How would you compare the accuracy on the Desktop, Buddy and PRS?

[wespor]

I created a thread in the Buddy forum a couple weeks ago asking about spindles, it morphed into a desktop vs buddy vs full size discussion and while I got a lot of good feedback there, the more the better.

I need accuracy (for carving stamps for leather embossing, printmaking and fabric dying) and z height (for creating concrete molds). The more of both the better but it seems like i'm at an either/or crossroad.

[coryatjohn]

What about speed? Is that a factor?

[wespor]

What about speed? Is that a factor?

I don't think it would be, as long as it can cut the detail and the stamp is durable. My hydraulic clicker is my main fabrication tool, the router will only be used to make tools. Same with my concrete casting, I only use the initial mold to make a urethane rubber mold that I then reuse to cast concrete.

[Ajcoholic]

I created a thread in the Buddy forum a couple weeks ago asking about spindles, it morphed into a desktop vs buddy vs full size discussion and while I got a lot of good feedback there, the more the better.

I need accuracy (for carving stamps for leather embossing, printmaking and fabric dying) and z height (for creating concrete molds). The more of both the better but it seems like i'm at an either/or crossroad.

All Alpha machines have step resolution of 0.0004", while standard machines have resolution of 0.0006".

But ALL the machines have positional accuracy of 0.002".

So to answer your question, the desktop, buddy or gantry machines should all be equally as accurate.

[dana_swift]

Andrew has stated my position- the machine is able to do 0.002". Is the operator? Technique really matters, no machine will get 1" accuracy if the material is not held down.

My question is how accurate is the operator? A new CNC operator will not get high accuracies any more than a beginning manual machinist. Setup, skill, & experience are the key factors if both precision and accuracy are required. It is not automatic with any brand of machine, or product line.

Have you determined your error budget for each step in the process? Its like the old saying: "measure it with a micrometer, mark it with caulk, cut it with an axe".

Setup of the machine matters. How tight to you keep it? How often do you tighten the gears? keep track of the backlash? Check the squareness? Use fresh bits for each cut? How much dust is allowed to settle on the table? How much variation is there in the density of the stock? These things in the hands of the operator, not the vendor.

If 0.1" is your finished precision need you will find it easy to get. 0.01" is not difficult but requires attention to detail. 0.001" requires finesse, setup, calibration, meticulous cleaning, and advanced technique. ShopBot doesn't claim to be able to reach 0.001". However it is possible with some materials and setups.

Most of us on the forum are machine owners who make things. Our customers are usually not on the forum. Each of us has our unique use of the machine and our own expectations and standards. We dont represent ShopBot. We like their products, and most of us have found they exceed our requirements. For your application, consider going to Durham and have them show you what is possible with your specific needs, space constraints, and budget. Talk to them about the maintainence requirements to maintain your precision needs. They can give you the best answers.

Sorry if the forum digresses. We are just humans taking a break from the real work that pays the bills.

Best wishes!

D

[coryatjohn]

>> consider going to Durham

I think this is excellent advice. I learned more in a weekend in Durham than I could have in months of breaking bits by myself.

[wespor]

Thank you, that helps. I guess I got a little confused about the individual shopbot specs. I read that the Desktop's resolution is .00025” compared to .015” with the larger models. I just don't have a really big grasp on how much of a difference that makes or just how resolution relates to accuracy.

I'm expecting a learning curve on the amount of accuracy I can get as the operator, I just need to know I can get there one day with the equipment I've purchased otherwise I've wasted a lot of money.

[dana_swift]

"compared to .015” with the larger models." Where do you get that value? The largest gantry machine they make is specified:

Step resolution of .0004”.
Positional accuracy of +/- .002”.

That is copied from http://shopbottools.com/mProducts/prSalpha.htm That is for their largest gantry router. A buddy is also an Alpha and has the same specs. My personal experience is that the a buddy "can" be better than they specify. My experience with gantry machines is it preforms exactly as they specify. I dont own a gantry machine, but I have worked with several of them.

I dont see any links to the standard models anymore. Is SB no longer offering them?

