First alloy on the 3040 cnc (with 2.2kw spindle)

There are times when words are not needed. When you see a 3040 or 6040 cnc without any enclosure there is a good chance that the machine doesn't see heavy alloy cutting. It only takes a few videos to see how chips are thrown around when a 24krpm bit touches a block of alloy. As a prelude to any alloy being cut I enclosed the 3040 in a "terrarium". This was itself an interesting build and as usual I overdid the design. The top and bottom box frames are made of 5cm square timber with a fairly solid base panel. The back is just light junk with plywood bolted to tabs on each side so I can replace things as I feel. The door opens beyond 90 degrees to get right out of the way and closes to rest on the base 5cm timber at the front of the enclosure.

For anybody reading this I have one word of advice, any gaps in the first 50cm from the machine base will have chips thrown at them. So make sure that the angles the chips might come from near the spindle have been accounted for with your air venting that allows some cooling into the mix. The sides of this case are more than 80cm in height.

The next modification is a mister to help clear local chips and bring some light amount of cutting fluid into the cut zone. The first runs were just using a light spray of CDT over the cut zone before job start.

The very end of one of the first runs is shown in the below video.



The parts being cut are wheel mount crossover plates to allow an outdoor robot to have larger wheels attached. The wheels want M8 bolts, the motor mount is an actobotics pattern, so an M4 hole was a good fit there. Because it's CNC the part itself was cut with many splines to include material where it could do structural good and exclude it otherwise.

I found it useful to cut templates in MDF to test the fit before a final run. This fed into part 3 which includes mounting holes for all 4 bolts of the hub mount. The alloy version 4 also has rounded ends and is shown attached to the wheel. This will let some cheap $10 wheels which are 12 inch across mount to an actobotics based robot.

I'll have video of the "houndbot" in action using these mounts next time.

3040/24,000 CNC first dry run in place

The progression has finally reached an upgraded CNC with high power spindle. Things still move around fine to the eye, the next step is likely to do some test drills at known distances to see if the additional weight has had an impact on the steppers that can't be easily seen.

A few interesting times when spinning up to 24,000. At around 320hz there was a new loud rattle. I think this turned out to be resonance with either something that was on the cutting plate or the washers on the toggle clamps.

There is going to be video once this machine starts eating alloy. The CNC needs to be lowered into an enclosure (the easier part) so that chips and the like go into a known location. The enclosure itself needs to be made first ;)

Ironically a future goal is to be going smaller. Seeing if twice the number of microsteps can be pulled off in order to get better precision and cut QFN landing zones on PCBs.

3040 spindle upgrade: the one day crossover plate

Shown below is the spindle that came with my 3040 "engraving" cnc next to the 2.2kw water cooled monster that I am upgrading to. See my previous blog post for videos of the electronics and spindle test on the bench.

The crossover plate which I thought was going to be the most difficult part was completed in a day. I had some high torsion M6 nuts floating around with one additional great feature, the bolt head is nut shaped giving a low clearance compared to some bolts like socket heads. The crossover is shown from the top in the below image. I first cut down the original spindle mount and sanded it flat to make the "bearing mount" as I called it. Then the crossover attaches to that and the spindle mount attaches to the crossover.

Notice the bolts coming through to the bearing mount. The low profile bolt head just fits on each side of the round 80mm diameter spindle mount. I did have to do a little dremeling out of the bearing mount to fit the nuts on the other side. This was a trade off, I wanted those bolts as far out from the centre line as possible to maximize the possibility that the spindle mount would bolt on flat without interfering with the bolts that attach the crossover to the bearing mount.

A more side profile is shown below. The threaded rod is missing for the z-axis in the picture. It is just a test fit. I may end up putting the spindle in and doing some "dry runs" to make sure that the steppers are happy to move the right distances with the additional weight of the spindle. I did a test run on the z-axis before I started, just resting the spindle on the old spindle and moving the z up and down.

I need to drop out a cabinet of sorts for the cnc before getting into cutting alloy. The last thing I want is alloy chips and drill spirals floating around on the floor and getting trecked into other rooms.

3040 for alloy

I have finally fired up a 2.4kw 24,000 rpm spindle on the test bench. This has water cooling and is VFD controlled. The spindle runs on 3 phase AC power.



One thing that is not mentioned much is that the spindle itself and bracket runs to around 6-7kg. Below is the spindle hitting 24,000 rpm for the first time.


