MDC ECM series (ECM93, ECM95, ECM97) pick and place info/SW WTD

[philpem]

Hi folks,

One of my many ongoing projects is repairing an MDC (also sold by Manncorp and Mekko, but made in Japan by MDC aka Mamiya - not the camera company!) ECM97 pick-and-place machine. Thus far, I've got most of the mechanical issues sorted, but I'm curious if anyone else happens to have one of these machines, any documentation or software.

I've scanned the manuals I have for mine (an ECM97CWL - Conveyor Feed, "not sure what W means", LaserAlign) and have an image of the hard drive - including the LaserAlign diagnostic software and the Panasonic PANATERM servo drive config tool (which I haven't been able to get working... yet).

What I'd be really interested in is a copy of any of the following:

- Original software install disks (rare as rocking horse manure from what I can gather)
- Documentation
- Documentation for the PC I/O cards - the Kumagai KP422 motor control card, and the "digital I/O card". I'd also love to know what the connectors on the Kumagai card are. They're almost like Hirose FX2s, but not quite.


Nice to haves:
- A spare servo drive (Panasonic MINAS Ex Series or similar) and motor to run up on my bench. Needs to be cheap. I'll probably start a thread in For Sale And Wants for this thing.
- A replacement LaserAlign HD sensor head. Mine has a half-dead laser. I'd like either another one with a dead laser (so I can play surgeon and still have one working-ish unit) or a working unit. I'm not picky.


I've got a load of scanned documentation for my machine, which I should be able to put on an FTP soon-ish. The only thing that's proving difficult to get scanned are the A2-size fold-out schematics (!).


My eventual goal for this thing is to replace the clunky (read: buggy) Pentium PC and DOS software with something a bit more modern (probably a RasPi mated to a custom STM32 or similar motor control board), then add on bottom-up and top-down vision via a few USB video capture pods. But that's a pipe-dream project. Something to keep me occupied over the course of a year or so, or longer 

And then make boards with it. Of course. 

Cheers,
Phil.

[sandyc]

Hi Phil,

I have 2 EMC97 machines (one with laser no conveyor and one without but with conveyor).
Documentation (or what there is) is atrocious. I do have some manuals and importantly the schematics to repair, but no original software - although what I have on the screen seems to be older than the software manuals I have managed to find on the web.

The operating software isnt much better - although to be fair I have seen some "real" machines that have the same type of interface.
However that said as an operating machine they seem to work pretty well (touch wood) and we have boards running through them regularly now without too much grief.

Like you, one of my background projects is to separate the control system from the drive system so that Im not reliant on maintaining 15 year old PCs and floppy disks, and then to create a modern user friendly interface for operating these monsters.
Swap out the scratchy camera/TV interface with a couple cheap HD webcams.

Im heading toward a distributed motion system using something like CAN with a Raspi CNC control/setup interface and then ethernet/wifi for remote control/setup from my desk in the warm of the office.

Happy to swap ideas or help out.
Sandy

[philpem]

I have 2 EMC97 machines (one with laser no conveyor and one without but with conveyor).
Documentation (or what there is) is atrocious. I do have some manuals and importantly the schematics to repair, but no original software - although what I have on the screen seems to be older than the software manuals I have managed to find on the web.

I'd be interested to know what schematics you have -- I have a wiring diagram for my machine and some schematics, but the main one I'm missing is the schematic for the board which hides behind the circuit breaker box in the cabinet.

One of these days I need to open up the control box and read out the tape feeder sequencer EPROM...

The operating software isnt much better - although to be fair I have seen some "real" machines that have the same type of interface.
However that said as an operating machine they seem to work pretty well (touch wood) and we have boards running through them regularly now without too much grief.

We had a Multitroniks at work until it was scrapped off; it wasn't much better from what I recall. Perhaps slightly more user friendly and a few less bugs.

Like you, one of my background projects is to separate the control system from the drive system so that Im not reliant on maintaining 15 year old PCs and floppy disks, and then to create a modern user friendly interface for operating these monsters.
Swap out the scratchy camera/TV interface with a couple cheap HD webcams.

Mmm. My main sticking point is time (or lack thereof), but the secondary sticking point is not being able to find any info on the Kumagai I/O board or obtain any of the connectors for it...

Im heading toward a distributed motion system using something like CAN with a Raspi CNC control/setup interface and then ethernet/wifi for remote control/setup from my desk in the warm of the office.

That sounds pretty similar to my idea -- CAN sounds like an interesting idea. I was thinking USB, but CAN or LIN might be easier to implement. Hmm!

Cheers,
Phil.

[wzehler]

Hello Guys,

I just purchased two used ecm97 machines and I am looking for any manuals or software I can get.   One machine seems to work just fine and the other tries to home but fails on the y axis.  The bios reports 3.7 volts and should be 3.3 volts and yells at me.  I will change out power supply on the computer but I don't think that is the problem.   If anyone out there has spare parts for sale please contact me.  I purchased these to play around with and maybe write some new software.  I will share what I find in the future.

Thanks, Bill 

[philpem]

The homing failure on the Y axis is, I'll bet, a dirty or misaligned home sensor or a slack belt. There's an adjustment procedure for this in the manual.

I think the homing sequence is something like this:

- Move the head slowly towards zero until the "home" sensor activates.
- Move the head off home a few steps (may only be one step) - if the home sensor doesn't clear, raise a homing failure.
- Move the head back home

If the sensor is stuck, the head won't move and it'll fail homing.

If the belt is slack (tensioner not tight enough), you'll get misalignment. I can't honestly remember which axis is which, but the working machine should give that away.

I'm still hunting for any info that's floating around - I hear the BASIC source for the machine control software is "around", and I'd love a copy if anyone's willing to provide it. I'd love to replace the DOS control software with a Raspberry Pi, but while the electronics look easy (make a board to run Marlin and add some G-code commands for Pick-and-Place?), the head change movements have thus far eluded me!

[philpem]

Just a brief update from me -- please pardon the necrothread but I thought it best to keep things together!

Someone very kindly contacted me to offer the parts from a broken ECM97. I now have a complete set of the major components to keep mine running, including some extra nozzles.

This also included a hard drive from their ECM97, which turned out to include part of the ECM source code as deleted files on the hard disk. Sadly they were partly overwritten; there's about 10K of code missing.

If anyone has the hard disk or install disks for an ECM series machine (ECM93, ECM97, ECM98), it would be great if you could provide me with a "dd" format disk image. Hopefully with data from another machine, the files can be reconstructed.

I'm going to try and update my website soon - I have some wiring diagrams scanned which will be added too, along with the incomplete source code.