Shapeoko 2 E-Stop Button

[Eric Hubert]

After looking around there does not appear to be an estop button prewired for the shapeoko, shapeoko 2, or the grblshield (which controls the shapeoko).  We are going to have to build one ourselves.  I created this thread to coordinate that.

Luckily grbl  (the firmware that runs on the arduino controller) has support for e-stop, feed hold, start/resume cycle, homing switches, and limit switches.  I was able to find the pinouts yesterday. 

This image is from https://github.com/grbl/grbl/wiki/Connecting-Grbl.  We just need to make a circuit for the e-stop.  We should add buttons for feed hold, reset/abort, and cycle start as well.  We need more information to know if the feed hold, etc switches should be momentary or what.  The e-stop button instructables sells (https://www.inventables.com/technologies/e-stop-button) has two contact blocks.  One for normally open and one for normally closed.  This is ideal for our circuit. 

The spindle can be controlled from grbl as well.  Pin 12 will turn the spindle on or off.  We should put in a relay connected to a simple wall outlet type plug in order to cut the spindle power or turn it on.  This setup will work with any ac powered spindle such as the dewalt router or a dremmel.  We may also add a DC powered spindle with a speed control.

Since we aren't going to use cutting or cooling fluid on the shapeoko we can reserve coolant enable (analog pin 3) for vacuum enable if we get around to adding a vacuum attachment to the mill.  This should be attached to a relay/outlet combo just like the spindle on and off.  It can be powered by a shopvac type vacuum.  If we add a vacuum table bed we can control that separate.

We may or may not add a rocker switch for enable and disable steppers.  The main reason we may not want to is because if someone disables the steppers and then moves the machine it is no longer homed.  Without someone being aware of this they may assume "it will just know" where it is leading to some kind of problem.  More brainstorming should be put into this.

[Ryan Mcdermott]

Shouldn't e-stop be independent of the control system?

Could the e-stop work more like a breaker that cuts the power to everything (spindle included)?

--
You received this message because you are subscribed to the Google Groups "HeatSync Labs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to heatsynclabs...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Corey Renner]

Ideally, the e-stops do both.  They act on the control, but also directly on the actuators themselves.  That's the safest way.

cheers,

c

Blog:

http://coreyrenner.tumblr.com/

 

YouTube Channel:

http://www.youtube.com/user/vandal968

[david farmer]

Basically you need it to cut power to the spindle, cut the 24v power for the steppers and also send a pause for the gcode, so you don't ruin your work. If you can get a power strip with a breaker built into it that may be easy to hack it to be triggerable.

[Mike Bushroe]

Ryan,

  I have not worked at all on the Shapeoko so I am only guessing here.

SHORT ANSWER: yanking power is greatest assurance of stopping, but also requires most work to get going again, software stop may be almost as good and be faster to recover.

A 'kill all the power' button is extremely useful on large, powerful, manually controlled machines like the lathe and the mill. When you get to the table top machines you are rarely talking about human physical danger, just danger to the piece being built/worked, the tool, bits, or extrude in contact with the piece being worked, and the machine itself. So these 'disasters' are what you want to avoid or stop early and they should be balanced with the 'costs' of yanking the power immediately.
 
   From what I understand, if your 'red kill button' kills the 12V going to the stepper motors, then the software controlling the machine will completely lose track of where it is and need to re-zero every axis. This may not be a big deal compared to the above dangers, in which case this would be a good choice as it assures and instant stop to the damaging movement. On the other hand, if re-zeroing is a major hassle, then telling the software to halt is better because it does not lose position for each axis but will still stop if software is running, and if software has crashed, it is vary rare for the machine to keep moving, thus almost as safe as yanking the stepper motor power.

  If you take the more extreme measure of disconnecting the AC power in to get a guaranteed stop, then you have to reboot, reload, and re-zero everything to get started, otherwise same comments as above.

   The correct answer is a balance of the risks and costs. Having never worked on the Shapeoko, I can't make any recommendation. I am just trying to point out that there are alternatives to implementing the safety red stop button on a CNC machine exactly the same way as a manual machine.