With this and some other bits a 3040 should be able to machine alloy.

Houndbot progresses

All four new shocks are now fitted! The tires are still deflated so they look a little wobbly. I ended up using a pillow mount with a 1/4 inch channel below it. The pillow is bolted to the channel from below and the channel is then bolted from the sides through the alloy beams. The glory here is that the pillows will never come off. If the bolts start to vibrate loose they will hit the beam and be stopped. They can not force the pillow mount up to get more room because of the bolts securing the 1/4 inch channel to the alloy beams coming in from the sides.

I'm not overly happy with the motor hub mount to wheel connection which will be one of the next points of update. Hopefully soon I will have access to a cnc with a high power spindle and can machine some alloy crossover parts for the wheel assembly. It has been great to use a dual vice drill and other basic power and hand tools to make alloy things so far. But the powerful CNC will open the door to much 2.5D stuff using cheapish sheet alloy.

But for now, the houndbot is on the move again. No longer to the wheels just extend outward under load. Though I don't know if I want to test the 40km/h top speed without updating some of the mountings and making some bushings first.

Houndbot suspension test fit

I now have a few crossover plates in the works to hold the upgraded suspension in place. See the front wheel of the robot on your right. The bottom side is held in place with a crossover to go from the beam to a 1/4 inch bearing mount. The high side uses one of the hub mount brackets which are a fairly thick alloy and four pretapped attachment blocks. To that I screw my newly minted alloy blocks which have a sequence of M8 sized holes in them. I was unsure of the final fit on the robot so made three holes to give me vertical variance to help set the suspension in the place that I want.

Notice that the high tensile M8 bolt attached to the top suspension is at a slight angle. In the end the top of the suspension will be between the two new alloy plates. But to do that I need to trim some waste from the plates, but to do that I needed to test mount to see where and what needs to be trimmed. I now have an idea of what to trim for a final test mount ☺.

Below is a close up view of the coil over showing the good clearance from the tire and wheel assembly and the black markings on the top plate giving an idea of the material that I will be removing so that the top tension nut on the suspension clears the plate.

 The mounting hole in the suspension is 8mm diameter. The bearing blocks are for 1/4 inch (~6.35mm) diameters. For test mounting I got some 1/4 inch threaded rod and hacked off about what was needed to get clear of both ends of the assembly. M8 nylock nuts on both sides provide a good first mounting for testing. The crossover plate that I made is secured to the beam by two bolts. At the moment the bearing block is held to the crossover by JB Weld only, I will likely use that to hold the piece and drill through both chunks of ally and bolt them together too. It's somewhat interesting how well these sorts of JB and threaded rod assemblies seem to work though. But a fracture in the adhesive at 20km/h when landing from a jump without a bolt fallback is asking for trouble.

The top mount is shown below. I originally had the shock around the other way, to give maximum clearance at the bottom so the tire didn't touch the shock. But with the bottom mount out this far I flipped the shock to give maximum clearance to the top mounting plates instead.

So now all I need is to cut down the top plates, drill bolt holes for the bearing to crossover plate at the bottom, sand the new bits smooth, and maybe I'll end up using the threaded rod at the bottom with some JB to soak up the difference from 1/4 inch to M8.

Oh, and another order to get the last handful of parts needed for the mounting.

Houndbot rolling stock upgrade

After getting Terry the robot to navigate around inside with multiple Kinects as depth sensors I have now turned my attention to outdoor navigation using two cameras as sensors. The cameras are from a PS4 eye which I hacked to be able to connect to a normal machine. The robot originally used 5.4 inch wheels which were run with foam inside them. This sort of arrangement can be seen in many builds in the Radio Controlled (RC) world and worked well when the robot was simple and fairly light. Now that it is well over 10kg the same RC style build doesn't necessarily still work. Foam compresses a bit to easily.

I have upgraded to 12 inch wheels with air tube tires. This jump seemed a bit risky, would the new setup overwhelm the robot? Once I modified the wheels and came up with an initial mounting scheme to test I think the 12 inch is closer to what the robot naturally wants to have. This should boost the maximum speed of the machine to around 20km/h which is probably as much as you might want on something autonomous. For example, if your robot can out run you things get interesting.

I had to get the wheels attached in order to work out clearances for the suspension upgrade. While the original suspension worked great for a robot that you only add 1-2kg to, with an itx case, two batteries, a fused power supply etc things seem to have added up to too much weight for the springs to counter.

