That's a lot of info to cover there lol... my company makes precision gauges for industry and some have dimensional call outs that I can hold with the Makino. Flats, certain lengths, relief cuts, special profiles, etc that I can already hold to within a tenth or 2. I also use it as a micro boring bar, going from .020 to .030 to whatever size I need, I just bored 2 rings this morning from a .02 hole to .0594 and nailed it. The rings get lapped by hand 3 tenths or so to a mirror finish on the ID. I can cut those types of profiles perfectly but if I ask it to run as a basic cutoff with one rough pass E 1066 D.00787 (or so, I have the E and D offsets on my desk) across a pin the machine does not cut correctly. It hisses, pops, breaks wire, does myriad clusterpucks. My biggest problem now is bobcad post processing for the Makino U3. It causes errors in the machine to the point that I find myself re writing posts more than actually cutting. @cameraman how much editing to your NC gcode do you do before you cut your pieces? Please let me know, I want to collaborate with as many EDM makino owners as I can si we can all benefit from shared knowledge and experience! Feel free to email me ZerocheckEDM at gmail.
OK I kinda get where you are coming from on this now, thanks for the greater detail.
OK so coming at this as an Engineer that was a materials scientists and migrated to CAD/CAM and VR software development + machining, electronics etc. + design engineering.
Having looked at the 1000+ pages of all the permutations for different cuts (rough 1st pass, second, third... in some cases up to seven with associated Ra(s) and Rz(s), for different wires, different part thicknesses/ heights and different materials... ) That you are dealing with a "Matrix" with somewhere between 7 and 14 simultaneous variables all acting on each-other (simultaneously) with different rule sets for various in-cut real time conditions.
However these parameters and setting are derived from a set of core equations (electrically) but also empirically from test cut data. + whatever algorithms that are used to govern the various cutting strategies "Real time".
To my way of thinking "Intuitively" it would not be easy to write a comprehensive post for a pretty much any Wire EDM that has a lot of advanced internal cutting strategies that are not immediately obvious. I.e. not dialed back to simple/ dumb / very slow and conservative settings for very generic cutting conditions of various materials.
I would say in the case of BobCad unless you reach out to Makino and vice versa
to obtain / grab their full table of parameters which is about 1000 pages long (and deeper key information), and start to reverse engineer or back-map effectively some of the core equations + get some idea of the fuzzy logic that drives the machine or get specific development information from Makino themselves (like a development kit), then I could imagine things could be a bit hit or miss for a very generic post. If someone at BobCad doesn't have a U3 in front of them , then I would imagine writing a post for the U3 would be challenging / hit or miss unless Makino provided a good development kit for the machine. I.e. sounds like some of what you are experiencing @CleverUserName1
. Their post would have to be "verified" by them with a machine / U3.
That's a sh*t load/shed load ? of testing that needs to be carried out (even with core equations and algorithms) / and or a development library.
Maybe Monday or Tuesday I'll be able to dig through some of the data I have to see if there are inconsistencies or errors.
Sounds like you are pushing the machine to it's limits in a good way... I'd never have thought of using the machine that way... I'm mainly after a few straight cuts between one and two micron-ish over 30 mm and overall part / form tolerances of roughly 5 micron / two tenths or wider. For taller parts we can fiddle stuff and the design of the components we need have certain adjustments and are custom fitted to each other so high tolerances and close fits can be achieved but NOT for super high tolerance interchangeable parts
. <<<---- The exact opposite of your application @CleverUserName1
where you literally make high tolerance interchangeable parts with full traceability.
I do believe there is a visual way to represent these seemingly disparate "Multidimensional" parameters for the Raw cutting conditions and settings and modes, in a simplified visual form.
The thing that's interesting about the U3 / U3i ARE the development possibilities as a second CAD/CAM / PC driven / linked screen is possible as an option, that's probably why... In respect of what you are having "Plug and play" issues with. For me I see the opportunity , for others that may be a pain in the arse/ ass and not what they wanted to sign up for.
I have heard the 'Notion" of the complex stuff is "easier" and the simple stuff is more time consuming / fiddly on the U3 ? But haven't been able to pull sufficient focus yet as to why and how.
, I'm treading carefully here as I could see there could be "anti-competitive" law suits etc. / legal ramifications if such deeper development information was offered to one CAM development company and not another. ~ That's my IP / legal hat on there. Historically there have been some hideous law suits along similar lines.
Best to contact Makino directly.
Also what @SeymourDumore
hints / mentions with flushing considerations for incredibly tiny "apertures/ holes and features " yet very tall ???
Unless I have driven the U3 / U3i for at least 18 months I'm probably not going to be much help for the physical limits of the machine and the physics of surface tension vs. electrostatics and surface fluid dynamics / micro turbulence etc. That's very difficult to model and requires a lot of "Suck it" and "see" hands on work to map out the limits of extreme or unusual cutting conditions and geometry. Sort of reciprocity failure that might require very different settings and cutting conditions to execute properly.
On the other hand simple stuff should be simple ? Maybe there's a some hidden bugs ? (shrugging shoulders) - not easy to sort out without being on the shop floor hacking out code right next to the U3i etc. for about 18 months ...