Feasibility of hacking/rebuilding commercial plotter

Projects General Guidance
1

Hi all, ignorant noob here looking for advice on starting a project...

I have an old Graphtec FC4100-75 plotter that took a poop on me, the best I could troubleshoot was bad rom/ram, which lead me to the idea of replacing the plotter 'brain' with an Arduino. This is a heavy duty plotter I used to cut sandblasting stencils, also capable of cutting rolls of magnet and other tough media. It's a well built machine, in my opinion, other than the dead board.

I've long admired the ranks of you maker types, this is my excuse to start learning.

In discussion of taking on this project we concluded doing away with all the bells and whistles of the plotter's sensors for simplicity sake, really just looking be able to send a path file like SVG or Gcode from a design software such as Inkscape to plot my designs.

From what I understand as a bystander of what you fine folks do here with Arduinos, I thought it should be a simple enough project to control the two motors and pen up/down (X,Y,Z) but after a great deal of searching I have come to the conclusion I may be an idiot for taking this on without previous coding/hacking experience, hoping you guys could confirm or deny my conclusion... ha.

My partner and I opened up the plotter and determined it has two optically encoded DC motors for the x and y (NOT STEPPER MOTORS), but I have not been able to find any specs on the motors, I would think that would be the first step. Closest thing I've found is info on the 'series' of motors called the Micro J series by Yaskawa, but no specs/details. Info printed on the motors are:
Minertia Motor
UGJMEE-A7MGR71
905745-1

Pointed questions are:

  1. Is it a feasible project to replace the computing components of an old serial plotter with an Arduino?
  2. Is there a good starting source of info for setting up Arduino driven plotters from a hack like I've described above?

So far I've filled my brain with hours of searches on arduino x-y plotters, motors, Gcode, drivers, and shields... like I said, total NOOB here and feeling a bit lost, thanks in advance for your time and expertise.
Best wishes,
Neal

2

Yes it should be entirely possible to control the plotter with an Arduino, some motor drivers, and a transistor or mosfet to control the pen lift solenoid.

If you can't find the motor specs, can you work out what voltage they work on, by looking at what voltage power supplies the plotter has? My guess is that it will be about 12V, but it could be lower. Also look on the circuit board for the motor driver ICs and see what type they are, which will give you an upper limit on the current they need. If you are unlucky, it may use discrete transistors in an h-bridge configuration instead of ICs.

Does the plotter have X and Y endstop switches for detecting the home position?

3

the whole unit is designed for a specific purpose. if you think it just needs new memory chips and probably new caps, then a few hours of re-work may bring it back to life.

if you want to swap it out, then figure out the encoders wiring and get new servo drivers. simple enough, guess that they are 1,000 pulse per revolution, command it to move 1,000 and then measure. if it moved 4 inches, then i is 250 pulses per inch.

I would opt for MACH and servo drivers and just convert it to full CNC. if your intended purpose is to run it as a PC based unit, with pictures and such, then that is the way to go.

4

Thanks for the quick responses guys.

in response to dc42:
We took an initial look at the transformer side of the unit, looks like several voltage supplies soming out, I feel confident we can figure out voltage at the motor side. There are several ICs on the board, I assumed some to be the drivers and will do some research in that direction, thanks for the tips. This is a rolled goods plotter, so only X has the endstop switches you asked about, the Y axis is a roller that feeds the media through, it has IR sensors for the front and back edge of media, but we were a bit intimidated from trying to integrate all that at first.

in response to dave-in-nj:
we did an initial sweep of the unit to see if there were any popped capacitors, we did not see any visual evidence of such. After scrambling my brains with initial Arduino research this weekend, I may be wiser to look into replacement ICs, but the idea of a complete CNC overhaul to a modern USB interface sounds heavenly. What do you mean specifically by 'opt for MACH'?

Thanks again to both you fellas for taking the time to response, I appreciate your assistance.
Neal

5

http://www.machsupport.com/

6

While it should be perfectly possible to control encoder motors with an Arduino I suspect you could spend a loooonnnng time fiddling around to get them to work properly if you have not done it before - even if you are good with software.

