eMachineShop is a threat to small machine shops?

My opinion is that as long as one is forced to submit designs via a proprietary format the potential market will be quite limited. Anyone (me for instance) who has spent a couple months modeling a design in a parametric CAD package, in order to create an accurate 3D model with linked detail drawings, is not going to spend more time redrawing the parts with klunky software just to get a quote. Or to find out that the part simply can't be created with the system.

The service may work for the occasional customer who is content to draw a simple part and doesn't require part documentation (as far as I can tell eMachineShop's software does not create dimensioned drawings.)

Neat idea and perhaps a good start, but useless to me at this point. If the system were smart enough to accept native 3D CAD models then they'd have something really slick. While the current system must represent a considerable investment, what I'm suggesting would be an enormous task.

Ned Simmons

Nothing ever springs forth fully developed. If there is a demand, a product soon will be forthcoming.

One of the problems with a lot of CAD drawings is that the draftsman "cheats" the CAD by drawing something approximately and then manually dimensioning it correctly. The paper drawing is fine, but the CAD file is full of "bugs."

If someone has a specialized CAD program or some sort of translator for common commercial CAD programs that automates the bridge between the drawing and the CNC program, it will start turning up in the industry as shops seeking to improve efficiency and shorten turn-around buy it.

Of course, the early versions of any such things will not be without their problems.

I have made a great deal of use of a similar service for circuit boards. You use the company's propriatary layout program, e-mail them your finished design and the UPS guy hands you your finished boards a couple days later. It's cheap and a great way to make a proto or one of. But it hardly replaces the conventional board houses who can do far more sophisticated work.

The small machine shop has one huge advantage: the machinist can talk to the mechanical designer (customer). Many mechanical designers are glorified draftsmen who couldn't find the power switch on a Bridgeport. We all know that it is very easy to design something that is very hard (read, costly) to make. A little communication can go a long way toward making a customer satisfied and keeping him coming back.

My take: There is a niche for "eMachineShop" type operations, but they're not going to make much of a dent in the typical shop's business. And, if there is a significant demand for such services, it won't be long before reasonably priced tools become available to enable the local shops to provide them.

Jerry

I tried to get a few parts made but couldn't get through their CAD program. It took me about 20 tries to get the drawing right, then when it came time to detail which tools were going to be used I had no idea what the options meant. If their had been a local machine shop to go to it would have taken me half as long and I would have had a finished product. I have drafting experience, have read a lot about machining, but have never operated a lathe or mill.

my 2¢

Eide

Sounds very similar to the scheme employed by a popular hobbyist printed circuit board house... but it only attracts hobbyists.

Are you talking about a CAM program, or something else?

One big problem I've had is that local shops generally don't want to even talk to me unless I'm giving them $500 or so in business.

I think their software can import DXF. Of course that may be useless if you have a complex 3-d part. But still, I can see it being good enough for a lot of people.

I know for a fact that many machine shops out there have a web site and the ability to provide quotes on line from a CAD file. Why are they not taking the next step and doing something like eMachineShop? Is the market too unprofitable? Or is it plain luddism?

"Soon" seems awfully optimistic. Based on the fact that reliable translation between CAD packages, for which there's a huge demand, hasn't improved in twenty years (in fact the prevalence of 3D modelers has made things worse), I'd say that even "someday" might be optimistic for a system that can accept data from 3 or 4 of the most common modelers and generate code reliably. You have to remember that you're shooting at a moving target in the form of proprietary file formats that are changing every 6 to 12 months.

It's pretty easy to create an accurate solid model of a simple prismatic part of the complexity that eMachineShop is capable of producing. Surface models of "swoopy" shapes are another matter entirely.

Aren't you describing an ordinary CAD to CAM translator?

I wonder if eMachineShop didn't model themselves after this sort of service?

Have you tried eMachineShop's software? It does, for example, attempt to help you decide which process is appropriate for the part you're creating based on geometry and material, and will tell you if you attempt to include a feature that can't be made, or is difficult to produce. Pretty ambitious, though I found the current implementation rather flaky. It would do things like recommend a switch from laser cutting to water jet, then inform you that water jet is incomaptible with the specified material.

Agreed.

Ned Simmons

I think you're right. I've played with the software and for something like a plate with a bunch of holes, importing a dxf seems to work fine.

Do you think the quoting is automated, or is someone opening the file and manually generating a quote, or some combination of the two?

I suspect it's a combination of a perception that it's not a large market, along with the fact that it's not an easy thing to do. It appears to me that eMachineshop has made a large commitment to date, and while what they've accomplished is impressive, has a long way to go.

Ned Simmons

In fact, according to PC Magazine, the president of eMachineShop also created pad2pad, a circuit board service.

The quoting at eMachineShop is fully automated. The recent Wired magazine article mentions how it's done.

I'm not a professional machinist, and don't know exactly how quoting is done in the online world. I assume that an experienced person takes a look at the design or CAD file and gives a quote based upon lead time, quantity ordered, current workload, and machining complexity. More complicated parts may require CAM and a CNC simulator for quoting.

Am I missing anything in there?

My two main machinists for robot parts are in Nevada (for waterjetting) and Wisconsin (everything else), while I live in Georgia. Right now we get everything done through e-mail. These guys do good work at a great price. I'm sure I'm not the only one who does this.

I have had a few parts quoted on E-Machineshop. They are returned much too fast for it to be a human. At present, they are not a viable option for me. I owe them a debt of gratitude, because their quotes were so high, I bought DRO's, and a new 6-jaw chuck, learned to run my machines better, and are now making the parts myself. It would have been nice if it has worked out.My problem with local shops is not prices/part, but quantities. 10-20 pieces. What they guys really want to do is to load up their CNC's and go watch Oprah, I guess. Whatever the issue, I did not appreciate being told to GFMS. I am having waterjet cutting done, which elimnates a few steps and reduces material removal for me.

Does anyone else smell spam all of the sudden?

I suppose it's possible that this isn't an astroturfing shill, but...

I'd wager there is a push to change proprietary formats in order to force customers to contunue using a certain software. There's really only so much difference between two CADCAM programs. They all let you draw parts, they all let you cut them on any machine. Other than specific software taylored to a specific industry...

I can imagine that these surfaces would be virtually impossible to quote automatically anyway. I have spent days taking machined surfaces to mirror finishes using stones and emery cloth. Certain applications cannot tolerate unfinished surfaces.

I think the idea of trying to get software to infer what a designer/engineer has in mind (or what the appliction demands) is ludacris. The worst case sceanario is that you get people like Iggy who don't really understand what they're tying to make and have no idea how to make it trying to use software to design workpieces.

Perhaps I'm being a stickler becuase I firmly believe things should be done

*right*. This software could do for manufacturing what the internet has done for literature. Now any idiot can instantly publish their own work in front of an audience of millions (are we at billions yet?) of people for next to nothing -and it certainly doesn't have to be correct, or even good.

ME degrees and machinist apprenticeships aren't 3-4 years long just for fun...

Regards,

Robin

"Will eMachineShop and services like it put anyone out of business? Or are they simply tapping into a new market?"

Emachineshop is certainly NOT going to put any decent machineshop out of business. Their software is pretty much limited to 2D (I belive they call it "2-1/2 D"). Their service and supply chains are terrible, and their lead times are OUTRAGEOUS.

See my writeup here:

-M40-