Mori MV 35/40 retrofit

4 Aspirin later, I think I understand how the parts are chosen.

Working backward,,,

From the number of inputs and outputs you determine the Isolated I/O card requirements divide your needs by 8 output, 16 inputs to know how many 7I37 you need

From the number servos you determine which ( and or how many) Analog Servo Interface are needed.
For 3 axis and a spindle can a single 7i33TA 4 axis analog servo interface be used?
For more axis either multiple 7i33TA or a 7i48 would be needed.

From the number of Isolated I/O cards and analog servo interface cards you know how many connectors you have.

If it is 3 connectors, you can use a 5i20
If it is 4 connectors, you can use a 5i22
If it is between 5 and 6, you can use 2 5I20
If it is between 7 and 8, you can use2 5I22

My machine has servos for 5 axis but I don't have the rotary tables so I think i will go 3 axis for now just see if i can do this.

So my parts list would be:
2, Isolated I/O cards @ $79each = $158
1, 7i33TA 4 axis analog servo interface = $79
1, PCI I/O card 5I20 = $199
Misc cables = $100
1, HP xw6000 leftover computer

Total est cost = $536

Is this about right?

Just wondering what parts did Chris use on his MVJR shown in this image?

I started laying out the logic for the toolchanger today and have most of it figured out and should be able to be done pretty easily in ladder form. the only thing that I'm having an issue figuring out is how keep track of which tool is in which pot and how to call that pot to the tool change station. the magazine has a binary rotary encoder to track its location.

I was told this by Chris: It might be similar to your machine too.

"The only hardware modifications I made to the machine were to fix the
feedback from the carousel. It has three proxes but would still lose
position if it went past a pocket or stopped at the wrong time. (it
uses a contactor to run DC injection for braking the carousel; if it's
at all sticky it will sometimes miss the target pocket) I have heard
the original control on these would occasionally get confused about
carousel position and even load the wrong tool. This is just a design
defect in the feedback, and I fixed it (one prox is an index sensor
for pocket-1 position, and the other two were set up stupidly. I
moved one to make them quadrature so it can't get lost.) I have never
had a problem with the fixed tool changer."

OK thanks john it is a random arm style tool changer.

Jon 16 outputs will not handle the machine operations at least if the tool changer is setup like mine. my machine has en encoder on the magazine instead of prox sensors. it is all hydraulic with a slow down circuit that can be initiated one bit before the requested bit.

Tool changer ouputs -13

magazine rot cw
magazine rot ccw
mag slow
mag free
pot up
pot down
arm up
arm down
arm rot 0 deg
arm rot 180 cw
arm rot 0 deg
arm rot 75 ccw
air thru spindle to remove chips

other outputs -(10-13)

tool clamp
tool unclamp

coolant pump on
air blast on
oil shot on
coolant thru tool spindle on

4th axis lock solenoid

tool orient
spindle forward
spindle rev

hyd pump start
spindle pump start
servo on power
could maybe pull these 3 off of servo card? or combine into 1


every time i make this list i come up with more lol.

If you are not in a big hurry there a some cheaper and denser I/O options coming.
For example a 5I23, a 7I48, a 7I44, one 7I70 and one 7I71 would give you 6 analog servo axis plus 6 encoder inputs(7I48), 48 24V 300ma isolated outputs (7I71), and 48 isolated inputs(7I70). with expandability to 384 real time I/O points by adding more 7I70s and 7I71s. This also leaves one 5I23 connector free.

When will this hardware be available and supported by emc2? this sounds like the way to go.

Thanks
Cory

Regarding the mother board, if I understand correctly you will need two PCI cards so the D510MO won't work for that, you need to find a board with 2 PCI slots.

I'm wondering, how to choose a mother board; what lead you to the D510MO to begin with?

f you are not in a big hurry there a some cheaper and denser I/O options coming.
For example a 5I23, a 7I48, a 7I44, one 7I70 and one 7I71 would give you 6 analog servo axis plus 6 encoder inputs(7I48), 48 24V 300ma isolated outputs (7I71), and 48 isolated inputs(7I70). with expandability to 384 real time I/O points by adding more 7I70s and 7I71s. This also leaves one 5I23 connector free.

When will this hardware be available and supported by emc2? this sounds like the way to go.

Hardware will be available by the end of this month but The hostmot2 driver needs to include new code to support the 7I70 and 7I71. I expect the may take a couple of months.

Also there are so limitations on the 7I70 and 7I71. The 7I70 48 input module takes active high inputs and has a common (isolated from PC) ground. That is it senses + voltages on its inputs, so your field wiring need to match this. The 7I71 48 output module is a sourcing driver, that is it has a common +5 to +24 control power input and sources power to 48 loads. Again field wiring needs to match this standard, with load common connected to control power V-.

Jon 16 outputs will not handle the machine operations at least if the tool changer is setup like mine. my machine has en encoder on the magazine instead of prox sensors. it is all hydraulic with a slow down circuit that can be initiated one bit before the requested bit.

OK I see, your plan seems right for me too then.

So my parts list would be:
4, Isolated I/O cards @ $79each = ~$320
1, 7i33TA 4 axis analog servo interface = $79 (assuming this can run 3 axis and a spindle)
2, PCI I/O card 5I20 @ $199 = ~$400
Misc cables = $150
Mother board with 2 PCI slots, any thoughts on that?

I'm looking at:
Intel® Desktop Board DG41WV
www.intel.com/products/desktop/motherboa.../DG41WV-overview.htm
It seems to have two PCI slots, but I'm not up to date on mother boards anymore.

Touch screen, any thoughts on make, model and source for that?
I'm looking at:
ELO TOUCHSYSTEMS 1522L(E518492) 15"
www.newegg.com/Product/Product.aspx?Item=N82E16824217102

Total is getting closer to ~$2,000, but it is looking like an interesting and rewarding challenge.
The old yaskawa boards are probably worth that, unfortunately i know what it is like to need those boards.