Shopbot VS Mach 3

[Cutter99]

Can anyone give me an informed, objective opinion on the pros and cons?

Also please tell me which you would go with...

Thanks,

Cutter99

[Richards]

Both programs will do almost anything that you need a CNC router to do. I have used Shopbot's SB3 program since July 2004 and Mach 3 (test bench only) for more than three years.

First, I doubt that you would want to hand program either SB3 or Mach3. I use PartWizard and Vectric to transform my AutoCAD LT DXF drawings into tool paths. PartWizard, furnished by Shopbot, only produces tool paths for SB3. Vectric produces tool paths for a number of different machines, including Shopbot and Mach3.

G-Code is the industry standard, but in my opinion, is not as extensible as SB3; however, SB3 lacks a few commands that might be necessary in your work. For instance, in Mach3, I can choose any two axes when I want to cut an arc. Shopbot limits you to using only the X and Y axes. That means that you'll have to either write your own code to do a simple arc using the Y and Z axes or the X and Z axes (or buy some full feature tool pathing program).

What I really like about Shopbot SB3 is its ease of use. It is more like a very limited traditional programming language (similar to BASIC). It gives you GOTOs, GOSUBS, labels, variables, the ability to call a program from another program, PAUSES, prompts, eight inputs and eight outputs.

That comes at a cost. You have to buy at least a V201 card from Shopbot. The V201 card is bundled with the software. I don't know the price, since Ted Hall gave me a couple of cards to play with about a year ago. If you're comfortable building a circuit board to interface to that card, you can do almost anything imaginable. On the other hand, if you mostly need a plug-and-play controller, be prepared to spend upwards of $6,000 for an Alpha model Shopbot controller (including motors).

Mach3 is also extensible, but you'll need to have some programming experience to add features. It is also limited to only a few I/O lines (unless you use more than one break-out-board). I have the Gecko G100 and Gecko G101/G102 modules that were supposed to be the ultimate in pulse generators, but Gecko announced about a year ago that they were discontinuing the G100 due to lack of demand and unsatisfactory integration with Mach3. Those modules had plenty of I/O and they worked very well will very few glitches after the G100 plug-in was released by Mach3, but they're obsolete. I haven't purchased a SmoothStepper module, so I don't know whether it's ready for use yet or not.

Personally, I'm satisfied with SB3. The main reason may simply be one of familiarity. It does what I need it to do. But, to be completely fair, if I had to use Mach3 exclusively, I would not be disappointed. It might be, that at my age, I've grown tired to the extensive learning curve that new software (any new software) requires.

[Gerald D]

For most people, the fact whether one is using SB, Mach or EMC, etc. should be "invisible". They are actually machine languages . . . . . most times only receiving a coded file and making the motors move. That coded file being a product of your CAM program.

Having moved from SB to Mach over 2 years ago (SB has obviously been upgraded since), the things we miss are:
- Hitting "pause/stop" on SB was nearly instantaneous, Mach takes a while before it stops
- the trajectory planning in SB seemed to be better. SB could figure that a lot of short lines actually made a long line and keep a consistent speed going.

The things we do not miss are:
- SB was obviously written for the inch environment and we found that running it in millimeters caused too many problems.
- Some CAM programs would not generate SB-code. Everybody else used G-code.
- SB could only run on SB hardware. There were too many reports of that hardware being glitchy, particularly in the area of communication breakdowns on the USB line. (Because of the inch/millimeter issue we were obliged to use an older version, before the USB port was used)
- SB stops running every 100 000 lines or so, to load another batch of data, leaving a cutter burn mark.
- SB could only interpolate circles in the XY plane. Vertical circles needed lots of straight segments.

[isladelobos]

who axis supports each program? sb and mach.

Is possible use this in a 6 axis table or more?

Thanks.

[Gerald D]

Another difference between Mach and SB . . . .

In Mach we set the acceleration during motor tuning and then forget about it. In SB, they don't have the concept of "acceleration" - they advocate the changing of "ramps" for each job. link (http://www.talkshopbot.com/forum/messages/29/1432.html)

Ros, sorry, I did not understand your question??

[jhiggins7]

Gerald and Ros,

I think Ros' question is:

How many axes does each Program ( Mach3 and ShopBot) support?

