulley - something that wasn't originally going to be a problem, but the datasheet on the stepper has a much longer shaft in the drawings than than the actual part that arrived.

[Eclipze]

Servo cams with bearing. Used to lift the Z-Axis.

[scsi]

That's a fast servo you have there. Maybe even too fast for the task. Unless you can control the accel/deceleration of this actuator, you can end up dropping heavy parts off the nozzle.

Here's the scenario. The servo is lifting the Z-assembly up and when it stops moving at the upper position, the axis will continue to travel up (there is a play in there), then it will meet the bearing. That's going to be quite fast deceleration which could cause the part to shift on the nozzle. After that, the Z axis will drop back down a couple mm's and meet the bearing again on the way down.

See if you can add a cushion of some kind around the bearing on the servo horn or otherwise reduce the play by as much as possible.

Another way is to use a "high torque" servo that has much higher gearing ratio and is several times slower. Many good "digital" servo's can be programmed for slower speed and acceleration.

-scsi

[Eclipze]

Servo mounts are done. I was quite happy to find the photo interrupter that goes on the back of the carriage platform (black part, last picture), aligns perfectly with the plate that blocks the beam. That is for a limit switch when the Y-Axis travels to the far rear.

[phomann]

Hi,

I'm not sure the servo speed will me a problem. On my Yamaha P&P machine, each head is is on a pneumatic slide. It had a strong spring return. When actived the head is just pushed down until it hits something. That something is the part in the feeder or the PCB when placing the part.


The servo is a bit complicated in reality, A small pneumatic slide may be simpler and lighter. Also being a on/off control, it only requires a digital output.

Cheers,


Peter.

[scsi]

phomann -

I bet those pneumatic actuators on Yamaha had air speed controllers on them. I'm describing a different problem with the Z-axis hitting a hard stop while traveling down with the part hanging on the nozzle.

Another acceleration related problem to consider. I'm working on a solder paste dispenser and researching this stuff a lot. Turns out that solder paste does not like fast motion (shaking) and won't stay very long in a homogeneous suspended state when riding on a shaky Z-axis. It tends to separate - the solder spheres accumulate in one place as they are heavy and the flux cannot not hold them in suspension.

-scsi

[Eclipze]

I have looked into solder paste dispension myself, and also concerned about the separation of the paste components. The use of air pressure for dispensing the paste is also something I'm not keen about, as the amount of paste ejected has a dependency on the amount in the syringe. I have been thinking more about using a screw based actuator to plunge the paste out a measured amount, with secondary suck back. But only using a small custom vessel that will need frequent refills I otherwise use stencils, but for custom jobs or a few prototypes a small paste tank would suit. Importantly, one I can unclip and only load when required.

I'm using Mach3 for the stepper control, where acceleration rates can smooth out any rapid movements that could bump the part off the nozzle :-)

[scsi]

I have been thinking more about using a screw based actuator to plunge the paste out a measured amount, with secondary suck back.

Speaking of screw based dispensers... Here's what I built today. Works surprisingly well and I can dispense 0.1uL volumes with 0.5mm nozzle while pumping thick grease. It's called auger valve. A proper one costs ~$5K : http://www.techconsystems.com/fluid-...ry-auger-valve

[Eclipze]

Oh very nice! How did you make the auger? Need some more pictures/details of how you have accomplished the mechanism please

[scsi]

The auger you can see on the desk next to the pump head. Took my Dremel and hacked the stepper motor shaft so that it resembles a flat screwdriver. Drilled couple of holes in a piece of HDPE and put some fittings here and there. Next time I will get proper luer fittings and machine the piece on my mill out of Delrin so it wont look so ugly.
The uC in the background is driving stepper by issuing step/direction commands following a simple program. Speed and pullback is adjustable via trimpots that uC is reading. Stepper driver is from Pololu. They also make cheap mylar stencils for $25+ http://www.pololu.com/catalog/product/1202

[Eclipze]

Drilled couple of holes in a piece of HDPE and put some fittings here and there.

Hahaaa... now your teasing without enough details

[scsi]

Literally, just two holes in a piece of HDPE not counting mounting holes. This was an ugly attempt of making something similar to the proper auger valve like Techcon ones on the picture attached. They have molded in luer fittings, which is nice. Techcon sells these Disposable Material Path valve assemblies in bags of 10 for something close to $500. Hence the attempt to make my own. It's not as easily disposable, but costs just a couple cents. The hardest part was to match the hole size of the screw well to the screw OD. I ended up making the hole slightly smaller and then used the screw itself to make it larger. Again, the screw that I'm using is the one you can see on the picture.

