It looks like its been a while since I have posted anything, but I have updated the Garden sensor project and added the Rover joystick project. As always, fairly busy so don’t get to these projects much. Progress was made on the garden sensor, but it has grown a bit too large for the housing I am using and so I need to consider either making a dedicated board to bring the bulk down, or build an extension under the housing to hold the bulk of the electronics.
Regarding the new project for a rover joystick, I did design and order boards … then redesign and order .. then redesign and order. I really need to design. Take a step back and review it again step by step to cut down on wasted boards.
Focus is an issue so projects actually finishing is a fun challenge.
Just a notification that more work has been posted on the garden sensor. I should add more pictures … too many words. New/updated schematic uploaded with a the updates being the explanation as to why.
Really has it been 7 months since I sent off for this?
I tried to design some modules with surface mount pins.
Well, today I finally soldered one together. The implementation is not as good as I had hoped. It worked. It was rather nice that it worked, but only one implementation. Pulling the module out of a breadboard snapped the delicate contacts off for some of the module. The pins were not cheap to do surface mount pins with, so this was rather disappointing. I will redesign it to have castellations. It wont be as svelte as it is now, but it should be a bit more robust. The surface mount pins just snap off too easily. I don’t know if that is a fabrication issue, or just a weakness inherent in the design. I will assume the later.
Edit: As a side note, Epoxy would probably help secure the pins well enough for repeated use.
Finally got the printer back. It turns out they shipped me a whole new one.
This one, so far, is working far better than the last one. The prints are a bit smoother and they complete. I haven’t done anything too tall yet, but some of the Z settings seem to be off as the nozzle starts to drag on the print as it gets taller, and it doesn’t take too much.
Printed 4* TUSH++ brackets to build my spool roller. Before it was sitting on a swing-arm rod with no bearings. It was temporary until I could build the spool holder, but needed to build the spool holder.
The brackets had bearings added (608 bearkings) per the directions for the TUSH++ setup. https://www.thingiverse.com/thing:2451368
I did not use the cross piece as part of the print. I did print it, but I don’t think it is needed. I put the holder and spool into an airtight container from Walmart (approx $8) and drilled a hole and put in teflon tubing that I then used Silicon adhesive on to secure. Will see if it holds. Put it on both sides to try and maintain the airtight-ness. Only air getting in now is from the teflon tube itself. Tube goes to the printer cable and is zip-tied to it a little loosely. Once that was setup, the filament had no issue coming off the spool. the spool rotates at the same rate as the extruder without any jerking in the line.
With the filament delivery option setup, I started printing some other calibration prints and they are coming out nicely.
Hopefully the Z belt, extruder, and heated bed last a while. The heated bed has already lasted 3 days, so that is indeed an improvement over the one day on the previous one.
I don’t think the printer is capable of jaw dropping prints, but it is able to do a decent rough job at present.
Still waiting on the M3d Pro to come back. Received a mail recently indicating it should be on it’s way back next week. We will see.
Havn’t worked on any electronics lately. Need to finish the projects that were started.
Found that Unity development, initially, is free as well as using Blender3D is free. Have bought training courses on these two pieces of software and have been enjoying them. We will see if it leads anywhere.
Received word on my printer that I sent off. It will be about 3 weeks or so to get all it’s parts fixed and sent back to me. The tech originally said they would upgrade any parts that had upgrades. From the sounds of it, it had quite a few.
I remain cautiously hopeful.
So as I received the items, they were much smaller than anticipated. I probably should have looked at the product dimensions, or looked at the connector as a size reference. The modules are indeed bright, but small. I think I will need to use all 4 in a single application to get the lighting results I was aiming for.
On the plus side, they also come with a resistor inline, so I don’t need anything to get them to run. A lot of people have reported that they stop working after 6 months or so and I attribute this to the LED being over-driven a bit to increase their output. Most figure they are cheap and disposable, so more can be sold that way, no?
I put a standard diode inline with the positive power rail to drop the power from the 12V brick from 12.5 to 11.5. I imagine in a car environment, the voltage can vary a bit, but using a wall-wart, it should be pretty steady. The ones I got put out 12.5V under load pretty consistantly. With all 4 running thats about half an amp of current and all 4 light up fully. They get warm, but not hot. Many complaints on these were also that they can get hot so I feel I’m on a good path for extending their life. They will light up with as low as a 9V battery, tho not nearly as bright.
