Build Thread- Upgrading an old (1993) Minitech Mini-Mill 2.

[LegendCJS]

Hello all:

The short version: My lab has an old Minitech Mini-Mill 2 from 1993 (?) that is in an almost pristine condition except for a heavy layer of dust. The story goes that it never received much use because the steppers on it were just too weak to make it do more than carve words into wood and plastic.

A couple weeks ago I started to under take the task of reviving the mill. All documentation has been lost, as well as all tooling. The only thing we had was the mill itself. I got the trial version of Mach3, and a brand new computer, and went to work. The early version of my trials can be found in the Mach3 forum under the thread: "Need Help!- Mach3 with old (1993) Minitech Mini-Mill/2"

We succeed in getting around the low current issue with modern parallel ports by abandoning Mach3, and trying out EMC and setting it for 50us pulses. I think we will stick with EMC2 because I really like its HAL abilities. The Hardware Abstraction Layer and the software programmable logic controller should allow me to do some fancy stuff in software like programming the drivers to only allow jogging away form triggered limit switches. Plus EMC is open source, and I like to save money (even if its not my money)

Anyway I decided to start this thread because the rest of the project will have nothing to do with Mach3. I plan an asking advice about new steppers and drivers and a general upgrade of the electronics for the mill. And I will also document the progress.

Things accomplished so far:
1) The mill has been disassembled, cleaned, lubricated, and reassembled. After some research I decided that any kind of generic lithium grease was good for the lead screws, and we used way oil (Lubiplate No 3.) on the rails.

2) Decided that the current servos are too puny. They are a 16+ year old design (round) and the specs follow:

Superior Electric "SLO-SYN" Type: MO63-LS-502
rated for 4.6Amps, (2.25v - not sure what the voltage rating means on steppers...) and only 150 oz-in of holding torque.

3) Began searching for a new set of drivers, a descent breakout board so I can add an E-stop and limit switches, and stronger steppers. In the next post I will describe my proposed shopping list and ask people if they foresee any problems or incompatibilities.

Thanks everyone for following my thread

[LegendCJS]

I plan to get three new steppers, three new drivers, an isolated breakout board, and e-stop and limit switches.

For my Z and Y axis which need to move the most weight (and are the easiest to stall with slight hand pressure with their current steppers) I plan to get: two RS23-570s (570 oz-in rating for homeshopcnc.com). I would get one for my X-axis but I don't want to have such a heavy motor (4.2 pounds) added to the load of the Y-axis if I can avoid it. Plus the X-axis has an easier job to do anyway.

For the X-axis I plan on purchasing an 23-305-DS8A 425oz-in bipolar rating stepper form hobbycnc.com. I am reasonably sure it weighs less than the 570 even though no weight is listed on the hobbycnc.com site, so it makes it a better choice for the X-axis.

The 570 requires only 2.5 amps per phase in bipolar series configuration, and the 23-305-DS8A requires only 3amps (I don't know if that is per phase or total).

Each motor has dual shafts so I can attach hand wheels for manual operation easily if I wish.

To drive the three motors I plan on getting three ProboStepVX drivers. They are rated up to 3 amps per phase. I know form reading other posts on this site that you have to be careful with your heat sinking and ventilation when running ProboSteps higher than 1.5amps, but I'll take care here. I actually want to put a temperature safety switch on my driver chips in series with my E-Stop switch to shut it all down if things are getting hot.

Providing all the signals to the driver chips will be the BOB form Probotix: the PBX-RF isolated break out board. And I'm aware of the nice wiring options for connecting the BOB to the drivers that Probotix also sells. I will also buy my E-stop switch form them.

One area I am insecure about is where to find good repeatable limit switches. I'm aware that it is desirable to have a levered switch maybe with some kind of roller on it so there is still some travel room left for the stage to come to a stop after the switch is triggered, but that is the extent of my knowledge.

So my questions to the community are: Does anyone have any experiences or suggestions with the steppers or electronics I listed above that would be useful to know about before making a purchase? Does anyone have any good alternatives? The travel on the mill is very limited compared to the average CNC router, and as such I am not worried about getting the high speed torque out of my 570s. So the inability to run them in bipolar parallel at 5 amps a phase does not bother me. Does anyone know of an off the shelf heat sink for the SLA driver chips in the ProboStep drivers? Is anyone aware of a temperature based switch that cuts off at a given temperature so I can make my own thermal protection circuit? And does anyone have any limit switch sources to recommend or knowledge to share?