I believe the answer for Mach3 is 6 axes, X,Y,Z,A,B and C.
Mach3 can also control the spindle (On, Off, Speed). Of course to control 6 axes, you must have the right Break-Out-Board (BOB) configuration.

I don't know anything about the ShopBot software other than what I've read on this Forum.

Regards,
John

[Gerald D]

Shopbot software only runs on shopbot hardware. So, the axes question is related to the capability of the V201 control card from ShopBot - the last time I looked, there wasn't much published about the V201.

[Richards]

The V201 board has provision for 5-axes, X, Y, Z, A, B. (Shopbot's SB3 software only handles 5-axes.)

The V201 board has a male DB37 connector that handles all of the stepper driver signals as well as all INPUT signals. The +5V input voltage to the board and the OUTPUT signals are handled by a separate 16-pin IDE-type connector (two rows of eight-pins each, 0.10" spacing).

The V201 connects to the computer via a USB connector.

With the V201 board, the X-axis only has one step/direction signal. The is no provision for a "slave" axis. So, the X-axis slave motor would either be driven directly from the V201 in parallel with the X-axis master motor, or by buffering the signals through a TTL gate (7404 to invert one set of signals, or 74245 (or similar) to simply amplify the signals. Remember that the SLAVE motor runs backwards when compared to the MASTER motor, but you can easily reverse the direction of a motor by interchanging ONE set of connections between the stepper driver and the motor, i.e., Black/Yellow to the A-coil on the SLAVE motor would be wired Yellow/Black.)

The step/direction signals are the correct polarity to work directly with the Gecko G203v stepper driver.

The OUTPUT signals are reverse polarity (compared to most TTL signals). When OFF, the OUTPUT sends a 0V signal. When ON, the OUTPUT sends a +5V signal. In other words, the OUTPUT port SOURCES the signal. Most TTL devices SINK the signal where ON = 0v and OFF = +5v. (I know that is confusing to many people, but most early TTL devices could interface to 10 other devices when they SINK current, but they could only interface to 1 other device when they SOURCE current. Because of that, 0v became the accept voltage level for ON and +5v became the accepted voltage level for OFF.)

[Cutter99]

What about the problems with cuts and gaggies etc? I have read some posts that show Shopbot to have lesser quality cutting than Mach 3...

Are these reports fixed or still better with Mach 3?

C99

[Gerald D]

In terms of cut quality, there should be no difference between Mach and and SB running on their V201 controller. Both will be doing the same level of micro-stepping of the motors. Mach has always done 1/10th micro-stepping, older SB was something like 1/2 or 1/4 stepping.

[Cutter99]

"- SB could only interpolate circles in the XY plane. Vertical circles needed lots of straight segments."

Is this true and if so is it fixed or still a problem? How does it affect the quality?


C99

[Gerald D]

As far as I know, it is still true. It is not a problem if you divide the vertical arc into enough straight segments. A good CAM program will take care of this for you.

[Cutter99]

If I upgrade to the new Shopbot Stepper Controller Box, will I be able to run Mach and or Quantum on it as well?

Thanks,

C99

[Gerald D]

All the ShopBot controller boxes that I know of will only run their software.

[Richards]

The latest several versions of the Shopbot SB3 software include a G-code converter so that native G-code files can be fun. When I tried it a few months ago, many of the G-code instructions had not been totally defined (especially arcs). The latest word is that SB3 handles all G-code files produced by Vectric (V-carve). I have not tried generating and running a G-code file with my copy of V-Carve Pro.

[Gerald D]

Yes, the boxes will run the linked limited G-Code (http://shopbotwiki.com/index.php?title=GCodeMain) but it will be under SB's software. They will not run Mach, EMC or Mach's Quantum like our generic boxes.

[Cutter99]

Quantum sounds like the perfect thing to smooth out my old machine.

Any suggestions, info, anything about the shopbot vs Mach Quantum / mach 3?

Should I get the SB controller then upgrade to a mach controller? What are the downsides of going outside of SB?


Thanks,

C99

[Gerald D]

Richard C99, let's start at the beginning . . . . . what ShopBot do you have? What are the model numbers on the motors? What does the controller look like?

