I built a homemade brake to bend sheet metal and it works OK, not great. The corners are really radius’s and fairly large ones at that. The second and main issue I have with it, is the inconsistency that comes from the bends. I find that even though I try to align things perfectly it does not work that way. Some of the parts I have made lately I have cut some slits with the CNC to weaken the folded edge making it more consistent and I wish I had done this for this part of the project….Anyway, a quick reminder of last years effort.
As you can see the hopper is quite small and I ended up adding an extension later. The burner is integral to the hopper and auger, making it very difficult to make any changes to the burner. Lastly the legs are welded on and have no adjustment and seriously two wheels what was I thinking? So while the principal is the same there are some major differences.
It’s pretty challenging to align four pieces of steel of such odd shapes and weights, so in this iteration I cut some gussets with tabs to fit into a slot in the end pieces of the hopper. This keeps the end pieces fixed from side to side because of the slot and plumb as it is butted up to the square the gusset forms. All from welding a gusset plumb. So that was an aid to success, of course as soon as I finished I tossed some ideas around with a friend and we thought up a perhaps better way by making two peices instead of four and doing a different bending technique. Next time. In the meantime this is straight, level and functional. Almost done with the hopper/auger assembly, The last task is to mount the electrical box and paint the entire assembly while it is still apart and then assemble it into a working machine and test the auger drive.
After last winter’s experience, I decided to start from scratch on the auger/hopper assembly. I varied the height several times through different burner designs. Now it looks like it was modified once too many times. The other main reason to rebuild the hopper assembly is the difficulty removing the burner from the boiler because it must stay balanced. I don’t remember exactly my thoughts at that point, I probably just had two wheels around the shop and was in a hurry.
So this fall I am rebuilding the hopper with a different auger drive system, and additional wheels and supports to make it much more robust and simple. The foundation of improvement in the new assembly is the ability to draw the parts in SketchUp as well as cut the parts with the CNC. Here’s a view of the cut out parts.
The parts laid out I call the saddle, angle iron, and side alignment plate.
The parts below are the angle iron plate with the side alignment plate. The angle iron plate is tabbed to fit into the slots of the side alignment plate, this way the parts are self aligning and jigging. Adding strength and ease of assembly. This makes the welding so much easier.
Finally here is the assembly on the welding table ready to weld, note the threaded rod, which also aids in the rigidity and ease of adjustment to make sure all the pieces are square and parallel prior to welding.
And finally the partially finished welded assembly, this assembly will be the foundation to cantilever the auger in the feed pipe as well as support the auger drive motor and gears.
The CAD designed parts combined with the CNC, combined with self jigging design for success make a nice finished assembly with light material for cost savings combined with good strength.
Notice the title didn’t say completely finished since the last few steps are to improve the acme screw feed as well as add a water table, ventilation system and wire up a better connector for the torch control. But the paint looks great and the electrical components in a box is great. The part you see cut is intended to be the plate that is the interface between the boiler and the burner. I call it the burn plate. It will go into the opening in the boiler, the holes in the very bottom will allow a resistance heater to light the biomass either pellets or chips, the holes above are for draft. The large hole in the top center is for the auger feed and the remaining holes are on the left, a fire detector sensor and a hole to be tapped for attaching the sensor. The two holes in the upper right are intended to be the swivel point and observation port to observe the flame. I drew the part that you see pictured up in Google Sketch Up. What really impressed me about Sheetcam is that the piece to swing and hide the flame in the observation port was made at the same time from the scrap that was cut out of the auger hole. Sheetcam automatically figured out the hole in this swinging part should be cut as a hole with the offset of the kerf to be on the inside of the hole, then cut the remaining profile as a outside kerf, then cut the auger hole as an inside hole. This was all done in the correct order automatically. Cool! Sheetcam is a great program.
The plasma cutter is working in the CNC mode! I have successfully cut parts that meet dimensions and are usable. I made the plates that mate with the casters for the bottom of the machine this morning. I have them welded on and working. It was a treat bolting them together with no alignment issues without spending the time to make them on the Bridgeport.
The CNC is not finished, I have to mount the cable tracks, as well as extend some of the stepper motor wires to allow the full movement of the machine. Before I move on I want to put a coat of paint on the steel parts to make it look finished and professional. But an important milestone has been reached. This machine will allow me to make parts designed on the computer as well as improve the quality of the chip boiler parts significantly. This week I intend to finish the CNC and rebuild the hoppers support wheels. More pictures when I finish the CNC.
After assembling all the lead screw nut to carriage connections the machine is testable. I ran a program supplied with Linux Enhanced Machine Controller that required coordinated motion between the axes. This came out in an interesting way, mirror image. I figured out that the y axis was opposite from the expected direction, or in other words, when the program called for a positive y axis move, the machine delivered a negative y axis move and vice versa. This was a simple fix, I physically picked up the axis and moved it end for end with a helper and remounted it. After that the test program ran just fine.