And another illustration - meat grinder that uses the same principle of operation. Some pressure is necessary on the inlet for the thing to work properly. Something like 5-10 PSI air is all it needs.

-scsi

[12oclocker]

Eclipze, I have been following this thread since it started, figured I'd sign up, I've been looking to either build my own pick and place, or buy one. You design looks good, if you can think of a way to add support on your floating axis I'm sure that would increase placement speed and stability, as well as accuracy. Look forward to seeing how your's turns out.

[phomann]

Eclipze, I have been following this thread since it started, figured I'd sign up, I've been looking to either build my own pick and place, or buy one. You design looks good, if you can think of a way to add support on your floating axis I'm sure that would increase placement speed and stability, as well as accuracy. Look forward to seeing how your's turns out.

Yes. I had a really old A3 scanner (SCSI bus) by Sharp, that had a similar arrangement, but had a dolly wheel on the out-board end. The head on it was very light.

Cheers,

Peter.

[Eclipze]

It wouldn't be to difficult to add a dolley wheel setup for vertical stability, or to use a rod and bearing for full support. I have specifically made provision to support this, as shown in the picture. But until it can see the necessity, I want to avoid over engineering the design and making it more complexity. It will mostly depend on the acceleration used, carriage mass, flex in the Y-Axis rail and the stability of the supporting X-Axis.


I'm interested in the Auger system, however it seems a rather messy mechanism to deal with. I'II post a drawing of the system I'm thinking about for paste dispension... as soon as I draw it hahaa...

I re-machined one of the brackets, as I didn't provide adjustment for the C-Axis stepper and it's mating with the gear. The carriage holes were also mis-aligned - blunt drill bit wandered.

The biggest issue I've had so far has been the stepper coupler. I had a massive 0.3mm offset error, which wobbled. That is a massive oscillating error introduced into the accuracy, so I had to machine another paying particular attention to getting the hole pockets perfect. It really is a part that should be done on a lathe. The new one was within 0.1mm before I used a dremel to carefully grind the pockets with constant checking. Managed to get the offset down to 0.01mm Took me a few hours though The X-Axis coupler is much larger/deeper... I'm going to have to get that done on a lathe.

[phomann]

Hi,

I also have a positive displacement paste dispenser. A paste syringe is
screwed into the unit and has a small +ve air feed into the end of the
syringe to help feed the paste. The unit has a small DC motor that runs
contineously.

The paste is fed by an auger (looks like a 3mm ballscrew). The auger is
driven by the DC motor via an electronic clutch. To dispence the paste,
activate the clutch, to stop, deactivate. This head is off a
commercial machine and is quite heavy. It's about 10" tall 1.5"
diameter. Also has a mic adjustment on a dovetail slide to set the
nozzle height.

Haven't played around with it much but it does seem promising.

[scsi]

The auger is driven by the DC motor via an electronic clutch.

It always surprises me how some engineers would do anything to avoid using stepper motor. Also, when I see an instrument of almost any kind with a stepper or two in it, it typically costs over a thousand. Why it's so hard to use a stepper?

-scsi

[Eclipze]

Hahaaa... I know what you mean.

This is just a rough concept of what I've been thinking of implementing for the solder paste dispension. The yellow rod through the stepper is a threaded, fine pitch.

I've only been thinking of having a small solder paste reservoir (for example, 2mL), for a few reasons. I'm concerned about long term vibration separating the paste components for a large amount. I want to use a stepper that isn't too heavy/large to plunge the paste (low torque). More accurate control over a smaller amount (actuation length vs total volume). Finally, it's something that I'd use for small jobs. Anything that would use a lot of paste (multiple board) I'd just make a stencil.

What I'm looking to achieve is having a paste block I can easily un-dock and put back in the fridge. The approach shown has two bolts to remove the paste holder, and the rod can be driven straight out of the stepper so I can leave the rod with stopper in the top, which means less mess and avoids the unnecessary introduction of air from removal of the stopper.

Having a thin/narrow reservoir will give me a higher precision with dispension, as well as helping to support using a stepper with lower power.

What I actually need to do with this design is make a slightly squarer profiled stopper, which will importantly prevent the rod from free rotation. The shaft isn't keyed, and if the rod is allowed to rotate it won't traverse up/down. This approach avoids having to provide a more complicated method to stop the shaft rotating (less weight, less complexity).

Using a linear stepper will hopefully give a nice consistent control, as well as supporting a suck back function.

Thoughts, ideas?