Now I need to finish the shelf in the garage where it is badly needed and have them installed on an overhang. I need to solder together a small distribution board and get a cable extension for the power. The wall-wart is rated up to 1 amp and the diode is rated up to 30amps. With this in mind, I may order another set and put 2 more in the mix. for the garage to ensure maximum luminosity.
So while I wait for the printer to be repaired and continue down that path, I have a new project. One that probably should have happened while I was building out the crafting space.
Space. That is the problem. Work table in the garage is covered with tools and parts. Work table upstairs is cluttered with tools and parts. No good way to clean them up currently as all surface space is used. Time to go vertical. I have a shelving kit for the garage Work table that I just need to put together. The crafting room will require some shelves, but that should be about it. Doing so may add some shadow to the work surface.
Well, the garage was never lighted all that well to begin with and the upstairs work area
has a nice bright 5 bulb directional octopus lamp, but it sits on one side. I picked up some LED arrays built for cars to put under the shelves to help light up the work space. https://www.amazon.com/gp/product/B00J23EA2U/ref=oh_aui_detailpage_o00_s00?ie=UTF8&psc=1
I can buy $20 current controller devices … or build our own. Looking up on solutions, I settled with one by Dan on Instructables: http://www.instructables.com/id/Power-LED-s—simplest-light-with-constant-current/
The parts listed were a bit more then I wanted to pay for, so I looked for alternatives. I often avoided using mosfets if possible as I was unfamiliar with them, but this was a simple enough project and helps me to learn the basics of them… in addition to making the control circuit cheaper than $20.
I modified the above schematic to:
The basics are there, but added a dimmer switch and soft off capacitor as well as a diode… that is probably unnecessary and will be removed. I originally added it to drop the voltage slightly from 12V with the intent of not over-driving the LED array, but I can do that with the resistors and dimmer switch by controlling the current flow.
Waiting on parts at the moment.
Problems with the printer seem to be ongoing. Heated bed died within 24 hours. Z belt snapped within a week. Now the extruder gear is somehow destroyed. Unable to return based on the Indigogo method. Damn shaddy buisness model, but stuck with this machine for now. They do promise to fix it and it’s been shipped back to M3D to get fixed and updated if newer parts are available. Will see if it comes back in a workable condition.
Thoughts: Should have just got the Prusa Original MK2.
Last night, the Z-axis belt on the Pro broke. I’m not sure what caused it to snap, but this appears to be a very common issue right now. Machine is a week old. I’m not sure there is anything that can be unconsciously done to force this to happen. I wasn’t in the advanced tabs trying to force it. It was working on a gearbox print and I noticed it wasn’t going up in the Z axis. I tried restarting it twice before confirming that it wasn’t working. Then I went into the advanced settings to try and raise it manually, and zilch.
Took the bottom plate of the machine this morning and it is confirmed, a snapped Z belt. The belt looked very small for the load it is supposed to support. It was also toothed on both sides. As I understand it, it just runs around the outside of the track, so that was additional material that could have been there to reinforce the belt, but I didn’t have the whole bottom open to look at the whole track. I will do that tonight or tomorrow night. Certainly when the replacement comes in. I created the support ticket this morning and their support system estimates 2 days until the ticket can be looked at. I imagine they are a little busy with these issues.
The belt looked to be about half the size of the gear provided for it. When I look at it closer, I will verify that initial thought and if M3D doesn’t provide a better belt, I will see about sourcing one from elsewhere. Looking up timing belt profiles and seeing what may be a suitable replacement, I noticed the tooth type that is in this machine: HTD: High Torque Drive. It goes on to say that this type is great for high torque applications, but is the least precise of the different tooth types. Really M3D? “The HTD profile is typically used on applications that require minimal positional accuracy. The HTD profile or curvilinear profile is admirably suited to transmitting high torque.” I think it could be stated that a 3D printer may need positional accuracy just as much as the torque, if not more.
Looking into the different tooth profiles, STD appears to be the proper belt profile to use. “The positive engagement of the two sets of meshing teeth allows synchronous timing belts to transmit large torques and withstand large accelerations. Because of the positive engagement there is little relative motion and most importantly NO SLIP between the meshing teeth. Synchronous timing belts are extremely useful in applications where indexing, positioning or a constant speed ratio is required.”
It may be worth replacing the belt (and 4 stepper heads) to have this better tooth type. If it doesn’t work, I could always trade it back out.