Image with corrected axis, you can see a little of the same program showing the E in the correct orientation, as well as a 5″ x 5″ box program that I was able to import from StepCam, although I did have to edit the program to get it to run.
Next on the CNC list is to order some additional materials from McMaster-Carr to finish the Z axis as well as some electrical routing accessories to finish the limit switch installations. Last but not least I will need to improve the mounting of the sharpie, plasma cutter and whatever other option should be available for this machine. No doubt a router head would be useful as well. I have the material for the fabrication of a base so I will start on that this coming week as well. Of course this is not the only thing on my list, I still have to make a living so progress will be measured in hours. Thanks for reading, and tell your friends.
After doing some internet research I have decided to go with Google Sketch Pro for the drawing of parts to produce a .dxf file., this will cost me a dollar or two but I have gone through the Google sketch tutorials and it is a powerful program. Google Sketch Pro is a CAD program, CAD standing for computer aided design. The CAD program is just the first step in a three step process of drawing the part and from that part drawing producing a .dxf file.
The second part of the process is to import the .dxf file into a program that will convert the geometry into standard G & M code this is a CAM program, which stands for computer aided machining. The program I have decided to use for this is SheetCam. SheetCam is a 2.5 axis program which makes it a little simpler to use. Of course the machine moves in the x and y directions but the third direction the z axis on a plasma cutter is really just up and down without much coordinated motion between the x and y axes while the axes are moving, we aren’t routing a 3D sculpture we are cutting a profile.
The third step is to use a third program which is a Linux based program called EMC which stands for enhanced machine control this takes the G & M code and translates that code into stepper motor signals that are exported through a parallel port to a Xylotex box which controls the motor motion to the 3 axes. Xylotex is the manufacturer/ integrator of the stepper motors and stepper motor drive boards.
At this point I have drawn a 5″ x 5″ square, downloaded that dxf file from Google Sketch Pro and created the code with SheetCam. The SheetCam program has a EMC post processor as a choice and I have the program ready to load into the laptop running Linux EMC to test the motions and accuracy of the CNC machine. I was able to draw the simple rectangle and convert it to G&M Code in about 3 hours this morning. Not a record but this does speak well for a well written intuitive CAM program, I had previous experience with Google Sketch so that doesn’t really count but I do like that program as well.
It hasn’t all been desk work I have also been turning the Bridgeport handles. The Y axis carriage plate has been built and awaits longer screws for final assembly.
I finished the assembly of the two carriages on the x axes of the CNC machine. The y axis is mounted and this next week the goal is to assemble the carriage to Acme nut connections and build the linear carriage for the y axis. So if all goes as planned next weekend or sooner could be the first prototype test of the machine, the z axis will not be workable but I am sure I can find a way to attach a pencil the axis to see movement. That is a great goal for the week.
My SparkFun package of an FTDI Adapter board and and second board will allow me to reload the firmware on the OpenLog board I ordered in January. I am dreading this a bit, electronics and computer area things are my weakness so I am a bit nervous but will tackle this job and test in a example sketch this week. I’ll bet that will take any free time I may have up but will get me that much closer to my goals.
Having taken most of the spring and summer off after last winters heating season, it is now time to review progress and regroup to make more progress. I have been spending part of my “free” time working on a CNC Plasma cutter. The goal of this machine is to improve the speed of prototyping and aid in the quality of the parts produced.
I’m not a very patient person so having a CNC make the cuts instead of me trying to either free hand or follow a straight edge with the plasma cutter will definitely improve the quality especially of the welds. At this point I am building the carriages and exploring what software to use in conjunction with the cutter. I am intending to use Mach 3, but I need to review and find a CAD program and CAM program that work together.
In conjunction with the CNC project I am working on getting a SparkFun OpenLog device to log data to a MicroSD card. At some point I know this will be useful to the project. I ordered the OpenLog from SparkFun in February so with the new Arduino IDE (1.0.1) the firmware for OpenLog needs to be updated. This requires ordering a FTDI board or using a FTDI cable. a second board is recommended as well. These items are on order and should arrive this week. Of course after updating the firmware and config file, the next step will be getting the software integrated into the Arduino sketch to make it a practical useful device. I think when I get that done I will make that section of the code a function and post it to the Arduino forum.
Lastly I am working on a PID control for the software, this is really in the development phase and not a lot of time has been spent on this aspect. To enable the boiler to convert from pellets to chips to sawdust to acorns? I know the software will need to adjust the burn parameters to make the product burn seamlessly.
Oh yeah I almost forgot, I have about 6-8 tons of chips in my driveway waiting to be dried. The drying needs to be done by me building a dryer…for now I am focusing on the CNC cutter and when McMaster-Carr or UPS fail me I can fall back to working on OpenLog functions for practical data collection.