Mike

[Corey Renner]

There are really two different types of stops that are often referred to as e-stops.  There is a "Feed Hold" where you tell the controller to stop moving the axes, and there is a true e-stop that cuts everything.  Resuming from a "Feed Hold" is as easy as clicking resume, and sometimes having to restart the spindle depending on the configuration.  A true "e-stop" cuts everything typically in multiple places.  Spindle is cut, motor drives, etc.  Resuming from a true estop is a little more involved, but if the machine has an accurate way of homing (and was homed prior to the interrupted run) it is trivial.  Mechanical home switches such as the ubiquitous grey/blue Westinghouse ones are repeatable to about 0.0004" and decent inductive prox switches can do better than that (mine are good to 0.0002" ~$20/ea on eBay).  Three axes, three decent switches (assuming that none already exist) and you're looking at $60 or less to be able to accurately recover and resume from a true estop.  Worthwhile from a safety standpoint IMHO.

cheers,

c

Blog:

http://coreyrenner.tumblr.com/

 

YouTube Channel:

http://www.youtube.com/user/vandal968

[Eric Hubert]

From what I have been reading it is better to hook the stepper power source through the estop button as opposed to through a relay.  http://www.shapeoko.com/forum/viewtopic.php?f=7&t=3045#p22826  I am looking at some of the components needed to outright make this board but I haven't found them in our electronics section.  I am attaching the eagle files provided in the article with this post.  If anyone more still in electronics could substitute components in the design with stuff we have in the lab or donate components It would be greatly appreciated.  Let me make an attempt to list a bunch of things I know little of.  I found a script, bom.ulp, that gave me this output.

Partlist exported from C:/Users/Owner/Documents/eagle/ShapeokoBoard/eagle/arduino grbl.sch at 7/3/2014 7:18:28 PM

Part    Value        Device       Package    Description                           MF                      MPN        OC_FARNELL OC_NEWARK

OK1     4N37         4N37         DIL06      OPTO COUPLER                          FAIRCHILD SEMICONDUCTOR 4N37       1612456    72K9948  
OK2     4N37         4N37         DIL06      OPTO COUPLER                          FAIRCHILD SEMICONDUCTOR 4N37       1612456    72K9948  
OK3     4N37         4N37         DIL06      OPTO COUPLER                          FAIRCHILD SEMICONDUCTOR 4N37       1612456    72K9948  
A1      22-23-2081   22-23-2081   22-23-2081 .100" (2.54mm) Center Header - 8 Pin  MOLEX                   22-23-2081 1756826    01C7592  

A2                   22-23-2061   22-23-2061 .100" (2.54mm) Center Header - 6 Pin  MOLEX                   22-23-2061 1462922    27C1933  

D1      22-23-2081   22-23-2081   22-23-2081 .100" (2.54mm) Center Header - 8 Pin  MOLEX                   22-23-2081 1756826    01C7592  

D2      22-23-2101   22-23-2101   22-23-2101 .100" (2.54mm) Center Header - 10 Pin MOLEX                   22-23-2101 1756828    94C0765  

D3      1N4148DO35-7 1N4148DO35-7 DO35-7     DIODE                                                                                        

D4      1N4148DO35-7 1N4148DO35-7 DO35-7     DIODE                                                                                        

D5      1N4148DO35-7 1N4148DO35-7 DO35-7     DIODE                                                                                        

D6      1N4148DO35-7 1N4148DO35-7 DO35-7     DIODE                                                                                        

LED1                 LED3MM       LED3MM     LED                                                                                          

LED2                 LED3MM       LED3MM     LED                                                                                          

LED3                 LED3MM       LED3MM     LED                                                                                          

PWR                  AK300/2      AK300/2    CONNECTOR                                                                unknown    unknown  

R1      3k           R-US_0207/7  0207/7     RESISTOR, American symbol                                                                    

R2      3k           R-US_0207/7  0207/7     RESISTOR, American symbol                                                                    

R3      3k           R-US_0207/7  0207/7     RESISTOR, American symbol                                                                    

R4      100k         R-US_0204/7  0204/7     RESISTOR, American symbol                                                                    

R5      100k         R-US_0204/7  0204/7     RESISTOR, American symbol                                                                    

R6      100k         R-US_0204/7  0204/7     RESISTOR, American symbol                                                                    

R7      10k          R-US_0204/7  0204/7     RESISTOR, American symbol                                                                    

R8      10k          R-US_0204/7  0204/7     RESISTOR, American symbol                                                                    

R9      10k          R-US_0204/7  0204/7     RESISTOR, American symbol                                                                    

REMOTE               AK300/4      AK300/4    CONNECTOR                                                                unknown    unknown  