If people are wondering, the reason I am not going for a Gecko 540 is that all the components listed above will cost me under $400, and a single Gecko 540 will eat the majority of that budget and leave little room for steppers and switches. Plus I don't foresee the need to have a 4th axis at any point.

Thanks again everyone for reading my lengthy posts and I will really appreciate any feedback.

[Fixittt]

you have some options,
One is a xylotex 4 axis board. I have had some great results with this board once you solve the resonance issues. A little under powered but for wax work it was ok.

but the type of steppers and drives are going to depend on what your plans for the machine are. Are you needing high resolution? high speed?

You need to determine what your end results are first then go with the setup that best matches it. If your wanting to do say like microfluidics, then your going to want a drive with higher microstepping and probably .9 degree steppers. If your going to cut out general shapes. Or engraving then maybe the 1.9 degree steppers and 8 to 10 microstepping will be enough.
I will help you in any way I can to solve the issues.
Robert

[jalessi]

LegendCJS,

The ProboStepVX driver is unipolar so the torque output of the stepper motors will be quite a bit less than the published bipolar ratings.

This wont be a big concern however since the Minitech has 10 tpi screws so you will have plenty of torque with the 23-305-DS8A motors, 305 ounce unipolar rating with be more than enough for all axis.

The larger RS23-570 motors will be complete overkill and the higher rpm limits may suffer if you decide to use them.

If you are looking for a very inexpensive 3 axis unipolar driver why not solder together a HobbyCNC ez driver, its only $64.00 and it has a built in breakout board.

The HobbyCNC kit uses the same driver IC chips as the Probotix unipolar drivers by the way.

http://www.hobbycnc.com/products/hob...ver-board-kit/

These micro switches work well>

http://tinyurl.com/kkxwqx

Jeff...

P.S. Fixittt is a Minitech expert, his experience and help will be invaluable.

[jalessi]

LegendCJS,

I neglected to mention if you decide to use the RS23-570 motor's you will need to bore out the stepper motor couplers.

The RS23-570 motor's have a .375 inch shaft.

The stock thrust bearing on the Minitech will not fit the RS23-570 motor either, they will need to be replaced.

Jeff...

[jrc347]

I'm pretty sure I have the same machine you are describing. I've done alot of modifications to the machine which I think are invaluable to the usability of it. The first mod was to get rid of the original steppers as you mentioned. I went with Step Syn 1.8 degree, 3.0a steppers that I got off of ebay for a great price. I don't really remember exactly what I paid, but it was less than $150 for all three. I'm not really sure of the torque of the motors either ( the number on the motor is: RM26A3D) but they are more than enough for the mill and I have never lost another step after the conversion. The motors are driven by a Nudrive 4 axis controller bought off of ebay for $500. The motors are 6 wire bipolar steppers and I think have 180 ozin of torque. They are very similar to these motors on ebay: item number 120413334314. And it just so happens that there is a nudrive controler on ebay right now too, auction number :290304446179. This combo has proven itself for the last 4 years cutting mostly aluminum. I also upgraded the spindle with a sherline 10k rpm unit that cost me another $500. Obviously I have spent a little more than your projected budget of $400, but It has been very much worth it in the long run for me. I've made countless "things" on the machine, and after 4 years its still going strong but showing some signs of wear and tear. My normal cuts are in the 26 ipm range and a very conservative .020 doc. But on some occasions (operater error) the doc was increased to a very lively .125 or more with no lost steps or other issues other than the spindle not holding up to the increased cutting pressures.

[LegendCJS]

Thanks everyone for reading and posting in this thread with all your helpful advice. I didn't know the ProboStepVX controllers I was wanting to get were unipolar only. I had assumed it was possible to take any controller and make it drive a stepper any way you wanted by just changing the wiring. So with that new information and the reassurances that 300+ oz-in of torque will be plenty with 10 tpi screws I think I'm going to go with the 23-305-DS8A or something very similar for all three axes.

The lab digital camera seems to be missing, and I keep forgetting to bring my own camera form home so no pictures yet, but soon I promise. And the Minitech people are being very helpful. They seem so happy that this 16 year old machine of theirs is going to be back up and running in support of an Electrical Engineering class at MIT that they are sending me some electronics to help out with the resurrection. I'll post more about it all when it gets going.

