Improvements of our DIY pick and place machine
Dear community,
As mentioned in the blogpost SMD Bestückungsmaschine (in german only), we built a DIY pick and place machine with BAM&DICE.
Because it worked very reliable, we made an effort to enhance the positioning accuracy and the diversity of components.
The result is presentable:
The rotation axis of the first version has been rebuilt completely. We replaced it with a NEMA8 motor with a hollow axis.
It is mounted on a new installed vertical axis which can be moved up and down with a second motor.
This way it is no longer necessary to move the whole Z-axis, which vastly improves the over all speed.
The hollow shaft is equipped with a cut 1ml syringe with cannula. For increased grip we slipped a rubber tip over the cannula.
The silicon tube of the first version was not leak-proof enough.
As you can see on the picture the whole construction is not perfectly straight so that the tip varies its position when turning, but this can be easily compensated in the software.
The next step was the replacement of the trays with real dispensers, so called feeders.
With huge components the trays are quite usable, but small components tend to bounce when collecting them.
So it happens that while collecting one component the 3 next to it bounce out of the tray.
These usually do not sit next to the tray but somethere on it, so that when collecting the next component not only does nothing stick to the tip but also the tray is shifted, because the tip collided with it.
That the tip is also bent by this is nothing I have to mention separately...
All things considered: suboptimal
The first idea was to buy used, professional feeders and integrate them somewhere in the machine. The problem was that these professional feeders are huge:
As comparison I inserted a typical spool of SMD-components.
The feeder is quite a clever solution considering the mechanical transport ot the components, but the crank for it is very stiff.
The mechanical construction to move this crank would be just too big - we don't have that much space in our DIY machine.
On the other hand such a feeder costs several hundred euros - a bit exaggerated when the feeders are more expensive than the whole rest of the machine.
So we looked for alternatives and found a very simple system:
In this system the components are also presented separately.
The feeding works by pulling off the cover tape which is wrapped around the motor shaft.
An optical sensor in the middle of the feeder tracks the tape of components at the side of the holes and always releases only one component.
As one can see in the picture this feeder is by far smaller than the professional one and so can be integrated into the machine much easier.
This system has some small disadvantages:
The wrapped cover tape has to be changed every 1000-2000 components.
You need one stepper motor and one sensor for each feeder.
The component tapes have to be respooled because the optical sensor is on the wrong side of the tape.
The only difference of the feeders among themselves is the size of the component slot.
This slot shoud fit the size of the component tape very well.
To not 3D print the whole feeder for each test, we only 3D printed the front part and adjusted the slot size until it fitted.
After the adjustment to the different component tape sizes the complete feeders have been 3d printed.
The controlling works still with BAM&DICE.
Since we use 4 of 5 stepper motor drivers for the axes, we had to connect a second BAM&DICE for controlling the 4 feeders.
The feeder controller monitors some digital inputs which are connected with outputs of the first BAM&DICE.
With these I/Os the movement of the feeders can be requested.
As soon as one of the I/Os turns high, the feeder controller turns the corresponding stepper motor until it gets a high-low state change of its optical sensor.
The software, openPNP could be adjusted to the new feeders with some small modifications.
After installing all 4 feeders we made a small video to show the outstanding performance of our DIY machine:
If you are interested in the feeders, the pick and place tip or the electronics or rather the interaction of all parts, you are invited to write an This email address is being protected from spambots. You need JavaScript enabled to view it.
to us.
We share the 3D models and send a download link to our version of the openPNP software.
The next adjustment is the further improvement of the feeders:
The separation of the feeding part and the part in which the components are released.
On the one hand one has more space around the motor shaft so it will be no longer necessary to change the cover tape every 2000 components. On the other hand the whole part should be designed in a way that one can use the SMD spools without respooling them.
Greetings from the lake constance
The 2PrintBeta - Team
Automatic cable cutter
Dear Community,
Once in a while we need different lengths of braid, PTFE-tube or shrink tube for a customer. Since the demand grows steadily and manual cutting takes its time, we thought an automatic cable cutter would do it better than us.
But as always: An industrial cable cutter is not really cheap. So only one solution remained: Do it yourself.
The principle is quite simple: A motor driven cable feeder and at the end something which cuts the cables. Both can be driven with a stepper motor, so we could use a BAM&DICE board with two DICE STK stepper motor drivers.
The feeder was realized with a set of RC wheels, which had been printed for test purposes some time ago. But what should we take to cut the cable?
The ideas ranged from razorblades to a holder for a dremel with a cut sheet. But a short test showed: scissors are sufficient. So why not use scissors?
After the idea comes the realisation. A motor was equipped with a drive disc which had a protruding bolt. With this the scissors can be moved while the second blade is secured with screws:
Before and after the drive wheels we mounted a guide which prevents the cable to slip aside.
After a few cutting tests with a very curly braid we found a little problem with the cable guide:
Right at the moment the cable gets cut, it tends to bend behind the blade of the scissors. The result is that the cable will not be driven between the blades anymore. It slips beneath Instead, gets blocked and the feeding is over.
To put things right we installed a small spring made of POM. This makes sure that before the next cut the cable gets fetched behind the blade. Besides that the spring is jagged which makes sure that larger cables don't slide along the blade but get caught by a jag.
As mentioned we used BAM&DICE to control the whole thing. We integrated it into the cutter:
So in the future it will be easy to cut the cables to the desired length at the push of a button.
It is not really fancy but it works remarkably reliable and the over-all costs are about one-tenth a bought automatic cable cutter would cost.
Last but not least a short video where you can watch our masterpiece at work.
Greetings from the lake constance
The 2PrintBeta - Team