PRSalpha vs. PRSstandard

[sciencekinetics]

Hello,
I have been researching CNC machines and ready to take the plunge and make the investment. Shopbot is at the top of my list but I having difficulty with one variable. Will PRSstandard meet my needs or do I need to go with PRSalpha? From what I understand, the primary advantage of PRSalpha is speed. For my application, speed is not at all important. Accuracy is of slight importance, however from the specs I've viewed, there is not a significant difference between the two versions. The material I will be cutting is primarily plastic. 2D and 2.5D only. I am considering the 48" x 96" platform. Therefore, from what I understand, PRSstandard should be perfectly acceptable. Am I missing out on some other factors to consider? Thank you to anyone who can offer some advice!!

[david_gardner]

The simplest way I can answer this is that the standard can lose steps and the alphas don't. Speed is the next biggest difference. I went through these same questions before I bought mine. I ended up with a used alpha with a spindle and if I had it to do over I would get the same again.

[erik_f]

I bought my standard for cost reasons. I just couldn't justify the extra expense for what I do with my machine. Besides if things are really going that well in the future you could upgrade to the Alpha. Working within the limits of the machine I have never had a problem with missed steps. The only time I have had stepping problems is when I've done something bone headed or the machine was out of adjustment. Both these conditions would cause poor cuts and problems on an Alpha as well. Both machines should work equally well in most situations, although the Alpha would be better suited for high speed, high production. I find that even though my machine can cut at 5ips that most of the time I'm cutting around 1-1.5ips anyway. I would say if you can justify going with an Alpha it is the SB's top of the line cats meow...but there are plenty of SB standard owners out there who, if they are like me, don't feel like they are missing out on anything either. They are both great machines. You have to think about the future too...if you are making money with your CNC and if you may end up using it for things you didn't intend to use it for.

[tkovacs]

My experience is the same as Erik's. I have a standard and working within the machines parameters it has be perfectly accurate as far as I can tell. Of course I would love to have an alpha but I used some of the saved $$ to add a spindle an I have not regretted the decision. My cut speed is typically ~2 ips, jog speed ~4 ips.

[toych]

I agree with Erik and Terry.

I used the price difference to purchase a Standard with a spindle and it has been the right choice (for me) so far.

You always have the option to upgrade if you find your needs exceed the limits of your machine.

The only second thought I had is table size. I somtimes find I could have used a 60" wide table (if I had a bigger shop) to handle larger sheet goods but the 48x96 handles 90% or better of what I do.

Regardless of which you chose your new Shopbot will deliver exceptional bang for the buck.

[tkovacs]

A correction - my xy jog speed is set to 8 ips

[richards]

I have a PRT-Alpha 120X60, but I would not hesitate to buy a PRS-Standard. The PRS-Alpha and the PRS-Standard use the same mechanical components (as far as I've been told). That means that the difference is the electronics.

The Alpha motors are excellent, but they are still stepper motors. They will loose steps and they will cause a fault.

The Gecko G203v, used in the 4G board, can drive motors up to 7A, which includes the Oriental Motor PK299-F4.5 motor. That motor, when wired bipolar parallel is rated at 800 oz*in of torque. Multiply that by 3.6:1 if you attach a belt-drive and you'll have an incredible 2,880 o*in of torque - or 180 lb*in of torque. That is 100 lb*in GREATER than the Alpha motors.

With a 3.6:1 belt-drive, you will also have the same excellent resolution as the 7.2:1 Alpha motors, since they require 1,000 steps per shaft revolution compared to the 2,000 steps required by a Gecko.

Cost-wise, four PK299-F4.5 motor and home-made belt-driven gear boxes will cost just about the same as ONE Alpha motor/driver.

So, if you're mechanically inclined, you could add four high torque stepper motors with belt-drive gear boxes for about $1,000. You would probably have to upgrade the PRS-Standard's power supply, but that would only cost about $150 for a 50V 600W unit.

The Alpha series machines have other nice electronic features, but we're not dealing with rocket science. If I can build an electronic package to control a Shopbot, anyone can.

[bcondon]

I went the same way as David. I started thinking that the standard was the way to go and went to class. Because I could be machining some large items like Kayak parts, a miss step gets expensive quick. At the class, I changed my mind.

The first piece of info is that I continue to be thrilled with the machine and all the people associated with it (which includes the company, the teachers, the entire support staff and of course, this forum).

IF you decide to go with the standard, you can upgrade from the standard to the alpha for $6000. You then turn in your standard parts for a $1000 return, so for about $1000 more (The price difference up front was $4000) , you can delay your decision to see if your system works well for you (This data was from Dianne when I attended the class)

So, you have options, options, options which shows that the Shopbot crew really understands their market.