Along with selecting new hardware for the mill I've also been learning cam programs for it. Since the target audience won't have a mechanical engineering undergraduates experience with cad programs like Solidworks or whatever, I've been searching for a simple and user friendly set of software. I currently really like CamBam and plan on getting the full up license for it if nothing better comes along. Could anyone who knows answer this question: once you've gotten CamBam pro do you get upgraded versions for free? And secondly: what is an easy to use 3D modeling program suitable for creating 3D files for importation into CamBam (STL, 3DS and RAW formats)?

The goal of the mill resurrection is to allow electrical engineering undergraduates to exercise some design creativity in a hardware sense by helping them make custom electronics enclosures for their EE projects. Everyone can imagine the standard EE class project leaves you with some circuit board based gizmo that does some neat trick, but is actually not very visually appetizing and is a fragile collection of components and wires that gets you flagged at any airport when you try to take it home and show your family (especially Logan airport.) Currently the only way around that is some standard off the shelf blocky enclosure.

The outline of the plan is to use the mill to carve single use (or low use) molds (i.e. could be wood, plastic, plaster...) These molds would then be used to form "clamshell halves" of their case designs on a thermal/ vacuum forming rig our of a sheet of plastic of their choosing. Then we cut the shapes out of the sheet of plastic and snap them together (if possible) or use rivet/jack nuts to give the thin shells threads and screw them together (more likely).

So imagine your an undergrad excited about taking this project class and you want your case to look just like an ipod- what would be the easiest/best program to use to make your 3D design in for importation into CamBam? After you've made your model the the future TA of this proposed class will be the only one to actually run the mill and will be in charge of approving any gcode or doing the gcode generation.

Thanks again for reading my posts.

I also ask the question about good tooling and raw material for these thermal/vacuum forming molds I want to make. Cheapness is ideal for the material- would something as flexible as dry bags of plaster powder that is mixed and set in the classroom make a good mold for this kind of thing? And I'll get those pictures up- both before and after. (Sorry the mill as already been cleaned- so I can't show anyone a picture of a 16 year old layer of dust...)

[LegendCJS]

A quick question for everyone:

Are there any sources of sub $50 NEMA 23 stepper motors with over 300 oz-in (unipolar) of torque that isn't using freaking PayPal for payment??

That rules out HobbyCNC.com, Keling, and of course all of Ebay. And HobbyCNC and Keling are the only places I've found with motors of the above description in the price range ($49 each)

The problem is that we (the professor and all his graduate students over the years) use a communal lab credit card for buying things online in our lab, and PayPal won't let the same credit card be used on more than one account. Someone must have (in years past most likely) made a paypal account on the fly to order something and we have no idea what the log in information or account is. And PayPal's "security" policy prevents use of this credit card in any other account until it is removed from the mystery account.... So good bye paypal and good bye to any business with paypal users....

[jalessi]

LegendCJS,

Did you call John at Keling, he may take a credit card over the phone.

630-254-8655

Jeff...

[escott76]

Originally Posted by LegendCJS Thanks everyone for reading and posting in this thread with all your helpful advice. I didn't know the ProboStepVX controllers I was wanting to get were unipolar only. I had assumed it was possible to take any controller and make it drive a stepper any way you wanted by just changing the wiring. So with that new information and the reassurances that 300+ oz-in of torque will be plenty with 10 tpi screws I think I'm going to go with the 23-305-DS8A or something very similar for all three axes. The lab digital camera seems to be missing, and I keep forgetting to bring my own camera form home so no pictures yet, but soon I promise. And the Minitech people are being very helpful. They seem so happy that this 16 year old machine of theirs is going to be back up and running in support of an Electrical Engineering class at MIT that they are sending me some electronics to help out with the resurrection. I'll post more about it all when it gets going. Along with selecting new hardware for the mill I've also been learning cam programs for it. Since the target audience won't have a mechanical engineering undergraduates experience with cad programs like Solidworks or whatever, I've been searching for a simple and user friendly set of software. I currently really like CamBam and plan on getting the full up license for it if nothing better comes along. Could anyone who knows answer this question: once you've gotten CamBam pro do you get upgraded versions for free? And secondly: what is an easy to use 3D modeling program suitable for creating 3D files for importation into CamBam (STL, 3DS and RAW formats)? The goal of the mill resurrection is to allow electrical engineering undergraduates to exercise some design creativity in a hardware sense by helping them make custom electronics enclosures for their EE projects. Everyone can imagine the standard EE class project leaves you with some circuit board based gizmo that does some neat trick, but is actually not very visually appetizing and is a fragile collection of components and wires that gets you flagged at any airport when you try to take it home and show your family (especially Logan airport.) Currently the only way around that is some standard off the shelf blocky enclosure. The outline of the plan is to use the mill to carve single use (or low use) molds (i.e. could be wood, plastic, plaster...) These molds would then be used to form "clamshell halves" of their case designs on a thermal/ vacuum forming rig our of a sheet of plastic of their choosing. Then we cut the shapes out of the sheet of plastic and snap them together (if possible) or use rivet/jack nuts to give the thin shells threads and screw them together (more likely). So imagine your an undergrad excited about taking this project class and you want your case to look just like an ipod- what would be the easiest/best program to use to make your 3D design in for importation into CamBam? After you've made your model the the future TA of this proposed class will be the only one to actually run the mill and will be in charge of approving any gcode or doing the gcode generation. Thanks again for reading my posts. I also ask the question about good tooling and raw material for these thermal/vacuum forming molds I want to make. Cheapness is ideal for the material- would something as flexible as dry bags of plaster powder that is mixed and set in the classroom make a good mold for this kind of thing? And I'll get those pictures up- both before and after. (Sorry the mill as already been cleaned- so I can't show anyone a picture of a 16 year old layer of dust...)