I now have some new small 'coil overs' in hand which are taken from mini mountain bike suspension. They are too heavy for what I am using, with around 600lb/inch compression. I have in mind some places that use coil overs in between the RC ones and the push bike ones which I may end up using. Also with slightly higher travel distance.

As the photo reveals, I don't actually have the new suspension attached yet. I'm thinking about a setup based around two bearing mounts from sparkfun. I'd order from servocity but sfe has cheaper intl shipping :o Anyway, two bearing mounts at the top, two at the bottom and a steel shaft that is 8mm in the middle and 1/4 inch (6.35mm) on the edges. Creating the shafts like that, with the 8mm part just the right length will trap the shaft between the two bearing mounts for me. I might tack weld on either side of the coil over mounts so there is no side to side movement of the suspension.

Yes, hubs and clamping collars were by first thought for the build and would be nice, but a reasonable result for a manageable price is also a factor.

Making surface mount pcbs with a CNC machine

The cool kids^TM like to use toaster ovens with thermocouples to bake their own surface mount boards at home. I've been exploring doing that using boards that I make on a CNC locally. The joy of designing in the morning and having the working product in the evening. It seems SOIC size is ok, but smaller SMT IC packages currently present an issue. This gives interesting fodder for how to increase precision down further. Doing SOIC and SMD LED/Resistors from a sub $1k CNC machine isn't too bad though IMHO. And unlike other pcb specific CNC machines I can also cut wood and metal with my machine :-p


Time to stock up on some SOIC microcontrollers for some full board productions. It will be very interesting to see if I can do an SMD usb connector. Makes it a nice complete black box to do something and talk ROS over USB.

Terry & ROS

After a number of adventures I finally got a ROS stack setup so that move_base, amcl, and my robot base all like each other well enough for navigation to function. Luckily I added some structural support to the physical base as the self driving control is a little snappier than I normally tend to drive the robot by hand.

There was an upgrade from Indigo to Kinetic in the mix and the coupled update to Ubuntu Xenial to match the ROS platform update. I found a bunch of ROS packages that I used are not currently available for Kinetic, so had an expanding catkin ws for self compiled system packages to complete the update. Really cool stuff like rosserial wasn't available. Then I found that a timeout there caused a bunch of error messages about mismatched read sizes. I downgrade to the indigo version of rosserial and the error was still there, so I assume it relates to the various serial drivers in the Linux kernel doing different timing than they did before. Still, one would have hoped that rosserial was more resilient to multiple partial packet delivery. But with a timeout bump all works again. FWIW I've seen similar in boost, you try to read 60 bytes and get 43 then need to get that remaining 17 and stuff the excess in a readback buffer for the next packet read attempt. The boost one hit me going from 6 to 10 channel io to a rc receiver-to-uart arduino I created. The "joy" of low level io.

I found that the issues stopping navigation from working for me out of the box on Indigo were still there in Kinetic.  So I now have a very cool bit of knowledge to tell if somebody has navigation working or is just assuming that what one reads equals what will work out of the box.

Probably the next ROS thing will be trying to get a moveit stack for the mearm. I've got one of these cut and so will soon have it built. It seems like an ideal thing to work on MoveIt for because its a simple low cost arm that anybody can cut out and servo up. I've long wanted a simple tutorial on MoveIt for affordable arms. It might be that I'm the one writing that tutorial rather than just reading it.

Video and other goodness to follow. As usual, persistence it the key^TM.

libferris 2.0

A new libferris is coming. For a while I've been chipping away at porting libferris and it's tree over to using boost instead of the loki and sigc++ libraries. This has been a little difficult in that it is a major undertaking and that you need to get it working or things segv in wonderful ways.

Luckily there are tests for things like stldb4 so I could see that things were in decent shape along the way. I have also started to bring back the dejagnu test suite for libferris into the main tree. This has given me some degree of happiness that libferris is working ok with the new boost port.

As part of that I've been working on allowing libferris to store it's settings in a configurable location. It's a chicken and egg problem how to set that configuration, as you need to be able to load a configuration in order to be able to set the setting. At the moment it is using an environment variable. I think I'll expand that to allow a longer list of default locations to be searched. So for example on OSX libferris can check /Applications/libferris.app/whatever as a fallback so you can just install and run the ferris suite without any need to do move setup than a simple drag and drop.

