"Tom Gardner" wrote in message
...

Lloyd had caused me to remember a Trilogi PLC that I had bought many years
ago for some project that got shelved and had forgotten I had. So, with all
the wonderful advice in my last thread, I found the PLC and started
reading.

I found a nice, free beginner's book in ladder.
http://www.tri-plc.com/reqplcbook.htm

I've passed my first stumbling block and I can "see" it in my head once I
understood some misconception I had. PLCs are WAY cool! I imagine all
kinds of possibilities. The only PLC machine we have now was farmed out to
a "guy" that one of the mechanics knew from a previous job. This guy did a
great job and the machine works just fine but I haven't even looked at the
program yet.

Thanks for all the ideas and support guys!

They sure beat the heck out of wiring and re-wiring a bunch of relays!
Sometimes I get something running quick and dirty, then clean it up and
refine later. I have found hind sight is generally better than fore sight
so after the machine has been running a while I see if there are ways to
make it better.

Flow charts are more like a series of steps while the PLC is more like
everything is running in parallel. Once you get your flow chart steps
mapped out to PLC conditions you can make things work like your flow chart.
Starting out I had some problems with things executing when I didn't want
them to, like performing a calculation, one shots or edge detection type
instructions (depends what they are called on your PLC) solve that problem,
they can be used to cause something to happen once per cycle.

Example
if step3 then increment some number.

this will cause the "some number" to increment every scan while the program
is on step 3. To solve this you would do something like:
If step3 and one-shot then increment some number.

The one-shot will be true only one scan cycle when "step3" becomes true
causing "some number" to increment one time per "step3".

When I was doing PLC programming for customers I had to use the PLC they
wanted. I learned to program using mostly a few instructions that were
common to most PLC's, it kept me and the customers technicians from having
to learn their PLC's unique advanced instructions.

RogerN

On 5/6/2012 7:43 AM, RogerN wrote:
"Tom Gardner" wrote in message
...

Lloyd had caused me to remember a Trilogi PLC that I had bought many years
ago for some project that got shelved and had forgotten I had. So, with all
the wonderful advice in my last thread, I found the PLC and started
reading.

I found a nice, free beginner's book in ladder.
http://www.tri-plc.com/reqplcbook.htm

I've passed my first stumbling block and I can "see" it in my head once I
understood some misconception I had. PLCs are WAY cool! I imagine all
kinds of possibilities. The only PLC machine we have now was farmed out to
a "guy" that one of the mechanics knew from a previous job. This guy did a
great job and the machine works just fine but I haven't even looked at the
program yet.

Thanks for all the ideas and support guys!

They sure beat the heck out of wiring and re-wiring a bunch of relays!
Sometimes I get something running quick and dirty, then clean it up and
refine later. I have found hind sight is generally better than fore sight
so after the machine has been running a while I see if there are ways to
make it better.

Flow charts are more like a series of steps while the PLC is more like
everything is running in parallel. Once you get your flow chart steps
mapped out to PLC conditions you can make things work like your flow chart.
Starting out I had some problems with things executing when I didn't want
them to, like performing a calculation, one shots or edge detection type
instructions (depends what they are called on your PLC) solve that problem,
they can be used to cause something to happen once per cycle.

Example
if step3 then increment some number.

this will cause the "some number" to increment every scan while the program
is on step 3. To solve this you would do something like:
If step3 and one-shot then increment some number.

The one-shot will be true only one scan cycle when "step3" becomes true
causing "some number" to increment one time per "step3".

When I was doing PLC programming for customers I had to use the PLC they
wanted. I learned to program using mostly a few instructions that were
common to most PLC's, it kept me and the customers technicians from having
to learn their PLC's unique advanced instructions.

RogerN



I understand just enough to start to "see" it in my head. This project
is perfect for learning, it's just replacing 8 mechanical relays and 8
switches. No counters, timers or math to do. I plan to build a trainer
and explore more uses and get all my staff interested.