Good luck with your decision.

[myxpykalix]

Jeff,
One thing you will quickly learn is that the primary reason for which you buy a bot sometimes gets altered and someone comes to you with a project totally unlike what you envison yourself doing. So for that reason I felt like the better course was to buy the most accurate best unit (at the time) so I wouldn't have to spend the money later because we all know prices go UP not DOWN.
So my recommendation is, if you have the money, get the better unit because in the long run you save money and when you need the upgrade down the line you may not have the money to buy it.

[richards]

There seems to be a misunderstanding about what happens when the PRx-Alpha machines go into "Alpha Mode". The Alpha step motor/driver package has a built in encoder that is monitored by the stepper driver.

Here is a greatly simplified explanation of what takes place inside the Alpha driver unit: As each step pulse is received by the driver, the driver increments a counter. As each encoder pulse is received (showing that the shaft on the motor has actually turned), the count is decremented. If that count starts to increment, the driver knows that the motor is not keeping up properly, so, if possible, it does its "alpha thing" and catches up. If it can't catch up, it sends a fault signal to the Shopbot controller card and the machine shuts down.

That's very good in theory, but it causes problems if more than one motor is active. There is no feed-back connection between motors, so one motor has no way of knowing that it needs to compensate for another motor that is struggling to run at the programmed speed. The result is often a ruined cut - and you'll never know that you have a problem unless you carefully examine each and every part (which we should all do anyway for routine quality control, but hey, I'm lazy and I only want to sample a few parts out of every batch).

The "Alpha mode delay" is factory set for about 1/2-second. That means that the Alpha motor/driver has about 1/2-second to catch-up before it sends out a fault signal. If you're cutting at 6-ips, that means that an axis could fall behind THREE INCHES before the fault signal is sent - and, if the motor/driver recovers successfully, you might not even know that a motor had slowed down during the cut.

Believe me when I say that I've learned to live within the capabilities of my PRT-Alpha (which means that I don't push the machine hard enough so that it ever enters the dreaded "Alpha Mode").

On the other hand, when a "standard" motor looses steps, you'll hear it - loud and clear. It doesn't hurt the motor, but it sounds awful. The cure is to slow things down a little so that the cutter cuts its way through the material instead of trying to force the cutter to take bigger bites than it should.

By the way, I am totally satisfied with my PRT-Alpha. When I had a chance to upgrade the motors from the AS911 non-geared motors to the AS98 geared motors, I bought the upgrade. The machine has worked wonderfully since July 2004. I've hardly done any maintenance outside of a little greasing and a little cleaning. The only part that has failed was one of the proximity sensors. That's just about perfect performance and much better than I had expected. One of these days, when an Alpha motor/driver finally fails, I'll replace all the steppers and all the drivers with standard steppers that have belt-drives attached. I ran the original motors through a belt-drive transmission and had excellent results, so I have complete confidence in going that route - when the time comes.

[Brady Watson]

Go for the Alpha if you can afford it. It is built for production and is better than a Standard in every way. You will not be disappointed no matter which you buy. Think of it like the choice between an F150 and an F350. Which do you need/want?

-B

[angus_hines]

The best answer I can give you is one I heard today.

"They'll all do something Cool !"

[ed_lang]

I heard that too! So it must be true. I saw two machines run today and both did some cool stuff!

Alpha if possible.

[erik_f]

I like Chevy better

[gundog]

Hi Jeff,
I cut plastic with my standard and up until a few days ago I was not happy with the finish quality. I updated from a router to a 3HP spindle and it made a huge difference. I can't tell you about the alpha but my machine works good for what I want to do with it since I upgraded to the spindle. Do not consider cutting with a router if you but the standard I don't think you will like the finish. I cut HDPE and UHMW I am not sure about the other types of plastic.

I do not do production work so to speak just runs of parts for my small business so speed was not an issue for me either. I find I get better results from using slower feed speeds. I am no expert though I have only had my machine for 1 year and I am still learning every time I use it.

Mike

[jerry_stanek]

I like the idea that if I buy the standard and not satisfied with the proformance than I can do the Alpha upgrade. My old PR will cut most anything that I want. I use a router and I have cut 3/4 inch plexi with mine and the finish just needed to be heat polished. Not many chatter marks. I checked the run out and had to replace the collet.

[richards]

The Shopbot is a tool. Buying the correct machine depends on the expected use of the tool. To get all the performance available from the Alpha model will require a large spindle (4hp or larger). It will also require the best vacuum hold-down that you can afford. Total cost for the essential spindle and vacuum blower accessories would be much closer to $20,000 than to $10,000.