For those interested, this is all pushed up to github so you can grab and use right now. Once I have expanded the test suite more I will likely make an announced 2.0 release with tarballs and possibly deb/rpm/dmg distributions.

New filesystems that I've had planned are for mounting MQTT, ROS, and YAML.

Through hole PCB Making -- Same Day

I initially thought that removing the multiple week wait for a board would be the true joy of making PCBs locally. It turns out that quick iteration is the best part. Version 2 and 3 of a board flows quickly and you end up with something unexpected after only a few days of tinkering.

I'm still at the level of making through hole stuff. Hopefully I can refine the process to allow some of the larger SMT stuff too. Throwing some caps, leds, resistors, dc jacks, and regulators on for a first cook round will cut down on the soldering phase.

My hello world PCB was an ESP8266 carrier with an mcp23017 muxer and a bunch of buttons. This is an MQTT emission device which I will be using to assist in the controlling of a 3d printer. While web interfaces are flexible, some tend to put buttons too close and you can fairly easily crash the bed by clicking down instead of up in some cases.

Today's iteration is an esp8266 breadboarder. This allows 3v3 intake, has a TTL serial header (on the left) and a resistor + led combo on pin 14 for blink testing. The button at top right toggles into flashing mode and the bottom of the board breaks out 7 gpios onto the breadboard. The 3v3 and ground also have a header under the hot glue to the power rails on the breadboard. Very handy for testing a breadboard layout before designing the next PCB to have an ESP8266 pressed into it.

The breadboard side needs a little trimming back. Turns out the older breadboard I used to measure was wider than this one :o

Making PCB with a hobby CNC machine

One of the main goals I had in mind when getting a CNC "engraving" machine was to make PCB at home. It's sort of full circle to the '70s I guess. Only instead of using nasty chemicals I just have the engraver scratch off an isolation path between traces. Or so the plan goes.

My "hello world" board is the above controller for a 3d printer. This is a follow up to the similar board I made to help use the CNC itself. For a 3d printer I added buttons to set Z=0.1 height and a higher Z height to aid in homing. The breakout headers on the bottom right are for the ESP8266 daughter board. The middle chip is an MCP32017 gpio extender. I've had good experiences using TWI on the ESP8266 and the MCP overcomes the pin limitations quite nicely. It also gives all the buttons a nice central place to go :)

The 3v3 regulator makes the whole show a plug in the AA pack and go type board. The on/off switch is the physical connection to an external battery.

One step closer to the design in the morning, physically create in the afternoon, and use in the evening goal.

CNC Control with MQTT

I recently upgraded a 3040 CNC machine by replacing the parallel port driven driver board with a smoothieboard. This runs a 100Mhz Cortex-M mcu and has USB and ethernet interfaces, much more modern. This all lead me to coming up with a new controller to move the cutting head, all without needing to update the controller box or recompile or touch the smoothieboard firmware.

I built a small controller box with 12 buttons on it and shoved an esp8266 into that box with a MCP23017 chip to allow access to 16 gpio over TWI from the esp mcu. The firmware is fairly simple on the esp, it enables the internal pull ups on all gpio pins on the 23017 chip and sends an MQTT message when each button is pressed and released. The time since MCU boot in milliseconds is sent as the MQTT payload. This way, one can work out if this is a short or longer button press and move the cutting head a proportional distance.

The web interface for smoothie provides a pronterface like interface for manipulating where the cutting head is on the board and the height it is at. So lucky that it's open source firmware so I can see the non obfuscated javascript that the web interface uses. Then work out the correct POST method to send gcode commands directly to the smoothieboard on the CNC.

The interesting design here is using software on the server to make the controller box meet the smoothieboard. On the server MQTT messages are turned into POST requests using mqtt-launcher. The massive benefit here is that I can change what each button does on the CNC without needing to reprogram the controller or modify the cnc firmware. Just change the mqtt-launcher config file and all is well. So far MQTT is the best "IoT" tech I've had the privilege to use.

I'll probably build another controller for controlling 3d printers. Although most 3d printers just home each axis there is sometimes some pesky commands that must be run at startup, to help home z-axis for example. Having physical buttons to move the x axis down by 4mm, 1mm and 0.1mm makes it so much less likely to fat finger the web interface and accidentally crash the bed by initiating a larger z-axis movement than one had hoped for.