SPINDLE              AK300/3      AK300/3    CONNECTOR                                                                unknown    unknown  

X1                   AK300/3      AK300/3    CONNECTOR                                                                unknown    unknown  

X2                   AK300/3      AK300/3    CONNECTOR                                                                unknown    unknown  

X3                   AK300/3      AK300/3    CONNECTOR                                                                unknown    unknown  

[Eric Hubert]

Progress has been made on the shapeoko.  Let me sum up what has been done in the last few months.  Since I had no money to put into the project it took a while to secure the simple funding for the supplies needed.  That has been accomplished.  The final design for arduino shield is done (I've attached the eagle file).  The shield has been milled to the new GRBL .9g pin layout that allows for variable spindle speeds.  The pictures are of the milling process and the first board which failed. 

The components are soldered and the board is connected to the UNO on the shapeoko.  Here is the populated board not connected.  I tried to use the vinyl cutter to do some text for each pin but it failed terribly.  So I put the backing for the vinyl on the board with some double sided tape.  It's better than nothing.   Does anyone know if we have new sharp knives for the vinyl cutter? 

The limit switches have arrived and the first of the mounts for the switches have been designed in Solidworks and have been mounted.  The new version of GRBL, the g-code controller firmware on the UNO, allows for variable spindle speed control via PWM and has a probing option.  I will see what I can do to hook up a probe for us to try out but that will be after I get the homing/limit switches install so we can use this thing.

[Eric Hubert]

I got a little more done on the shapeoko.  All the more steps closer until we can "unveil it".  It took a lot of creativity and about twelve hours of solidworks time and 3d printing but I made a mount so the z-axis has a limit switch at the top of it's motion.  There's nothing at the bottom.  The x and y axis have mounts installed but I am short two limit switches so only the maximum travel distances have limit switches at the moment.  For some reason GRBL choses to home to the max travel so that every machine coordinate is negative.  There must be some reason for this but I don't know it.  Homing was tested in GRBL and it works.  Repetition testing would probably be a good idea.

I recompiled the firmware and flashed GRBL version .9gHeatSync to the arduino controlling the shapeoko.  The only real change was adding our name and enabling the PWM for the spindle control. 

New mounts were make for the uno.  It is now attached in a more clean manner to the shapeoko.  I also created a probe which will make setting tool lengths much easier.  GRBL now supports G38.2 and G43.1 for probing.

[Jasper Nance]

Wow Eric, you are so cool!!

Thanks for all of your hard work getting this really cool resource functional <3

--

You received this message because you are subscribed to the Google Groups "HeatSync Labs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to heatsynclabs...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
----------------------------------------------
Jasper Nance - KE7PHI
Creative and Scientific Imagery
http://www.nebarnix.com/

[Eric Hubert]

I finished adding all the limit switches to the shapeoko except for the z-.  The problem with that one is I am having the hardest time finding a place to mount it.  With the way the table is setup (currently) the tool will bury itself through the table before the limit switch is even triggered so, _for now_, it will work.  GRBL has pins for start/resume cycle (start or resume the job), feed hold (stop tool motion but keep spindle on), and E-Stop (emergency stop).  The green start/resume cycle button and the red E-Stop button are installed.  Adafruit was out of the yellow buttons so the feed hold is not installed.  They have them in stock now.  The mill will function without many of these buttons but the E-Stop was very important.  The red button which was sent is defective.  It stays down when pressed.  I'll swap the green button to the e-stop until we can get this sorted out.  The next thing to do to get this ready for teaching classes is make something to hold the buttons and get a more stable computer to interface with the shapeoko.  In the future I'd like to machine a proper probe and a proper z-height indicator to make things easier for beginners, hook up a vacuum attachment with a Hepa filter, and create a vacuum table for it for easier work piece holding to name a few.

[Eric Hubert]

The "enclosure" to hold the buttons is done.  The picture I took of it didn't show up on the flickr feed and I am at home.  Due to this I won't be able to post it here.  The "enclosure" is a hacked together mdf laser cut traffic light looking thing attached to the right of the shapeoko.  The yellow button is not installed yet.  The red button still sticks but it sticks down which still makes it functional as a stop button.

