What's new
What's new

Deckel control emulator

legoboy

Hot Rolled
Joined
Nov 14, 2011
Location
Alberta
Has anyone taken the time to reverse engineer any of the popular Deckel controls and build an emulator so you can run your machine off a different computer? Similarly to the emulators they use to run old arcade games. If this is a ridiculous notion please be gentle;)
 
Not seen anyone make a Dialog simulator......If this is a simple key in setup (does not generate code from geometry) don't see much advantage to doing it off the machine.

Better i think to spend your time and money on a CAM system that incorporates a verify program and creates code off of graphics....(saves time and errors)

Dialog 3 and Dialog4 have built in graphics to verify the program,and its possible to check the geometry as it is written by verifying using the graphics in a spot check manner plus those programs do give error codes of programming mistakes providing a code for the error and the block number where the problem is located (sometimes).
Cheers Ross
 
I was thinking on the lines of being able to replace the existing computer with a much newer one so as to be able to increase the memory and such. By emulating the original control it could still operate the machine as originally intended but allow you to run much bigger programs.
 
Not sure i understand..Are you suggesting a new control (retrofit?) that works like the original?
Cheers Ross

I would like to upgrade the computer part of the control if there was a way of doing that. Or is there no way of seperating the "brain" from the rest of the control so i the case of the Siemens the end board where the software is written and the programs stored. So kind of a half retrofit.
 
Don't think what you are suggesting is very practical.
There were several folks years back that worked on making a more modern version of the "control"...while trying to retain the basic architecture.
One of the individuals was a professional working in motion control and pretty sharp...Project died because of the complexity and a very limited marketability.....

Retention of the original control's functions and user interface...one would be IMO better off working at making the interface of one of the open source control setups work like a Dialog while replacing all the
original hardware...(retrofit)
But for most to my thinking, if you are going that far, having modern features and functions would be more desirable.....
Cheers Ross
 
Is there any way to increase the memory? Other than to try and find old boards and pay way to much for them. Even then the best I could hope for is to double the memory which isnt saying much.
 
Better solution if you have a Dialog 4 control is to explore the Deckel DNC arrangement....
Allows unlimited program size.
I have this setup and it works fine....Have run programs that have well over 80,000 lines of code, no problem.
Requires a hardware change on one of the boards (NEP 52A i believe), plus the latest software in the control ( 3.07) and a hosting program that runs on the sending PC....

Think there is additional memory as an option..never looked into that and i wonder if even having more memory would jump the 9999 line limit for program blocks.

Cheers Ross
 
Think there is additional memory as an option..never looked into that and i wonder if even having more memory would jump the 9999 line limit for program blocks.

Deckel Doctor told me he either has or is working on a large memory option for D4s, it sounded cheap.
 
II also have a 3M on my FP3NC with NCT400 table. I think they did offer a larger memory as an option.
I have the 14k one. I don't think there is a dnc upgrade but check with DD. Are you using the canned cycles and subroutines? Do you use parameter programming?
I do a lot of 4 axis machining and havent run out of memory. I don't do a lot of 3 or 4 axis contouring,
but when I do I put part of the contour in a subroutine and rotate it or mirror image it.
You could delete some or all of the canned cycles and upload the ones you want when you need them.
Dave
 
The problem is the cam, it likes to write everything long hand. I have tried editing my post to work with canned cycles but it hasn't been a full success. I will just have to get better at hand writing my programs. I am using the machine for inlay work so lots of parts and pockets of 2d shapes. When I am at my day job everything I do is on cam. When I am at home I have to go back in time 30 years in technology lol.
 
If you are running CAM generated code and doing 2-D profiles....Check to see if your program can convert point to point to arcs and circles...that way your control will use its built in G03 and G02 code rather than thousands of individual points.
Does not work for 3-D contours ...but on 2-D stuff it saves lots of memory.
Cheers Ross
 
Has anyone taken the time to reverse engineer any of the popular Deckel controls and build an emulator so you can run your machine off a different computer? Similarly to the emulators they use to run old arcade games. If this is a ridiculous notion please be gentle;)

I have taken the time to dig into the Dialog 11 system. More than 1000 hours. Sometimes I wish I didn't but it's a fun project and this year I've spent around 300 hours on it.

All boards (CPU, memory mapping, IO chips, IRQs etc) except the NZP are now implemented in software. The NPC board which uses the Z8000 CPU was the hardest which I finished 2 months ago. Selftest pass and I can navigate around the menus.

Not sure what will come out of this in the end. If I don't get the time to finish it and make something good out of it I'll probably just open-source the software and knowledge I've learned.
 
This sounds like a cool project: I hope you open-source it for the rest of the world. I am curious about a few points:

- Did you copy the firmware/software off of your hardware and incorporate it into the emulation?

- Is it purely a software emulation (including real-time interrupts etc) or is it also a hardware emulation? By a hardware emulation I mean you can "cut a wire" or "burn out a gate" in software and observe/study the effect.

Cheers,
Bruce
 








 
Back
Top