ShopBot changed their configurations and options quite often. Some of them can be improved a lot with a new controller, some can't.

[Cutter99]

Thanks for all the help.

I built an older Shopbot with the older control box many years ago,
sold it and just got it back. It has a cut area of about 5.5' x 12'. I am a graphics guy with 20+ years experience in sign mfg.

Here are the motors:
Vextor PK296A1A-SG 3.6 / 5 degree Step / 2 Phase DC 1.5A 2.2Ohms
The Z motor is a PK268-02A 2A 2.25 Ohms
I was told they could be powered to 2.3A

I have an older Shopbot but it would benefit from the new controller I am told. My Z really sucks, plus I have 2 of them. It is the old ball screw type. I was hoping there was some way to fix it up without too much rework?

Actually everything sucks but the heavy duty table I built!
The rails are the old steel strut rails with BWC bearings but not the track.
But it did work, except for the jaggies, and the Z would loose it's height and it was slow etc etc...

I have heard Quantum might help out with the smoothing and racking?

Any advice weather I should go Shopbot, or Quantum?
Is their any quality differences? In functioning? In cutting? In expandability?

I have read many many threads, but they go back so far that it is hard to tell what is current information.

Then what about the CNC Brain? Bob Campbell is another very interesting option. What boards are the newest and the best technology for the money?

Does Shopbot have a limit of 5 drivers? Is there any limit to the drivers with Mach? (Anyone ever use 2 motors on one Z?)

Thanks for the interest and help. Anything you can think of that might help please share. There is another guy who emailed me in the same position as well.

Thanks,

Richard

[Gerald D]

Are all your ShopBot components from before the year 2000? Is it rack & pinion drive or cable drive?

[Richards]

The PK296A1A motors have high inductance when wired series. The Gecko G203v controller was not designed for motors with an inductance rating of more than 20mH; however, if you wired the motors half-coil, they would work very well with the G203v.

The PK268-02AA motors are excellent LITTLE motors. Just yesterday, I mounted two of those motors onto Oriental PAL2P motor mounts for an upcoming project. All together, I have fourteen of them that I've stripped out of old process control computers that I once built; however, they are SMALL motors. The PK29x size (34 frame size) motors are much better suited for use with a CNC router.

Although reusing old parts is sometimes possible, the electronics used in older model Shopbots are not nearly as good as the electronics that Gerald uses on his machines. The PK296A2A-SG7.2 motors that several MechMate builders have selected are excellent. The Gecko stepper controllers are all excellent, with my preference being the G203v. The PMDX-122 breakout board is excellent. A home-built toroidal power supply is perfectly adequate and the the reports of the AnTek supplied power supply seems to show that AnTek would be an excellent choice.

The non-industrial enclosures of earlier Shopbots also lack many of the features that Gerald has included in the controllers that he (and others) have built for the MechMate.

In other words, sometimes you need to just junk the old car and buy a new one. Chances are that by the time you tweaked the old electronics to give you the performance that you would want, that you would have spent more money and time upgrading the old than you would have to spend installing new.

[Cutter99]

Excellent information. Thank you. I also read Mike's post over here about SB motors... http://www.talkshopbot.com/forum/messages/32/35634.html?1225829441

I am on as budget so my time vs new machine won't come into play. I need to get this machine running and producing before I get a new machine. I will have to scrape the barrel to come up with the $1500 for the new controller. Also because of the Canadian dollar I need to add almost 25% to any of the pricing for the exchange rate.

I also was wanting to do this in stages. The new controller will give me much more power and resolution than I have now. My current controller only gives me 1/4 or 1/2 micro-stepping, but the new one will give me 1/10 micro-stepping, greatly increasing the resolution. The power supply also goes from 24V to 68V.

SB will charge me $1474-$250 for my old box. This will also come with some help. That part is attractive... But Bob Campbell will charge me much the same for a better box probably.

I think once I get moving though I won't need SB so much.

I feel my next weakness is the ball screw Z, motors, then the rails.

What about the smooth stepper? Quantum and smoothing? The CNC Brain?

So from an upgrade point of view, what would you do?