[Mike Bushroe]

Eric,

   lack of a bottom Z-axis stop doesn't sound good. How about one of those spinning laser levelers with a red filtered detector to detect when the tool body gets down to or below a certain point? it might mean adding a plastic or particle board fence around the bottom to keep the stray laser light from bouncing around the lab, but it would allow an adjustable stop height without needing something mechanical on the cutting head.

Mike

[Eric Hubert]

The main problem with an optical solution is dust from the cutting process.  I'd like a bottom z-limit switch too.  Do we have any other suggestions?

[Joe Modjeski]

The challenge is the waste board, in addition to tool length.

We could probe the waste board, but even then the end stop would have to allow for some depth beyond it.

I've wondered if there is a way to sense when a stepper stops moving.  Maybe current draw difference during a probe cycle?

Part of me thinks the solution is extremely hard, otherwise the big boys would have figured it out.  Crashing the tool seems to be part of the "milling experience".

This is one advantage to additive process.  We only care about zero at the start.

What about breakaway clamps for the dremel?  Those could be rigged to allow the tool to break away and shut off the power to the tool.

Just brainstorming...

Sent from my iPad

On Oct 13, 2014, at 9:21 PM, Eric Hubert <eric....@gmail.com> wrote:

The main problem with an optical solution is dust from the cutting process.  I'd like a bottom z-limit switch too.  Do we have any other suggestions?

--

[fixitquickservice]

The most accurate and probably one of the easiest ways to get high accuracy but keep the cost down would be to use a piezoelectric bump stop on an arm that travels with the Dremel. A mechanical wire could work also by being rigged off of the Dremel holder(think bug feelers). Use the electrical signal the piezo generates as a trigger to shut down the machine via a programmed arduino voltage limit.

Peizo's are what make the ferro arms work and is used when more accuracy than wavelengths of light from lasers can offer.

Working with one of the contractors that helps design partical accelerators controls for a high accuracy platform taught me that I when not confined to uber small collision detection chambers the limits for travel in every axis are always going to be thermal expansion of the bed and tools. Laser based optical measurement sensors are limited by optical wavelengths. Gaining more accuracy through controlled dithering of the lasers and eventually in combination with many micro piezo sensors mounted on tracks(think vinyl record) to send signals as encoders. Precision install of peizo's can be done but temperature differences in the bed and tooling will effect the accuracy thus the need for lasers which wavelengths of light go uneffected by temperature swings.

Of course the Shapoko doesn't need to maglev particles at the speed of light but if you wanted to set the bar high for accuracy there is a way.  

Anything more than a piezo stop is not going to be neccessary if using a good peizo sensor it can give 0.0001" accuracy or higher in some cases. Z axis resolution with the right platform could be programmed to stop before you broke through the copper clad if milling a pcb if the stop was zeroed out at the copper surface. 

Limitations on the shapoko z axis resolution are going to be clock cycle speed receiving the signal to process the stop signal and the quality of the peizo. Flex shouldn't matter if the peizo can trip the stop with less force it takes to flex anything.

This method requires zeroing out the z axis before each use and if the peizo reaction time to stop with the current build platform is a travel of say 2 mil. Throw down thin disposable 3mil and the bed will never see a bit plunge.

[Jasper Nance]

Are these the same force sensors that all the delta printers are putting under the beds to auto level?

[Eric Hubert]

I am a bit confused.  This is always a problem with a milling machine.  Each tool is a different length.  So if I put a limit so that one tool can not reach the bed then another tool could and a third tool wouldn't be long enough to reach the bed.  Would what was proposed be able to address this?  I do have to mention I like the idea.  I also have to mention that the controller (GRBL running on an UNO) has dedicated pins which when put low trigger the limit switches.  I hope that helps with any electrical evil schemes people are thinking of.

[Larry Campbell]

If it was mine I would swage limiting rings on the tooling so that all the tools WERE the same depth of cut... bit of a hassle with tool usage and breakage and people wanting to put in unswaged tools however that would somewhat solve your issue.

otherwise you need to make a feeler on an adjustable rod so that you could set it to depth after a tool change. 

On Tuesday, October 14, 2014 12:05:38 PM UTC-7, Eric Hubert wrote:

I am a bit confused.  This is always a problem with a milling machine.  Each tool is a different length.  So if I put a limit so that one tool can not reach the bed then another tool could and a third tool wouldn't be long enough to reach the bed.  Would what was proposed be able to address this?  I do have to mention I like the idea.  I also have to mention that the controller (GRBL running on an UNO) has dedicated pins which when put low trigger the limit switches.  I hope that helps with any electrical evil schemes people are thinking of.