On my PRT-Alpha, I upgraded to a 3hp spindle. I also installed two FEIN vacuums for material hold-down. Here, because of the 4,500 foot altitude, I have a problem with sufficient vacuum, even with both FEINs running. So, I have to make multiple passes with the spindle. The AMP meter on the Delta VFD hardly ever goes above 4A, so even a 3hp spindle is much larger than I can use - because of the vacuum. (When I use the GAST vacuum with proper masking, I can cut one-pass through MDF and melamine coated particle board, but that requires a custom vacuum mask for every part and I don't often cut long runs of the same part.)

Cutting speed is determined by the edge quality desired by the customer. 3-ips to 5-ips gives superior edge quality when compared with 7-ips to 9-ips. Keep in mind that I have the PRT-Alpha. The PRS machines probably flex less and are probably able to cut at higher speeds without the "chatter" that I get as the speed goes up.

In other words, it's kind of a package deal. To get the highest production out of the machine, you have to have the right sized spindle, the right vacuum hold-down (if you use vacuum for hold-down) and the right cutting technique. Other than the difference in backlash between the Alpha motor gearboxes and the standard motor gearboxes, the resolution should be exactly the same. The Alpha motors spin at 1,000 steps per shaft rotation and the standard motors spin at 2,000 steps per shaft rotation. So a 3.6:1 gearbox on the standard model generates the same resolution as a 7.2:1 gearbox on the Alpha model - assuming the same spur gear pitch diameter. I believe that the standard model uses a 25-tooth pinion and the Alpha uses a 30-tooth pinion, so the resolution for the standard model is 0.000545 inches per step and the resolution for the Alpha model is 0.000654 inches per step. In woodworking, I very much doubt that anyone would notice the better resolution on a standard machine (even if the resolution per step were 0.002 inches).

[beacon14]

If I can build an electronic package to control a Shopbot, anyone can

Thanks, Mike, that was good for a laugh.

To get all the performance available from the Alpha model will require a large spindle (4hp or larger). It will also require the best vacuum hold-down that you can afford.

I agree, but I don't believe one needs to think in those terms. Like you say, the ShopBot is just a tool, and as long as the tool does what you need it to do you don't necessarily need to get "all the performance available" out of it. My truck can go 90 MPH for 24 hours straight (if I keep it gassed up and maintained) but I mostly use it to drive my tools and a few cabinets around town. Most of the time it sits in the driveway. But it's still worth having because it does the job I need it to do,when it needs to be done.

If the Standard machine will do everything you need, to the quality and speed levels you expect, then it's the correct tool.

[richards]

David,

If you saw my electronics test bench, you'd really get a good laugh. Back in the mid 1970s, when I first needed to learn about electronics to keep some equipment in my photo lab working, I ran into George Balding, who had been chief electrical engineer for a company that built top-secret electronics for the military. Because he was trying to design equipment for high-end photo labs, he was kind enough to mentor me for several years. I furnished him with a test lab for his designs and he furnished me with an education in electronics.

One day, when I was wishing that I had an expensive HP calculator so that I could do the math required for my designs (when a HP calculator cost a month's wages), George told me that when he interviewed engineers, that those candidates who bragged about owning scientific calculators were only considered for employment after the "regular" guys had been interviewed. He simply said that if they needed a scientific calculator that they hadn't had enough experience with electronics to know what they were doing. At that point I began to realize that having common sense was important.

Now, when I design electronics, I look around my shop to see what I have on hand. Linking two or three resistors in series usually gives me a resistance value that is close enough for the test design. Linking two 24VDC power supplies in series lets me test a new Gecko G540 that is rated for a maximum voltage of 50VDC. Using four 24VDC power supplies in a combination of series and parallel gives me 48V at sufficient Amps to drive four motors.

Before a design is considered properly tested, I have jumper leads everywhere. Parts are strewn around and the smell of electrical smoke is heavy in the air (well sometimes I'm lucky and the smoke stays inside the components).

With that little bit of background, you get an idea of my approach to recommending a Shopbot. If I've experienced some chatter on my machine, I'm able to guess that going faster is not usually going to cure that chatter - so, if edge quality is the goal, running the machine faster will not improve the quality (usually). I also look at the Oriental Motor Alpha stepper/drivers. They are robust and have never failed me; but, I know that if a driver stops working, that there is nothing that I can do to fix it except to verify that power is getting to the driver. Keeping a spare Alpha stepper/driver is too expensive for my little shop. On the other hand, I have several extra Gecko G203v stepper drivers in a drawer, and I have spare stepper motors on the bench that I could use in an emergency.

