With a standard control unit.
With a standard control unit.
Looks like a great training/small project tool.
for that price for a desktop - I’d rather go with a 5 axis machine.
Now that would be a learning experience.
$8k is the starting price for a Tormach as well…
Whether or not the Haas desktop trainer make sense as a purchase for a makerspace depends on its goals. If one of the goals is to help to prepare someone for a career in CNC Machining, then they should be certain that they have the equipment that is broadly used across different industries and regions. I would guess that Haas is nearly as common as Fadal even though their controls are proprietary to their machines. Because Haas machines found a really good position in the market by providing a good-all-round machine at a very competitive price, they can be found in a very high percentage of shops in the U.S.
So to me, it makes sense to stay with Haas across the shop, its a simple straightforward control with 95% of the functionality needed, its versatile (I built injection molds on them for a decade), it can do everything and more than the ones mentioned, and knowing how to walk up to one and run it opens doors for people. It also makes no sense to go for a higher-end machine for a makerspace (like a Makino) because the liability for damage is so high.
this has been discussed in the past and no, that is not our goal. we are not a certified or certifiable institution. We are a community workshop that helps people/members/customers get access to tools for completing their personal projects. The class we offer would certainly teach how to run a CNC, thus aiding them if they wanted to be a CNC operator. A CNC machinist does more. A CNC machinist MUST be a machinist FIRST - and that is not something we can teach in a 2 hour class.
I agree that is the right approach for a makerspace. However it still makes sense to maintain a common control if there are multiple machines (like the VF1 and the desktop trainer). As you said the goal is to complete personal projects, which is best if your not having to learn multiple machine controls.
However, I think the makerspace should be open to someone with the personal initiative to combine the use of online training like at https://academy.titansofcnc.com/ and the conventional and cnc machines to become at least an apprentice level CNC Machinist. From personal experience I can testify that this is a rare (now, not 30 years ago) and valuable opportunity.
Thanks for pointing this out. Looks great for my personal projects and I have very limited space. Was hit by the tornado and will have to have the drive torn up. Makes the decision to run a 220 line to the garage very timely. Perhaps not good for the Makerspace but looks great for me as a member.
Just remember this is designed to cut plastics or machinists wax, not at all the traditional materials you’d expect out of a Haas.
Very good point.
True, but its still the traditional Haas controller and if you can learn that, you can run almost any Haas made.
One of the nice teaching things about this machine is that the controller is designed to interface to larger screens and classrooms in mirror mode.
This would be nice if anyone actually taught classes on g code
But given that we’re primarily about making things, I’d like to see a PocketNC instead for that 5 axis.
That’s what the simulator behind the HAAS is for.
I’m still amazed of how many shops still hand code. It’s usually a shop that is too cheap to buy software and yet they’ll waste 1000’s of man hours on trial and error trying to bang out a program. Now, knowing Gcode for modifying a program on the fly is another thing and quite useful to know. I still have problems with arcs but most other stuff I can hand code. With the exception of the HAAS, no other CNC equipment lets you modify code on the fly anyway. You always have to go back to the drawing and post process or modify and reload the gcode file.
I would suggest the only reason you need a 5-axis machine is either you have to have same-as-production materials or they are to large to be 3D printed.
That’s exactly why; like parts made of 6061
Hard to get students crammed around it and reading properly. As mentioned the controller of this system has presentation modes. Maybe we could hack something like that together on our similulator, bit unless we taught actual g code classes I don’t think it would be worth the effort
they don’t say how strong the spindle is, but even a 1/6th hp dremel runs 12000 rpm and will grind metal. Unless this spindle is a total POS, i’d bet you could cut aluminum and brass on it.
One of the videos I saw had it cutting aluminum and it seemed to be fine. However I’ve not seen any spec for it. I don’t suspect you would want to take out half inch paths at high feed but I’m sure it can handle aluminum overall
I would also favor the pocketnc if we were looking for another machine to expand capability. 5 axis + modern controller would handle 90% of a makers CNC wishes, like small intricate 3d geometries (that are massive in size in gcode, compared to what our HAAS can hold in memory).
Haas technically doesn’t have a limitation as it can dripfeed code. Did this for years cutting injection molds without ever having a catastrophic glitch.
that is true TODAY on newer machines. Our HAAS was born in the last century. Even in drip mode it has what is termed an ‘integer limit’ which is 256 Kbytes. Yep, it will just stop after 256k. I know - I was the first to hit that limit. Today’s controllers can stream up to the limit of the disk drive. I’m sure it all has to do with whatever 8 bit processor is in our controller. Imagine this, in 1969 we put a man on the moon (with non-metric tools) with what? 16K of memory? So how were the control designers of the day ever expected to ever create a program as large as 256K? Not to mention people were hand coding and producing code much more efficient than that of the sloppy programs available today.
I worked for AT&T Bell Labs when we produced THE worlds first 1 megabit memory chip. Cost in 1986 for that was about $1000/chip. What’s a terabyte cost now? about $30? Technology has progressed quite a bit since they cranked out our 1993 controller. And incidentally, there is no real upgrade path for ours, only a replacement path. We know - we’ve searched. A new controller today would probably cost more than what we paid for the HAAS.
There are of course workarounds. 2d toolpaths aren’t generally large and complicated and so you simply post each toolpath separately. 3D toolpaths are a different matter. My biggest one is 330K. that is one toolpath. that’s too big for our HAAS (which just FYI is the Robert Johnson III memorial HAAS) in streaming mode. It’s a little more complicated breaking up that file so that it picks up where the previous one ends. I can do it but not everyone can so they are kind of stuck if doing 3D.
IF we ever move, we have a CNC machine for the plastics area which is PC based and will be able to handle large plastics or composites jobs. Whether we allow it to be used for soft metals will be up to the committee to decide.
==== edit ====
I should point out that our controller on understands numbers up to 4 digits. It will give a range error once it hits line number N10000. You can remove all line numbers and white spaces from within CMEdit but you still have the 256K streaming limit in the controller.