If this is a real tool (rather than a pastime) I would go for an off-the-shelf solution if one exists and if it is almost "plug-and-play"

HOWEVER - if the off-the-shelf "solution" needs a huge amount of tailoring you could easily waste more time trying to learn how it works in order to tweak it than the time needed to develop something from scratch.

...R

7

Assuming this thing interfaces with a computer, the tricky part might be in re-creating firmware that properly communicates with the driver, so that you existing application works with your new electronics.

8

Daniellyall: thanks for the link to Mach software! This looks like a great tool.

Robin2: This is a real tool for rubber stencil production, although a fairly simple one, but too old to find anything remotely "plug-and-play"... I hear you on the learning curve. The more I learn the more I realize what I NEED to learn :slight_smile:

We did manage to make a little more sense out of the project today, the encoder on the motor has 4 wires, (power, ground, A Channel, B Channel) so we're going to keep learning and tinkering to see if we can pull this off. The alternative is spending around $700 for a lesser quality machine, so I'm certainly motivated.

At this point I guess we'll just have to start testing, my partner picked up an Uno today, excited to see what we can make happen!

Thanks for the advice everyone, I'll post further questions in the appropriate places to keep the forum tidy.
Best Wishes,
Neal

9

NMFH:
The alternative is spending around $700 for a lesser quality machine, so I'm certainly motivated.

I love tinkering with things.

But you have to think about how many hours it would take to earn $700 - not that long, I suspect.

...R

10

Robin2, indeed $700 is a good price for a competent machine as compared to years ago when I bought this old power house, and if I were paying an employee to do this rebuild I'd lose the profitability game, for sure, ha.
I'm more of a challenge based mind, so for me, this project is basically my excuse to get motivated in learning electronics and code, even if I did 'bite off more than I can chew' :slight_smile: Very thankful to find this community of helpful minds here eager to help.
Best wishes,
Neal

11

If the general experience with video monitors is anything to go with, I would think the first step is to go through and remove each capacitor (especially the power supply ones) and test it with a capacitance & ESR meter - or simply replace them all holus-bolus. What they look like is irrelevant. You could get an indication by scoping the power supply lines for a start - AC chop on the supply lines causes all sorts of funny stuff on computer devices.

If then you were going to perform a "brain transplant", the most practical thing is to identify the actual digital interface point to all the motors and sensors and connect to that. Re-designing the motor drivers and sensor interfaces is equivalent to engineering the whole machine from scratch - you might just as well go into production of them if you propose to spend the years to do that.

If you have performed the first step above and not been successful, then it would also be a good idea to figure out how to read the ROMs and consider disassembly. Understanding the original code would be a huge step toward replicating the functions and could well indicate where the present fault is (and so, simply repair it).

12

NMFH:
Daniellyall: thanks for the link to Mach software! This looks like a great tool.

"Kcam" (www.kellyware.com/kcam) is a simpler, all-in-one tool ($95). Kcam takes the vector file (DXF), converts it to GCODE, then spits it out through the parallel port. Mach3 starts with GCODE then spits it out through the parallel port; you still need a separate CAM program with Mach3.

...not that I mean to disparage Mach3; it's an awesome tool when you already have a good CAM package.

13

I think the OP is a long way from using any of these products. First he has to figure out the interface used to control the various motors and whether they are compatible with any of them.

The Arduino has the advantage that he could try different short programs to explore how things work - whereas I think it would be "like a pig looking into an airplane" trying to figure out why nothing happened when Mach3 or Kcam was "switched on".

...R

14

I did a quick search to see if there was any firmware update available for that model. (unfortunately not). I did find a page with all the currently available documentation still available for it from the manufacturer.
http://www.graphtecamerica.com/graphtec-america-support-downloads-cutting-plotters-discontinued-legacy-cutting-plotters-fc4100-series

You can find the user manual and drivers for it there. All of which will be useless to you unless you can get it working again.

If you could post some clear images of it's insides (control boards, any markings on the motors etc..) maybe we can make some inroads into working out a strategy to move forward.

It could also be worth contacting the manufacturer direct to see if they still have a copy of the firmware available. Then maybe you could burn another EPROM to bring it back up.