Re: real time linux

"John R. Carroll" <jcarroll@xxxxxxxxxxxxxxxxxxxxx> wrote in
news:b7ntf.48852$6e1.25806@xxxxxxxxxxxxxxxxxxxxxxxxxx:

<snip>

>> One of the technical issues that gets overlooked is that by its
>> nature this type of system is open loop. It really shouldn't be a
>> problem with the types of drives available today, but I can see where
>> it would be a factor in certain applications.
>
> The loop is closed Dan, just not the way you are used to seeing it
> closed. All of the data is precompiled and checked in advance so all
> you need to do is close the internal servo loop and not the entire
> process. You are still accurately moving the machine but the loop is
> "pre closed" if you will. As long as your hardware doesn't bomb out on
> you you have gotten the woopsies out of the process in advance of the
> drives ever seeing a single thing.

I probably should have been more specific. In a conventional control (if
there is such a thing) there is feedback from the encoders to the
positional command generator in the control (as well as the drives). In
the Tornos PNC sytem there are no positional commands being generated,
that data is compiled. The encoder of course feeds back to the drive. The
main difference is that if something starts to go off in the conventional
system, the control can generate some correction. The PNC system can not.
In either case, if the in position tolerence width is exceeded the
machine will alarm out. Like I said, it's a very minor difference and for
most applications it shouldn't really be an issue.

>
>
> This is what the Heidenhain solution looks like internally.
>
>
> 1) Tell the machine about the part:
>
> BEGIN PGM 1 MM
> BLK FORM 0.1 Z X-50. Y-50. Z0.
> BLK FORM 0.2 X50. Y50. Z100.
> ;
> * - BAUTEILNAME DMU50EV_MSSI.vmp
> ;
>
> Tell the control to pay attention:
>
> * - EILGAENGE
>
> Tell the machine to load the the stock model:
>
> FN 0: Q30 =50000 ; EILGANG
> FN 0: Q33 =10000 ; ABFAHRVORSCHUB
> ;
> * - A-45. C-90.
> * - Standard Drill
> * -
> MSYS/50.000000,-50.000000,75.000000,0.000000,1.000000,0.000000,-0.70710
> 7,0.0 00000,0.707107,0.707107,0.000000,0.707107
> * - GAGELENGTH/90.000000
>
>
> * - VHM SPIBO 9.9
>
> Tell the machine what feature to machine:
>
> TOOL CALL 0 Z S0.0 DL0 DR0
> M13
> CALL LBL 99
>
>
> All of this stuff comes from the uploaded CAD database, internal
> kinematics model and the internal tool library, just like Seimens.
> STEP is a standard so it is possible to have certain expectations
> about what is avaliable:
>
> FN 0: Q31 =0 ; FRAESVORSCHUB
> FN 0: Q32 =1000 ; Z-VORSCHUB
> ;
> L X50. Y17.678 R0 FQ33 M7
> L Z10. R0 FQ30
> CYCL DEF 200 BOHREN ~
> Q200=+2 ;SICHERHEITS-ABST. ~
> Q201=-15.283364 ;TIEFE ~
> Q206=+Q31 ;VORSCHUB TIEFENZ. ~
> Q202=+15.283364 ;ZUSTELL-TIEFE ~
> Q210=+0 ;VERWEILZEIT OBEN ~
> Q203=+0 ;KOOR. OBERFLAECHE ~
> Q204=+50 ;2. SICHERHEITS-ABST. ~
> Q211=+0 ;VERWEILZEIT UNTEN
> L X50. Y17.678 R0 F MAX M99
> L Z10. FQ21
> * - Standard Drill
>
>
> Now you go do it.
>
> The STEP standard is fully open. The machining algorithms aren't but
> can be made so if desired.
> As a former application guy, you could teach a control to do what you
> want, the way you want it done and then ship that knoweledge base with
> the other standard AI to the customer. IOW, custom canned cycles if
> you will. The necessary slicing and dicing is done internally and
> compiled into binary drive source, stored and then executed.
>
> Voila. No mas external CAM system and you can build increasing
> intelligence into the technology as you go.
>

That looks very similar to the limited amount of code I've been able to
see on the Siemens control. With the Tornos PNC system, the data tables
are generated by the CAM system. Their CAM system outputs conventional
CNC G&M code, so that manual edits can be made easily, then that code is
compiled into data tables. The main drawback is that you have to make all
edits on the CAM system, recompile, then reload the program. I've heard
that they are either working on or have a way of doing this on the
machine now. Clearly the STEP standard has some advantages over the PNC
system. The underlying idea of compiled data seems the same.

I still wonder about the patent. I looked it up and it seems to me at
least that these new controls would have to use technology which is
covered by the patent. Here is a link: <http://tinyurl.com/7bs3z> You
need a tiff plug-in for your browser to view the drawings.

It could be that since all of the controls that are capable of running
compiled programs are European, Tornos has licenced the technology. But
even that seems unlikely. If there is some fundemental difference in the
control technology I'd be interested in finding out what it is. I mean
outside of the STEP standard and the ability to edit and compile on the
control. (An obvious advantage IMO)

--

Dan

.

Relevant Pages

  • Re: Compile OpenNetCF - OpenNETCF.Windows.Forms
    ... http://vault.netcf.tv/VaultService/VaultWeb/login.aspx (user guest, password ... If you have the latest source you'll see that the HTMlViewer control has ... > I'm trying to compile OpenNETCF.Windows.Forms.dll and use it in my project ... > instead of standard OpenNETCF.Windows.Forms.dll. ...
    (microsoft.public.dotnet.framework.compactframework)
  • Re: 4-20mA Analouge input
    ... Im sure there is a standard im meant to be following but im not sure ... between field sensors and process control ... electronics in the control house. ... loop voltage also 'powers' the field sensor device. ...
    (sci.electronics.design)
  • Re: Stevey and overengineering
    ... It baffles me that this hasn't become standard C/C++. ... Fails to compile as C89 ... Scoping bug in some compilers doesn't lock down x to the loop body ...
    (rec.games.roguelike.development)
  • Re: Colors of butons with XP skins
    ... For what it's worth, I believe you can copy the XP loop, but it would require quite a bit of control customizing for each control and probably not be worth it. ... you will need to copy the XP-style images from each control into an external program and then copy them back into standard controls. ...
    (comp.lang.labview)
  • Re: C#Control.>VB.NET
    ... For the others who may be searching the archives, the "standard" way ... is to compile the C# control into a dll, add that dll to the project ... and use the control like normal. ...
    (microsoft.public.dotnet.languages.vb)
Loading