The most complex machine I want to control is for making twisted knot
brushes. I envision a machine that will be programmed for the many,
many variables that are a total BITCH on the mechanical adjustment/relay
systems I use now. I have four of these machines that are hydraulic
powered, A new machine will be servos and will make any size, any
twist, any other variation all with a change of programs and a change of
quick change of fixtures.

"Tom Gardner" wrote in message
...

snip

I understand just enough to start to "see" it in my head. This project is
perfect for learning, it's just replacing 8 mechanical relays and 8
switches. No counters, timers or math to do. I plan to build a trainer
and explore more uses and get all my staff interested.

The most complex machine I want to control is for making twisted knot
brushes. I envision a machine that will be programmed for the many, many
variables that are a total BITCH on the mechanical adjustment/relay systems
I use now. I have four of these machines that are hydraulic powered, A
new machine will be servos and will make any size, any twist, any other
variation all with a change of programs and a change of quick change of
fixtures.

Like electric servos with ball screws/linear actuators? At the place I used
to work we had a bender that was programmable. It used a hydraulic cylinder
with LDT position feedback and either a proportional valve or servo valve.
I designed a servo hydraulic press for a GM line that used an Allen Bradley
1771-QB linear positioning module, MTS TempoSonic linear displacement
transducer, and Atchley servo valves. This was high $ stuff but it can be
done for less money. The press for the GM line pressed control arm bushings
in the control arms and measured the position of the parts while being
pressed, the idea was to get more consistent front end alignment.

For the Allen Bradley card, once everything else was set up, you just had to
set a position and tell it to move, the card controlled the valve and read
the position to move the hydraulic cylinder to the set point. You should be
able to set up something to control position with an analog output and scale
the output to inches, mm's, or whatever units you want. It's do-able with
electric servo motors or hydraulics, whatever works best for your
application.

For your application where you change setups, you can use different sets of
data and select the one you want to run. For example if you have 100
variables to define a "recipe" for a certain brush type, then just select
the "recipe" for the type of brush you want to run. If everything is servo
adjustable your changeovers can become simplified.

RogerN

On 5/6/2012 5:41 PM, RogerN wrote:
"Tom Gardner" wrote in message
...

snip

I understand just enough to start to "see" it in my head. This project is
perfect for learning, it's just replacing 8 mechanical relays and 8
switches. No counters, timers or math to do. I plan to build a trainer
and explore more uses and get all my staff interested.

The most complex machine I want to control is for making twisted knot
brushes. I envision a machine that will be programmed for the many, many
variables that are a total BITCH on the mechanical adjustment/relay systems
I use now. I have four of these machines that are hydraulic powered, A
new machine will be servos and will make any size, any twist, any other
variation all with a change of programs and a change of quick change of
fixtures.

Like electric servos with ball screws/linear actuators? At the place I used
to work we had a bender that was programmable. It used a hydraulic cylinder
with LDT position feedback and either a proportional valve or servo valve.
I designed a servo hydraulic press for a GM line that used an Allen Bradley
1771-QB linear positioning module, MTS TempoSonic linear displacement
transducer, and Atchley servo valves. This was high $ stuff but it can be
done for less money. The press for the GM line pressed control arm bushings
in the control arms and measured the position of the parts while being
pressed, the idea was to get more consistent front end alignment.

For the Allen Bradley card, once everything else was set up, you just had to
set a position and tell it to move, the card controlled the valve and read
the position to move the hydraulic cylinder to the set point. You should be
able to set up something to control position with an analog output and scale
the output to inches, mm's, or whatever units you want. It's do-able with
electric servo motors or hydraulics, whatever works best for your
application.

For your application where you change setups, you can use different sets of
data and select the one you want to run. For example if you have 100
variables to define a "recipe" for a certain brush type, then just select
the "recipe" for the type of brush you want to run. If everything is servo
adjustable your changeovers can become simplified.

RogerN



I have the basic machine designed, all electric except for one air
cylinder. One servo twists, another pulls the twister back and a
stepper indexes the disk. The disk holder has two stations so the
operator is loading and unloading one station while the other station is
being twisted. The whole machine will be about 3 cubic feet. My
current machines are 4'x4'x8'. Ain't technology wonderful?