I think just a new controller for now would smarten the machine up enough for me to produce product. Currently it has too many jaggies, is slow, racks, and sometimes the Z depth goes out, and is underpowered. So what controller? What circuits are the most current and best bang for the buck?

SB mentioned I would need a capacitor changed...so the box will have to be able to upgrade with the motors...

I could always sell the old SB controller and motors as well? Or get the new controller and sell the old at a later date?

Or???


RB

[servant74]

ShopBot is a nice option, and their newer models are more 'industrial' that the older ones. The software upgrades over time have gotten away from many of the jaggies problems. They come with some great software (partworks, etc). They also sell a new Z axis in case you want to bite a BIG bullet, but it looks fantastic (I saw it in October in Austin, they gave one away at a conference). I have not found what dialect of G-Code that shopbot works with, but I know they say it does.

EMC2 and Mach are both great g-code interpreters and can work with lots of different controllers. But SB being proprietary, I don't think they can run on it easily (spelled: without lots of software hacking they don't stand a ghost of a chance to run on the SB controller hardware ... starting at their gecko's on, everything should be OK).

IMHO, if you have a shopbot and want to keep resale value, stay with shopbot components. Otherwise, you can just set aside the shopbot controller you have and build a new one for use with Mach or EMC2. If you like the shopbot software, like partworks, go to Vectric and see if you can buy an 'upgrade' to their current version from Vectric or to Aspire (With Aspire and Cut3D it appears you can do everything all the other Vectric software will do and in a more unified fashion, but it will cost some real $$, more than my play money budget.)

If I was buying a low to middle end CNC for a shop and had business where this machine is needed quickly, I would probably go ShopBot just to have the 'professional hand holding' available. If I had a longer time horizon, MechMate is the way, but you loose the opportunity of the possible business in the interim. Never an easy choice.

I know others will have different opinions, but your decision is the right one for you.

[Gerald D]

Repeated question: Are all your ShopBot components from before the year 2000? Is it rack & pinion drive or cable drive?

[Richards]

Jack has made some very valid points. The cost of upgrading vs the cost of lost business because the machine can't do the work is always a hard decision.

The first requirement is to thoroughly check your machine. Every mechanical machine needs maintenance. Go through the machine and replace or repair everything that is worn or out of adjustment. Add bracing, if possible, to stiffen the gantry. (80/20 aluminum is my favorite "erector set" material for one-of-a-kind projects. It is amazing how easy it is to modify an existing machine with some of their products. It's a little pricey, but all that is required is a chop saw and a few hand tools to tighten the screws. Steel is probably much better than 80/20 aluminum, but I'm lazy and I'm mechanically handicapped, so I go for the "easy" over the "best" when I'm prototyping.) Remember that adding electronics won't correct maintenance problems.

Once the machine is working properly, with every axis moving smoothly without snagging or binding, then it's time to look at the electrical.

Normally, I don't recommend things that I haven't tested, or at least seen; however, I have enough confidence in Gecko to recommend the G540 to those on a tight budget. That one device takes the place of four G203v drivers and a break-out-board (although at lower voltage and at lower current). The PK296A2A-SG7.2 motors pull 3A and require about 35VDC (32 X SQRT(1.5) = 39 Volts maximum), so those motors fall within the parameters of the G540.

A G540 costs $300. The PK296A2A-SG7.2 motors cost $257 each. An 11A AnTek PS-4N35 costs $94. Add in the price of a license for Mach 3 and you've got the basic electrical parts needed. Granted, you'll need to add switches, fuses, sensors, etc., but chances are, you don't have most of those parts on your Shopbot, so do yourself a favor and just build a cheap MDF or particle board case, add a fuse and a On/Off switch and begin cutting. After you've started cutting, use some of your profits to buy the other "elegant" parts and pieces as funds permit.

Every machine has its limitations. There are many things that I don't like about my Shopbot PRT-Alpha (too flexible, too much chatter, too hard to load from the side), but I've learned to work around the problem areas. When the machine finally breaks, and if I'm still able to do hard labor, I'll replace it with a MechMate; but, the old saying about a bird in the hand vs two in the bush is a good saying to remember. If you've already got something, make it work until you can get what you really want.

Edit: Gerald, you posted while I was typing. You've made a good point. The old cable-drive machine is probably not worth messing with, unless your needs closely match it's capabilities.