Have a look at pourable urethanes. I use a bunch from smooth-on, but there are several others. Their Task line is really nice for stuff like this. You can get "molded" plastic results if you plan right. I cut molds out of plastic (delrin) and with a release agent they come out really nice.

[LegendCJS]

Originally Posted by jalessi LegendCJS, Did you call John at Keling, he may take a credit card over the phone. 630-254-8655 Jeff...

I did call him and I learned that they have a normal credit card payment option that doesn't require having a paypal account. So now I have three KL23H286-20-8B steppers on the way. Thanks for the advice about the drivers I was interested in being unipolar only. And thanks for the phone number- it saved a lot of time.

To escott76: Can you tell me if you think pourable urethanes will work for a classroom situation? If it takes lots of time and skill to get right then that moves the job form the students to the TA. But if its an easily teachable operation then I guess the time goes into making the molds. How would you compare delrin as a mold material in relation to any other mold materials you may have experimented with? The molds for our application don't have to be particularly reusable, and it would be ideal if they were still workable with hand tools after the cnc stage- which leads me to think something like plaster is ideal. Remember these are not Mechanical engineering students, they are Electrical Engineering students, so they won't have even basic machine shop skills, and programming/ watching the cnc mill and using hand tools to fix up any mistakes in their molds will be about all that we want to allow them to do.

Thanks

[escott76]

Originally Posted by LegendCJS To escott76: Can you tell me if you think pourable urethanes will work for a classroom situation? If it takes lots of time and skill to get right then that moves the job form the students to the TA. But if its an easily teachable operation then I guess the time goes into making the molds. How would you compare delrin as a mold material in relation to any other mold materials you may have experimented with? The molds for our application don't have to be particularly reusable, and it would be ideal if they were still workable with hand tools after the cnc stage- which leads me to think something like plaster is ideal. Remember these are not Mechanical engineering students, they are Electrical Engineering students, so they won't have even basic machine shop skills, and programming/ watching the cnc mill and using hand tools to fix up any mistakes in their molds will be about all that we want to allow them to do. Thanks

Read more about the specifics of the materials I was talking about at www.smooth-on.com . A & B mixed together, generally by volume which is the easiest. Anywhere from 3 min to an hour and a half demold, with various times in between. You can pigment them as well.
Plaster isn't going to be the best stuff to machine. What will come off is a fine dust that you do not want to get into your ways, screws or spindle bearings. It is IMO a horrible material to use for machining purposes and will cause more trouble than it saves you in "cheapness". I use delrin because it gives good results, but there are some much cheaper recycled HDPE boards (search "plastic lumber" on McMaster) that I have been eyeing as a good cheap alternative. I keep meaning to order some, but I haven't had to place a McMaster order in a while. Delrin is workable by hand.
The other mold material that will work well and can be cheap in the long run is machineable wax. When you are done with a mold, remelt and reclaim it. No release agent required with a lot pourable resins, and it hardly wears tools at all. Absolutely hand workable after cutting.