Relays

Hi guys, just like to find out how can i get the interfacing of two machines to react faster. both are using 24vdc. currently, when machine 1 needs machine 2 to do something, plc from machine 1 will activate the relay, hence giving a 24vdc signal to the plc of machine 2. Seems fast enough but there's still a delay of approximately 0.3 second. Seems like there is some lag from the time the relay is closed to the time when the difference in vlotage can be sensed at the plc of machine 2. so is there any componenets to replace the relay in order to enhance the response time? Or is there any ways to solve this problem of lagging? loose connections?
regards

If the powers (-) are grounded....and PNP system...
Otherwise optrocouplers could be faster than relays.

The mechanical delay of the relay closing an issue as well as the voltage rise time and both will contribute to the delay. Typically a 24vdc relay will take about 15ms to close.

300ms is a huge delay, even for a relay. I guess if the PLC has a relay output driving this relay, the mechanical delays (plus I/O response time) MIGHT approach 300ms, but it still sounds excessive. You might want to look if the programmer added his own debounce timer to the input in machine 2 (or cranked up the filter time of that input). Keep in mind that he may have done that for a reason!...

Like Seppo said, first replace the relay with a solid-state device, then if the delay still exists, start looking for other causes.

beerchug

-Eric

relay

hi guys, thanks for the response. what is an optrocouplers and how do i use it? what is the solid state device that eric is talking about? I've presently have an Omron G6B-1114P-FD-US on hand. Can it be used? It is an NPN system. I've checked machine 2 board and there's action immediately once the two terminals are shorted. That means the delay is due to somewhere between the relay from machine 1 to the terminals of machine 2. the relay from machine 1 is contacted for approx 0.3 second before the terminals at machine 2 drop to 0v. Why is that so?

How far mascines are (PLCs) and what type of Input-unit/manufacturer ?

they are next to each other but the wires is all coiled up and the length is approximately 4 meters. the relay is from omron. I think the problem is not with the PLCs because the contact of the relay closes on machine 1 and the signal is only sensed at the terminals of machine 2 at 0.3 second later.

Relays not fast enough

Ken,

Have you considered that yoou could be getting some induced voltage on the relay connection wires? If there is a bundle of wires and some of them are carrying A.C. power (typical), then you could get enough induction to slow down the relay response. The solution would be to reroute the D.C. relay wires away from the A.C., or replace the wires with a 2/conductor shielded cable.

Just curious about the application. Why does this need to happen so fast. Just curious, because sometimes we can come up with another way to solve the same problem without speeding up the I/O.

Why don't you network the two PLC'c together. Pretty much all PLC manufacturers have protocols to network multiple plc's together. For example, Siemens uses ProfibusDP, ProfibusPA, MPI,Ethernet,etc. That will be your fastest way to get the PLC's to interact with each other.

Solid state relays is what some previous people have been suggesting. One of the major manufacturers is OPTO22. Here is a link to a 24Vdc input to 24Vdc output solid state relay. http://www.opto22.com/Ordering/ProductDrillDown.asp?IK=289

You still really havent stated what exactly you are doing. What is the relay inbetween the relays for? Isolation?

The process is actually a pick and place application. Machine 1, a plc controlling a servo motor, will enter machine 2, a plastic molding machine, when the mold is completely open. Upon retreival of the parts, machine 1 will move out of machine 2 and give machine 2 a signal to close the mold. I cannot hook up the two plcs together because there's no plc on machine 2. There's only a RCU board (what is this actually?) that allows machine 1 to communicate with machine 2 by means of shorting connection or opening connection. Present method employed by machine 1 is using relay to short or close the terminals in an attempt to signal to machine 2. However, I find that there's some lag in time starting from the closing of relay 1 to the activation of action by machine 2. Had used a meter to check if there's any delay set internally at machine 2 but found that it's actually the delay in signal that caused it. Will check for the AC wire and DC wires. Any recommendations guys? Keep them coming.
Thanks

From the information that you have given so far I personally do not understand exactly what you are trying to do. A sketch would be helpfull. My assumption at the moment is that a relay is enegised from machine 2 and a 24 volt supply is then passed through the relay contacts to initiate some action on machine 2. If the relay and its contacts are closing more or less instantaneously then the delay must be something to do with the 24 volt supply to machine 2 or else to do with the machine 2 action commencing.

Whoops:
My assumption at the moment is that a relay is enegised from machine 2 .

Should have been:
My assumption at the moment is that a relay is energised from machine 1

Ken,
It sounds like you are taking about a standard SPI robot interface which is very common in injection molding. This provides a means of interlocking the molding machine to automation, as in your case a pick and place robot.

Have you checked the robot program to insure no delay has been added for clamp close signal? Also insure that the molding machine does not have a close delay entered somewhere.

The reason for the relays in the interface is for isolation purposes. Keeping voltage of molding machine seperate from robot.

I've done lot of these interfaces and have never come across a 300ms delay in the actual signals. It has mostly been programmed delays in either the automation or machine causing a lag in close operation.

Walt