[Cutter99]

Thanks again for the great info.

Yes my machine is rack and pinion. Maybe one of the first, because they flew up twice at their expense to work on it, back in 1996-98 approx. But it has the old style BWC bearings but no track. The bearings ride on angles on the lengths of steel strut.

I also lost a very large contract back then because of the jaggies, and Shopbot did not get rid of them. I kept trying everything they said but the client finally gave up on my quality. I now know it was because the 1/2 micro step then to the 1/10 micro step now. Other problems and bugs that I was suffering from I see now from the forums were Shopbot Tools bugs, but I was told other things.

All I need to do now is replace the controller.

That controller should allow for a wide range of upgrade options, as I don't know what the path will be yet.

Motors: Keep the ones I have now, but I must have the ability to upgrade to really big motors later.

I don't know the advantages / disadvantages to different controllers, and driver software, and code etc. I think that with Shopbot I won't have the quality or the flexability I would get with a Mach system.

As far as the Shopbot Z, in my opinion, it is not strong enough or substantially engineered for me to part with $1600 Canadian plus $500 more for a new motor x 2 for the 2nd Z. They don't have the "machinists" feel or rigidity to them I suspect. There are also posts about it flexing and chattering and how one should learn to live with it. Especially the 12" slide. I could be wrong here, but I bet the Mechmate Z is stronger. For the money they are asking would I not be better with a K2 or other professionally engineered Z slide?

I think I can get my ball screw Z (I have 2 of them) to work better just with the extra power and resolution from the new controller. Next I would resquare them and weld the bolted strut together. I could also put a new, bigger motor on them. Finally, replace them with a proper Z slide. I have been thinking about 2 motors for the Z. If one looses a step the other might not? Plus twice the power...

RB

[Cutter99]

What steppers do good machines like Gerber or AXYZ Automation use?

How do you know how heavy the entire gantry can be before the motors can't handle the weight?

RB

[Gerald D]

Richard, is this what your ShopBot looks like?:

3345

I have no experience of that machine other than the stories I have read about it. Apparently the screw on the Z was prone to getting dirty and binding up - a new control box is not going to improve that. A bigger motor on the screw will allow it to get dirtier before it binds. Twin motors could make it worse because they might fight each other and cause each other to loose steps. So, if you keep the screw, the only improvement will come from a bigger motor. (A new controller might get more power out of your current motors, but I don't know enough about what the old controller is doing now)

The gantry is a flimsy structure and obviously prone to racking out of square. But the flimsiness also causes jaggies when the whole thing starts rattling. Quantum supposedly makes an improvement with its reduced jerk on take-off. But Quantum is un-supported experimental software - it is going nowhere fast.

Your X & Y motors are good, IF they can be wired half-coil - some ShopBot motors had only 4 wires coming out the case (black, green, red, blue) and that is a problem. Check if you have yellow and white wires accessible.

If you are not in the mood to build your own control box (under about $1000), then you will have to buy. Your cut quality with all boxes (DIY, Campbell, or ShopBot) will be very much the same and a heck of a lot better than you have now. Quantum might give an even better quality, but I don't have experience of it. DIY and Campbell boxes give you a chance to try Quantum, an SB box won't give you that chance.

[Cutter99]

Yes that is the machine...

I have beefed it up so it will get me by with the just the controller for now.

RB

[Cutter99]

and some motors probably...

RB

[sailfl]

Wow, that looks like the dark ages of CNC.

Richard, you must be good to get any quality out of that.

What did they sell that for?

[servant74]

It looks like some of the first designs from ShopBot. ShopBot was initially designed as plans only what could be built from hardware store components, I was told.

Personally, I would like to see a copy of the 'initial plans' they sold way back when. It is obviously a lot less of a machine and probably $$ than a MechMate is today!

[Cutter99]

Yes it was the dark ages...I started out with a kit I bought from them for a cable drive machine. I built a really solid steel table with outside dimensions of 6.5' x 13' long. It was not long before they offered rack & pinion and bigger motors. I used it to cut steel, aluminum, wood signs, plastic signs...I cut 10s of 1000's of letters out of pvc for the US Post Office. I drilled sheets for military products... I sold it in 2001 to a furniture maker and he cut tons of wood on it until a few years back he got into building luxury homes and mothballed the SB.