(Side note, I see that the SB site now requires google snoop ware: "ajax.googleapis.com" and "google-analytics.com" for the site to operate correctly. My browser has those disabled, and others, due to my concerns about snooping especially google. Also required loads are "typekit.com" and "feedburner.com". I really wish sites were not linked to snoopers, especially ShopBot. End of side note)

As to your question about resolution. That is the linear distance represented by a single micro step of the drive motors. A single micro step may or may not overcome the inherent friction of the rollers and result in any motion at all. ( How clean and lubricated are you going to keep your machine?) What will happen with a microstep is a force builds up. After several micro steps the force overcomes friction and the machine moves a tiny bit. How far will it move before friction holds it in a new position with some residual force? All machines do this. All brands do this. Physics is the law.

What the accuracy spec tells you is they guarantee that the force balance details, micro-stepping, vrollers, track hardness, backlash and other engineering factors combined result in a measured 0.002" position accuracy.

Only the best musicians can get the best results from a Stradivarius. Same thing with other tools. The best operators get the best results.

I understand the fear that it wont work out and be misused money. That can happen of course. The machine costs many thousands. A trip to Durham costs much less and makes cheap insurance. Most of the forum users only have access to one type of machine. Asking the typical user to compare them is not realistic. What we all can attest, is that they really work like the good folks at Durham say they do. Not only that but they have the worlds best tech support after the sale.

Get one and start learning. Paralysis by analysis was a problem that held me up too long :) Know that your results after two years of continuous use will be better than the results after two weeks of continuous use.

Just like violins.


D

[wespor]

"compared to .015” with the larger models." Where do you get that value?

It was the answer I received from SB support via email.


Question: I'm interested in detailed work, namely wood and corian block stamps for embossing leather and printing on fabric. What is the smallest bit your machines handle and are both the desktop and buddy systems equally equipped for detail work?


Those two machines are not equal in resolution. The Desktop model is by far the best for your application. It’s resolution is .00025” compared to .015” with the larger models.

A spindle has an ER20 collet and comes with the smallest as 1/8”. That said, you may be able to find a “sleeve” that could be inserted to hold a smaller diameter.


That is what really started my "paralysis by analysis" as you called it. The need for more cutting space and z height but the fear that accuracy would suffer to greatly. I will buy a ShopBot, probably the Desktop w/ spindle to stay on the safe side since I could potentially buy a BT/PRS with router down the road for larger, less accuracy work. That seems like a waste though if the difference in detail work is minute between the machines.

Here are some examples of the kind of stamps I need to carve.

[jerry_stanek]

What size is the Wharton stamp?

[wespor]

What size is the Wharton stamp?

Looks to be about 4.5" from the tape measure at the bottom.

[Brady Watson]

Weston,
The Desktop machine would be ideally suited to machining those stamps. It is a very impressive little machine that packs a lot of punch with fantastic resolution. The larger gantry tools with rack & pinion, just can't resolve those tiny letters and shapes as efficiently as the screw driven DT. Plus, the DT has a much smaller machinable area & is overbuilt compared to a larger tool. It is much easier to hold close tolerances in a smaller space.

So, here's the deal with 'tiny' bits. Everything 1/8" and smaller, by industry standards, has an 1/8" shank. You can go as small as .002" diameter - if you really had to, and it would still take an 1/8" shank. There are a few exceptions to the rule, such as specialty tooling like some Dremel bits @ 3/32" shank and dental tools with either 3/32 or 1/16" shanks. So no worries about getting bits small enough to work on the Desktop. The spindle is more money, but I think it is worth it, if for nothing other than the ER20 collet system.

I know you said you wanted to be able to do some larger projects as well - and that is possible to a degree on the DT by indexing the material. The only caveat being that it needs to be 24" or less wide, but it can be as long as you need it to be. For projects larger than this, you may consider either subbing it out or buying a BT or gantry tool in addition to the DT. Sounds far fetched, but once you make some money with the DT, it may pay for another machine. You never know.

-B

[dana_swift]

The material looks like HDPE (am I correct?) So how accurate is the material?

That has a thermal expansion coefficient of 60e-6 according to the website:

http://www.engineeringtoolbox.com/linear-expansion-coefficients-d_95.html

The rack gears of the shopbot (or lead screw on the desktop) are steel and have a value of 7.3e-6.

