Initially I thought I would make a rubics cube like thing for managing my production in EVE online. But after thinking things through I figured that it would not be challenging enough, and I wanted to improve my understanding of 3d printers. so I decided to make an articulated scara 3d printer, the primary advantage of it would be that it would be much more easier to bring into a clean room, than a normal cartesian like the one I built or delta printer would be. While scara on its own could in certain configurations be enough for this, articulation would give the printer several more degrees of freedom.
The benefit of the Scara is that it gives the printer accurate printing on the X and Y, and when you need the benefits of the articulation, you can just lock the scara in place and you have a fully articulated 3d printer. One of the weaknesses of the articulated printers is that the motions are controlled by so many motors, that errors multiply. So when you are printing normal layers, using the printer in scara mode, allows the number of problematic motors go from 7 to 2, making things much more stable, and as such easier to manage.
This setup also allows to have multiple printing platforms for a single printer, while it cannot be called mass production, it could be called continuous production. As one could have multiple jobs queued that would then automatically be printed on next free platform.
For this planning I used 123d Design, while its capabilities are limited, it is very useful in 3d drafting and allows changing designs quicker than initially using a full CAD program. For the purposes of this course I will be making the final designs on Autodesk Inventor, while I am curious about the Solidworks I believe that it is more important to choose my challenges apropriately and learning a entirely new program will be an unnecessary burden.
Well I have finally decided on the final project I want to do. I have been bouncing around from options on weekly basis, one week the liquid handler, other week a delta 3d printer around meter tall. Maybe a 3d scanner, but I have finally settled on a project. I will be making a multimeter, I figure it is as a project small enough that I get it done in time for Chile and meaningful enough to hold my motivation to complete it.
So far I have made boards for it, that can measure voltage, resistance and shortcircuits, the board that will measure current needs a shunt resistor that is not in the lab inventory. Still it is easily enough locally available.
I hope to add further features of capacitance measuring, diode and transistor measurements. Adding thermal sensor is also not very difficult. Also a interesting challenge would be to make a frequency counter for the multimeter. Initially I will design them as individual boards to make sure I get the feature working and then add everything to one chip.
I believe I have found a way to simplify the design further. Right now the standard of sorts in the industry is a twist knob. The problem is that if a lab has trouble sourcing the basic components, it probably also has trouble sourcing the materials for making a custom designed knob.
So while I will desing and cast something for the project, it needs to be for non-critical function, my own probe's for example. By using a n-mosfet from the lab inventory, I can switch on and off the power to the probes. this allows me to use internal power for testing shortcircuits and turn it then off for measuring voltage. I just need to add a button, so I can give the user some interface to switch features with.
In existing multimeters the twistknob does serve a purpose, it is a physical gate directing the flow of power, and as such used to choose the used feature.
Abd it is done, I have the board for it soldered, it has couple shorts I need to sort out, but once they are dealt with it is ready for programming. I already designed a case for it and it willbe 3d printed shortly. It still needs the ACS712 chip, and since the eagle didnt have the chip in its libraries I had to make a new part for it.
The BOM for it is as follows. The through hole pads are for the piezo buzzer, in case the LED is not bright enough to shine through the case. Altough the white LED can be painfully bright, might use a RGB led here in later version, because they are blindingly bright. the benefit of having a LED for the purpose of shortcircuit detection, is that then even deaf people can use the multimeter, as they wouldnt hear the beep.
In the 3rd edition of the board I found out that the regulator I use for the battery connection is not viable solution so I just bypassed it with a 0 ohm resistor and used an old voltage division board of mine to halve the 9volts the battery gives out, and in to the 4th edition of the board I will place the voltage division resistors directly on to the board itself.
Another issue I ran into with the 2nd edition was that once the piezo was in place the board was stuck in its case so I removed the LED and replaced it with a terminal connection so that the piezo cables can be detached from the board. Also I bought couple banana connectors for through plate install, and attached them to the lasercut plate, giving me proper connection to the probes. Into a later model I should get a second pair of banana connectors and leads, as the current sensor cannot be directly connected to the probes. The reason is that it connects directly to the line you are measuring, the vcc needs to flow through it so you cannot connect the ground to it.
Also I will need to come up with a more reliable way of controlling the N mosfet to get the benefit from it for the 4th edition of the board. Since it was suggested I also made the proper probes for it, with these I used the insides of a male connector to give me a proper probe head, also would be an idea to use a female connector for the ground, so one could just put it on a ground pin, and have a free hand while poking around with the other probe.