[Date Prev][Date Next][Date Index]
re: Cheaper PLC's (Koyo)
Pete,
I have been looking at the Koyo PLC catalog as well.
I even have purchased their DIRECTNet manual(there RS422 programming
networking scheme) as well as the manual for the DL405 controller.
I have decided against the small brick PLC's just from the
standpoint that I have no way of replacing an I/O point if
it becomes damaged without replacing the brick. In my bending magnet
FOE, I have found that I need to monitor a total of 58 relay/reed
switch contacts as well as 1 Analog signal for ring current (its not
important for the bending magnet, but it becomes more important
for the ID line). I need to produce 12 24V outputs to control valves
and provide status to the APS. (NOTE: we are splitting our beamline,
so our bending magnet FOE is rather complicated.) One thought
is to use 1 PLC for the entire FOE and place it in a shielded enclosure
in the front of the FOE. I just finished specing a system using the
Koyo stuff. Here is what I have:
# Part Item Unit Total
DF-440-1 CPU $407 $ 407
CF-RAM-2 15.5 Kbyte CMOS RAM $ 98 $ 98
CF-08B 8 Slot Local Base $111 $ 111
D4-DCM Data Communication Mod $215 $ 215
D4-16SIM 16pt Input Simulator $122 $ 122
D4-16TD2 16pt 12-24 VDC Output $ 95 $ 95
4 D4-16ND2 16pt 12-24 VDC Input $ 72 $ 288
D4-04AD 4 ch 12 bit Analog in $274 $ 274
Total Cost $1610 list
BUT, Koyo gives educational institutions a 15% discount, so
total hardware cost is $1368.50.
While 32 pt input modules do exist, I prefer the 16 pt modules
since they can source more current and the screw terminals for the
connections are on the module. Otherwise, I have to provide
a separate terminal strip.
The input modules have LED's on them which reflect the input
signal, so I do not need a separate status panel.
The 16 pt input simulator is a module with 16 switches on it
that I would use to provide manual input to the system (i.e.
to activate valve opens and closes).
Just like other PLC's you need to purchase
a one time windows based programming package for $495 ($420.75
with discount)).
This program is for their entire family of PLC's and communicates
via RS232.
The Data Communication Module would be used to configure the 405
as a master in a master/slave network with other Koyo PLC's using
either RS232 or RS422. This leaves the serial port on the CPU
open for communication with a host computer. The networking
protocol is in the DIRECTNet manual and looks very simple (basically
it just involves requests for reads and writes to and from memory
locations) and includes BASIC source examples. If I didn't have the
DCM, I could still use the 405 as a master in the network using
the interfaces on the CPU, but I wouldn't have an easy way to
read the data if I felt it was very important.
The Koyo line does have remote I/O capabilities of various descriptions.
The real difference is in response speed. Since the EPS system is not
especially time critical, I would use smaller PLC's rather than
the remote I/O options. The main advantage I see here is flexibility and
the fact that the PLC's take 110 V whereas some of the remote I/O
options require 24 V (the PLC power supply has an external 24 V).
I may find that it is more
cost effective to have distributed I/O even in the FOE. If this turns
out to be the case, I would use their DL205 PLC controller with the
appropriate number of I/O points and connect it to my master with the
RS422 DIRECTNet. With this option, I would need something like:
1 D2-240 DL240 CPU $143 $ 143
1 D2-06B 6 Slot Base $111 $ 111
1 FA-UNICON RS422/RS485 converter $ 69 $ 69
2 D2-08ND3 8 pt 12-24 VDC Input $ 35 $ 70
2 D2-08TD1 8 pt 12-24 VDC Output $ 38 $ 76
1 D2-FILL 1 filler module $ 4 $ 4
Total Cost $ 473 (list)
$ 402 (discount)
The differential is really in the cost of the top 3 items. This
needs to be weighed against the cost of cabling. I will certainly
need these in my white beam experimental stations. So far,
my experience with their tech support has been very good and
I haven't purchased anything other than manuals yet.
Now, I am going to sound like a heretic, but I think it is easy
to go overboard on the EPS system. If the beamline
is operational, it should just sit there and chug along its merry
way with valves open etc... While I do see a need to monitor vacuum
levels and feed them into EPICS or other status monitoring equipment,
I don't see a real need to feed valve position or whether or not my
flow meter has triggered into EPICS. Using LOVE controllers, I would
get the analog information through a separate RS485 data highway anyway.
The main difference in
philosophy is that it seems like the APS is actually using their EPS
system for 'data collection' to measure temperature rises and performances
of components that they have designed. This requires that they use
more i/o points and also have a much closer tie between their EPS
and EPICS.
- John
--
John Quintana Internet email: jpq@nwu.edu
DND-CAT Synchrotron Research Center Voice Phone: (708) 252-0223
APS/ANL Sector 5, Bldg. 400 FAX Phone: (708) 252-0226
9700 South Cass Avenue WWW: http://tomato.dnd.aps.anl.gov/~jpq/
Argonne, Illinois 60439