Over the period I used it I added a 5hp 600v Columbo spindle, 2 z's, Fog Buster misting system for cutting metals and spent soooo many hours tweaking, squaring, modding and upgrading. I redid the rails, tracks etc etc...

The 3 main problems I had were the ball screw z would loose its place, and the 1/2 or 1/4 micro stepping caused jaggies on the edges of certain parts of a circle. There were also real problems with files...weird cuts and movements that were not in the file.

SO a new controller would get me more power and resolution. That alone makes the machine much better than the last time I used it. IT will help with the cut quality, accuracy and speed. Less racking, and with more power to the Z it might hold its 0 better.

Next I will get new motors.

Then a new Z or 3. I read on a CNC site that 3 Z's will satisfy 99% of manufacturing needs, making a tool changer redundant.

Then the rails. I have a connection with a company that manufactures super high quality precision tubing for nuclear, underwater oil drilling etc. and they are really accurate. I may use this for the rails, as it will be free and I can get a 13' all one piece.

Yes a Mechmate is what I wanted the Shopbot to be back in 1996. I would like to rebuild my machine with a single y gantry beam.

RB

[jrabeneck]

I have a shopbot circa 1999 with exactly the motors that you have on your shopbot and mine looks almost exactly like the one that Gerald pictured in an earlier post. I recently purchased 4 Gecko 203v's and a pmdx 122 and mounted in the shopbot control box. The control box was a PRT 48x96 upgrade from the original PR96. I used the 44v power supply from the PRT box and wired the motors half step. Performance has increased dramatically.Jog speeds are comfortable in the 600-650 ipm range. It will jog faster but will sometimes lose steps. I am still using the original rails and rollers from the shopbot which lack rigidity. I have not done an a/b comparison on cut quality between the Gecko's and the shopbot controller but it seems that the Gecko's are very much smoother without the jerkiness.
BTW I am using Mach3.
JR

[Gerald D]

. . . . . I used the 44v power supply from the PRT box and wired the motors half step. . . .

Thanks Jimbo. I think you meant half coil. You can go up to 75 Volts for the half-coil configuration which will cure the lost steps for jogging. (I think that Campbell runs his boxes on a low voltage as well?)

[Cutter99]

Thanks jrabeneck.

What did you do with the Z? What speeds are you getting with the Z?

How did you find the learning curve of Mach 3?

That board is pretty cheap...

What did all the upgrading cost?

Anything else you can tell me?

Thanks,

RB

[jrabeneck]

Gerald.
You are right. Half coil is what I meant. I will eventually upgrade the power supply to the higher voltage. The 44v power supply is working just fine for now.

RB,
I am still using the Shopbot Z axis. I have not had any trouble with binding. As for Z rapid speeds, I am using 50 ipm. It may go faster. I haven't tried. I an planning on upgrading the complete x y and z axis to the mechmate system as soon as I gather all of the parts .
The learning curve for Mach 3 is not any more difficult than for Shopbot. It is just different. G Code is not quite as easy to understand as shopbot language. I have found that it is easier to find post processors for Mach 3 than for Shopbot.
Costs for upgrading the control box:
4- Geckodrive 203v @ $147= $588
1- PMDX @$81
Assortment of hardware ( heatsink,resistors screws etc) ~$25
Total less than $700
JR

[Cutter99]

How hard was it to get the proper settings in Mach?

Would you build your own or buy one if you could do it again?

50" per minute is pretty slow... I bet you could get it going faster than that.

I was thinking about welding the Z slide structure...

RB

[jrabeneck]

RB,
If you know your VU (unit value)settings in Shopbot, it should be relatively easy to determine steps/ unit values in Mach 3. My original Shopbot VU for the x and y were 127.324. For the Mach 3 the values are 2546.48 or exactly 20X the original values. There is a calibration function in Mach 3 that will help determining the correct value.
If I had it to do all over again and I had money to burn I may have let someone else build the control box. Shopbot wants like $1000 to $1500 for the 4 gecko upgrade. That is about $300 to $800 more than I have invested. That $300 to $800 will go a long way towards rails and rollers and laser cut parts.
JR

[Cutter99]

These are my motors:

Vextor PK296A1A-SG 3.6 / 5 degree Step / 2 Phase DC 1.5A 2.2Ohms
The Z motor is a PK268-02A 2A 2.25 Ohms
I was told they could be powered to 2.3A

How much juice can they take?
How many ounces would they be?
How would they compare to new motors?