The information in this message may be confidential.  It is intended solely for

the addressee(s).  If you are not the intended recipient, any disclosure,

copying or distribution of the message, or any action or omission taken by you

in reliance on it, is prohibited and may be unlawful.  Please immediately

contact the sender if you have received this message in error.

[Eric Hubert]

I looked into the tool which sets stop rings on 1/8th inch tools.  It was like 700 bucks!  There is no reason it should cost 700 bucks.  I am in total agreement with you Larry.  Stop rings would be ideal.   Here is a guy who made his own.  One can see the stop rings on his tools in the first picture.  We have some of these type of tools around the shop.  http://www.5bears.com/cnc35.htm  In GRBL .9g probing has been enabled.  This means we can hook a conductive wire to the tool using a clip and have a conductive object of a set height on the bed/top of work piece/etc which we can use to "touch off" and zero the tool.  Here is an example off ebay.  http://www.ebay.com/itm/NEW-Z-Axis-Auto-Zero-Touch-Plate-CNC-Router-Mill-Mach3-instructions-included-/281464709353?pt=LH_DefaultDomain_0&hash=item41889aa0e9   I am not sure if this is what you were referring to.

[Mike Bushroe]

Aaron,

 fascinating reading and much detail on an approach I had not heard of before. I assume that the piezos you are talking about are micromachined crystals that provide much the same function as micro-switches, but with a few orders of magnitude smaller slop. I can see where a micr-machined piezoelectric with amplifier stage could contend with the wavelength limits of visible light.

  The problems here are that we are looking for accuracies  of around 0.1mm, not 100 angstroms, and that the tool to tool length differences will 10 to 50 times the repeatability requirements. So having a single bottom Z-stop accurate to a 1/4 wavelength but that does not accommodate tool wear or tool differences doesn't help much. I also wonder about pricing on such ultra-high precision devices.

  But to follow something along those lines, if there was a microswitch or opto switch on the side of the work area that the software could slowly run the tool (not spinning hopefully) down until it trips the switch. The use a small stepper and drive screw to move the real Z-axis bottom stop that is mounted on the Z-axis slide so that it just strips at that distance too. That way you can re-calibrate for every tool change, and even check from time to time for tool wear. And still stick with inexpensive, commonly used and available parts.

Mike

[Larry Campbell]

That ebay touch plate is mega overpriced, you can duplicate that by soldering a wire to a piece of sheet metal and gluing it to a block of plastic

[Eric Hubert]

Even though I believe it is machined out of something like durian I also believe it is pricey (unless we can't make one ourselves).  I can see us needing three or four of these touch plates so being able to make them ourselves would save some money.  We'd need one for the big mill, shapeoko, the shurline (once it is running), and the eventual build/acquisition of a cnc router.  Having a precision, repeatable, easy to use, and (almost) fool proof way to z out a tool is of great value.  I made a "touch plate" of a solid scrap of aluminum and an alligator clip to test the concept.  Since the solid aluminum doesn't give we will break/wear tools over time.  Having one with some give will save us from that.  So, how would we make one if we were so inclined?

[fixitquickservice]

I know wavelengths of light are more than enough for the accuracy needed for the shapeoko. A clock cycle crystal could be used as a predictable laser pulse timer that could control the lidar (laser radar) with an optosensor to determine distance from the sensor to the table off of the frequency much like the arduino sonar modules except without the bounce. I was thinking something along the lines of something simple like a clamp that holds a pencil in a compass but instead holding a a rod with a flat tab on the end with a pressure sensor or other touch down that can pivot in front of the bit and plunge it until it touches and can be swung back out if the way of the bit or completely removed after calibration. Knowing the tabs thickness can be used as a tear thickness calculation like scales use tear weight to get the containers weight deducted so you only get charged for the items in the container. Use this data with a pre written program to auto fill the tear distance amount with the referenced laser measurement at the time it touches down and you got an automated non contact emergency stop program with laser accuracy that can be used with different bit lengths accounting for tool wear included. The laser readings would give the trigger signal once it's relative distance to the end of the bit is known so the touch sensor wouldn't be needed until the next bit change out. 

The parameters for what distance to safely stop everything would merely need to be set in the programming and will be limited by the stepper motors resolution.