With a temperature change of 20 degrees F, the 4.5" length of plastic shown in the Wharton image will change size by (60e-6 * 20 * 4.5) = 0.0054". In the meantime the shopbot itself will change size by (7.3e-6 * 20 * 4.5) = 0.00066".

The difference is around 0.005" for plastic. So if you cut a part at 70deg F, then stamp with it at 90deg F on a nice warm day, the plastic will be off by 0.005".

If the material is only stable to 0.005" over twenty degrees, and the various errors in the shopbot are around 0.002", the material is the limiting factor.

My guess is the value you got from tech support is the total effect of thermal expansion of a large shopbot over some nominal temperature difference between the designed operating temperature and the actual operating temperature. Given a large shopbot would cut 96" long materials and uses steel rack gears with a thermal expansion coefficient of 7.3e-6, if the shopbot is operated in a shop where the temperature swings are ten degrees from the design temperature, its size would change by: (7.3e-6 * 10 * 96) = 0.007". Seven thousandths of an inch. If they quoted you 0.015 they must be assuming you will be operating your shopbot around twenty degrees away from their designed operating temperature and cutting full size sheet materials (which also change size with temperature!).

My best suggestion is if you want real precision, one of the things you need to use is a thermostat. Set it at 75 and leave it there. Do all your work at that temperature and a large shopbot should achieve its rated 0.002 over its entire length. That isnt the way most shops work tho, so I can see where they get that value.

The point I made about about musicians and violins applies here. Are you going to worry about such details? What about the effects of humidity on the leather? I think about those kinds of details. I get amazing results.

Shopbots are not made from Invar. I have worked with that too. They are a fine machine for making most things. They are able to produce as good of a product as your skill allows. You will have to maintain the precision of the machine with lube, tightening and check it every few months. If you plan to make parts 4.5" wide, just expect to not need to get worried about these effects.

Its just my hunch that 0.01" is what most people would consider acceptable accuracy for furniture, cabinets, signs, etc. The machine can do better if you want to go to the trouble to chose material appropriate to the requirements, and calibrate the machine in its current operating environment with the specific bits, collet, and bearing wear in use.

The suggestion of go to Durham, or a trade show where ShopBot and other vendors are present, and challenge them to produce your part at your accuracy is still my best advice. What does a system that meets your needs cost? Who makes them? What is the environment you are going to operate in? Is that within your budget?

Reading the forum, there are a few people whose parts are not coming out the right size. When I see these posts they tend to be from new users. These posts do not seem to be a common issue, so my question is why? Do experienced users just not expect things to come out right (I think not)? Is it that with a little experience the problem is fixable with technique or proper settings? Or is it the user gives up? Both must happen, but I suspect most of the posters stick with it and figure it out. Often with suggestions from here.


Now my question is: "If tech support says the machine is good to 0.015inches, will you be interested in one, or buy another brand? (where they may not be so honest?)".

Your turn :)

D

[Ajcoholic]

If you want real world results, why not ask someone here with each type of machine to cut a sample for you? (Using your provided file) Even if you have to pay for shipping some materials and possibly some machine time- it is worth it to mend a decision on spending thousands of dollars.

I have an alpha buddy with a spindle that I keep finely tuned, tight & clean -
and I think it would have absolutely no issue doing what you show. Easily.

[Brady Watson]

I'm sure SB is willing to cut you a sample using your file & material.

-B

[curtiss]

Are these Corian stamps just pressed into the leather or heated and pressed?

Very nice looking work.

[wespor]

Are these Corian stamps just pressed into the leather or heated and pressed?

Very nice looking work.

The sample I linked wasn't Corian, it was another type of polymer, though I have seen successful Corian cuts (http://www.carvewright.com/wp-content/uploads/2013/01/mold2_closeup.png). I usually have aluminum, wood and steel dies fabricated for me and use a Tippmann 15 ton press to emboss them. It just a matter of pressure and sometimes moisture.

Back to the main topic, I'm overwhelmed with advice and food for thought. I can't thank you all enough, I think I've definitely come to a decision on the Desktop. It sounds like it can definitely do what I need at my skill level. With time and luck perhaps I can buy a larger machine down the road. In the mean time, I live in Houston and there are bound to be several places I can outsource larger jobs.

[wespor]

Desktop with spindle, I meant to say. I missed the edit window.