Thanks,

RB

[Richards]

The PK296A1A-SG3.6 motors have 7.7mH inductance when wired half-coil, so they can handle over 80VDC. The Gecko's limit of 80V means that you'll get maximum advantage of those motors if you can wire them half-coil. (Some motors supplied to Shopbot only have four wires Black/Green Red/Blue. Those must be wired series.) The SG gearbox limits the torque, so half-coil connection is not a penalty.

The PK268-02AA motor is one of my favorite motors. I've installed dozens and dozens of them over the years in non-CNC projects, but they are small and not ideal for use on a large CNC router. I would not use a 23 frame size motor on a CNC router unless I were stranded on a deserted island and had no other motors available. The PK268-02A motors are rated 3.6mH half-coil or 14.4mH series. I use a 50VDC power supply when I connect that motor half-coil. If you used a 70V to 80V power supply, you would have to wire the motor using the series connection. Wiring the motor series would give you excellent torque, but it would also limit the speed to around 200 - 300 RPM. With a 5-tpi ball-screw, that would give you a maximum speed of about 1-ips.

Running two power supplies, a 70V for the PK296A1A-SG3.6 motors and a 50V for the PK268-02A motor is a possibility, but it would probably cost about the same to replace the PK268 motor with a PK296 motor. Even then, the problem is that a high-performance 34 size motor (motor with low inductance) works best when wired half-coil and a power supply in the 35-40V range.

Those PK296A1A motors just have the wrong inductance when compared to most other motors being used on a CNC router.

[servant74]

http://blog.makezine.com/archive/2009/04/shopbot_open-sources_their_code.html

Apparently ShopBot has open sourced their code. ... Interesting.

This happened some time ago, I guess I just missed it!

[Gerald D]

Their input .sbp code (the stuff you see in the file with Notepad/wordpad etc.) has always been freely accessible and they promote it as better than g-code. This official "open source" announcement is just another push to get more recognition for their .sbp code, as opposed to g-code.

Their control software (actually firmware), serving a similar function to Mach3, LinuxEMC, etc., is definitely not open source

http://www.opensbp.com/

For people to use ShopBot machines they have to feed it a non-standard code (with .sbp extension). ShopBot is resisting a full swing to the internationally recognised g-code. (G-code is also very much "open-source", although their isn't an "official" site for it)

[Richards]

After visiting the OpenSBP web site, I fully agree with Gerald. The OpenSBP movement seems to be to "allow" anyone to use Shopbot "defined" instructions without worrying about getting a call from a lawyer for copyright infringement.

To RUN that code, you would still have to have a Shopbot controller. That code is not opensource.

I have written a lot of OpenSource code. OpenSource means that the source code, the actual code used to create the program, is included so that anyone can modify that code to allow the program to do whatever is desired. The only stipulation on OpenSource code is that it must remain OpenSource, meaning that if you modify my code and then pass that code on (sell or give) to someone else, you MUST include my source code and your source code.

Now, back to Shopbot. They have NOT released the code that they use to create the SB3 program; therefore, their code is not OpenSource. All they have done is to allow anyone to use their syntax and to encourage others to create tool pathing files using their code - so that they can sell more machines that use their proprietary controller hardware.

The OpenSBP "movement" seems totally self-serving to me. Frankly, I'm very disappointed with Shopbot for pretending that OpenSBP is something that it is not.

[servant74]

Now we just wait for a nice geek to do a SB to G-Code converter!

If someone did and got it really working, at least SB couldn't complain that they are ripping off SB code! :)

Still, nothing to lose sleep over at this point, one way or the other.

[Alan_c]

Shopbot actualy released one, I have a copy but cant seem to find it on their web page right now, go there and do a search.

[jrabeneck]

UX_SBP.exe. It is a utility included with their control software.