As far as making sensors nano piezoelectric powders are available pretty cheap. In the past I've used a mixing beaker with a magnet stirrer to mix evaporative solvents with polymer resins to bind the peizopowder ranging from 10-40% solids depending on the polymer, intended product strength, and if the strength is needed doping with conducting powders may also be needed to get energy transfer through the thicker resin gaps between the piezo particles. 

Pour or press into molds and you can have custom shaped piezo crystals. Bake if needed to cure the resin.

Make thinner solutions of 10% solids or less for spraying thin coatings on conductive surfaces to make them produce a voltage when pressed can be done as well. 

Haven't gotten to play with producing the thin film sensors yet but I intend to.  

If others are interested in what kind of results we can produce with a little experimenting maybe we make a touch off that is comparable to the high end sensors. I can check to see how much the powder is again and at what quantity as it has been a while since I last bought some but I think it was reasonable enough to justify buying some if others think they could use it at the lab. 

Last time I used 2% n methyl 2 pyrolidone with the remaining liquid being lab grade 99% isopropyl alcohol but that was for a expensive high strength polymer producing rather thick products that I had to ensure strength wouldn't be effected by off gassing during or after the full curing bake cycles. N methyl doesn't like to bake out thus only 2% solution but it's a great solvent and alcohol evaporates really well. So if thick crystals are made and the binding polymer will dissolve in n methyl or alcohol this is a great way to do it. 

Also piezo's are used in reverse  like peltier coolers. You can add voltage and make the piezo shrink and grow as linear actuators with pretty impressive force instead squeezing making a voltage. This is what they use for actuators for leveling precision tooling(particle collider pathways) and other amazing precision machinery.

Cheers

[Eric Ose]

Just curious. Who is fixitquickservice?

Eric Ose
azrobotambassador.com

[Eric Hubert]

I was looking for a probing example to post here and come across this.  It's not related to the estop or limit switch but would be a wonderful addition to the mill.  It's a digitizing probe.  It can be used to do positioning of stock on the mill.  However, it can also be used probe measurements of things on the mill.  I've seen people probe coins such as quarters with ridiculous accuracy.  Aaron, how difficult would this be to make using something more accurate than a nail and foam? 

http://www.instructables.com/id/cheap-super-sensitive-digitizer-probe-for-cnc-/

[fixitquickservice]

This is Aaron :)

Correct me if I'm wrong I know there are a few Eric's at HSL but I believe I met you a few times but most recently at the official HSL Vine social experiment. I am the guy that was speaking to you and the drunk russian aerospace engineer student about the heat control issue with the fighter jets at Mach speeds. 

I have the crazy car project I'm working on with the peltier coolers and other fun stuff.

[aaron carlson]

Making something more sensitive would most likely be the case a most mass production techniques don't make use of nanotech.

That digitizer is a great example of how something so cheap and basic can be made with existing cheap push sensors.

Amplification is usually needed to make it a more usable trigger but usually isn't a big problem.  Less contact area is going to squeeze less surface area thus producing a smaller voltage so making a thicker piezo is a good thing as it makes more usable voltage with a smaller area or said another way it's more sensitive for the same reason.  Sensitively can be gained for a thicker piezo design.

Steal the needle and arm off a record player and you can have the accuracy of that digitizer or probably much better considering the resolution of the bumps are easily read off of the vinyl records. Those are old school stones not nanofilm and they still work great due to how big the stones are.

On Oct 16, 2014 12:11 AM, "Eric Hubert" <eric....@gmail.com> wrote:

I was looking for a probing example to post here and come across this.  It's not related to the estop or limit switch but would be a wonderful addition to the mill.  It's a digitizing probe.  It can be used to do positioning of stock on the mill.  However, it can also be used probe measurements of things on the mill.  I've seen people probe coins such as quarters with ridiculous accuracy.  Aaron, how difficult would this be to make using something more accurate than a nail and foam? 

http://www.instructables.com/id/cheap-super-sensitive-digitizer-probe-for-cnc-/

--

[Eric Hubert]

That Eric was possibly Eric O.  I am the tall, blondish Eric  who enjoys the shop.

[Eric Ose]

Aaron,
Yes that was me. I remember your crazy car project and of course the drunk aerospace engineer.

Eric Ose
azrobotambassador.com

That Eric was possibly Eric O.  I am the tall, blondish Eric  who enjoys the shop.