Wiring a CNC machine has much more in common with wiring that add-on bathroom than it has with designing anything more sophisticated. You need a switch to turn the power on and off. You need an indicator light to tell that the power is on. You need a fuse to keep the wiring from melting. You need a contactor (relay) with a self-latching circuit so that you can quickly kill the power by hitting an emergency stop switch. Being able to read simple directions from the manufacturers of the various components takes care of the rest, although it does help to have some DIN rail and some 2.5mm or 4mm terminal blocks on hand so that you can experiment with the wiring.

A pragmatic attitude and having the spirit of adventure takes care of the rest.

[sciencekinetics]

Thanks to everyone who has responded! Your advice has been very helpful.

[beacon14]

A pragmatic attitude and having the spirit of adventure takes care of the rest.

Well said, two characteristics of successful ShopBotters and successful people in general.

[dana_swift]

Mike- you really struck a chord with your discussion of your electronics bench. When I was working as an EE, I occasionally worked with new graduates, currently one of the hot things is "spice", a tool for modeling analog circuts.

The trouble I have with the new EE's is they cant just visualize what a circuit will do in their heads. The dependency on sophisticated tools hides the real world problems like RF interference, ground loops, and poor connections. The ability to imagine the real world a system will run in helps make an appropriate design and then figure out what sub-rule of Murphy's law is in control at the moment.

Murphy's law for spice would read something like: "No real component works the way the model does".

Pragmatic attitude and experience beat a lot of modeling tools. I love the wonderful wizbang stuff we have now, but its just as important to remember that the world got along before it came along.

Now you are making me embarrassed that I actually know math and own one of those fancy calculators too..

I think it all has its place.

Thanks for the thought fodder-

D

[dauntless]

I bought a PRSStandard9648 last year, forgoing the Alpha for cost reasons, and now regret my decision. Mind you, I am carving 3D foam blocks for mold patterns, so I am not straining the machine at all. Still, tool paths are often 20+ hours long and the non-feedback steppers inevitably lose X or Z positioning in spite of slow cut speeds (typically 2 ips), and I end up having to repair and rerun.

In fact, I am cooling off right now in the office after discovering that the bot had lost x-position slightly, and it will complicate finishing this pattern.

And now I want a 12" z-axis, too... ;)

[erik_f]

Is it possible you are having a computer issue or connection issue more than a stepper issue? Just putting it out there. I've only run a couple 3d files and have had what looks like a shift in the model. This I'm pretty sure was due to a somewhat poor connection speed between the bot and the puter.

[dauntless]

Thanks Erik. I can't say for certain that it is not a computer or connection issue, though my installation is very "by the book". How would I go about confirming those factors are up to snuff? Stan

[erik_f]

You may already know these things, so I hope I'm not wasting your time. However, there is a usb connection speed test in the utilities of the control software. This will let you see not only the speed of your connection, but what your actual bandwidth is. I am not an expert on this at all, but I have seen people with as good as an 83% connection, mine so far I can't get above a 75%. 70% is the minimum that SB wants. Also I remember reading there is a file line limitation per how much ram you have. I included a link that helps fine tune your computers resources also. I would hunt around for more info on how to improve your computer/control box connection and also how to make sure your computer isn't getting distracted around hour 10 of a 20 hour file


http://www.talkshopbot.com/forum/messages/312/32195.html

http://www.talkshopbot.com/forum/messages/26/32297.html?1215604790

[cnc_works]

Stan, I can't be very much help in why you are having step loss, but I would be very surprised if it had anything to do with your stepper motors capability. I have an old PRT96 and often run it for many hours cutting 3d molds with never a loss in steps. Sometimes I have to flip the MDF over and cut the other side and I am continually amazed and thrilled to see the accuracy and repeatability of my CNC.

So, like Erik, I would be thinking your PRS has other issues causing your loss of position.

Good luck,
Donn

[dauntless]

Thanks for the tips, guys.

Erik, I have completed the steps Joe calls out in his thread, and also unloaded a bunch of unused programs. I also cleaned up the programs that were automatically starting up and running in the background (on startup). My USB efficiency is 73.9%.

I currently have only one ground wire, and will upgrade my bot's grounding today.

I did get a chuckle about the "100,000 lines of code". My current program has 758,000 lines.

[erik_f]

I don't remember the number...was it 100K lines per 256K of memory? I hope this clears up your problem...I know how frustrating it can be when things don't work out after waiting for hours!

[dauntless]

Thanks Erik...that makes more sense. My computer has 1.5 gb of ram, which according to that standard should support 600,000 lines of code. The computer has an unfilled memory slot, so I'll add another gb as insurance.