Recurring random issue when Jazz is switched on

[EJfr]

Hello, I have designed a machine for my French company that is fitted with a Jazz JZ10-11-T40.

The machine is working well, but I have a recurring random issue.

 

When the Jazz PLC is switched on : sometimes, all is running well (the 5 inputs values have correct logic value), and sometimes the PLC works and the 5 inputs values were inverted... and this occurs randomly.

 

I mean :

Sometimes, after having switched the PLC on, the 5 inputs are correct (input I0 to I4) :

-2 x laser sensors are on logic 1 (no detection because no objects to be detected)

-Two switches are on logic 0 (the switches are in off position)

-The last input from a 5th sensors is on logic 1.

Sometimes, after having switched the PLC on, all the inputs are inverted and results as following :

-2 x laser sensors are on logic 0 (detection should be 1 because nothing is present and detected)

-Two switches are on logic 1 (even if the switch are in 'off' position)

-The last input from a 5th sensor is also inverted in logic 0.

 

In any cases, the eight outputs are always correct.

 

I hope my description is clear enough.

Did one of you got same issue?

Do you have any idea about what can occur or what can be the root cause?

 

Think you in advance for your help.

[kratmel]

Please check connection (red on pic).  Is it  pnp/npn type selection  input connected correctly?

[Ausman]

some comments on this.

Kratmel may be correct, but if the system is running ok most of the time then I guess it is physically set up ok.

1).  This depends on how you have structured your program, but it might be a power down issue in that various sensors on your system either completely power off before the plc does, or vice versa.  The result will be something unexpected on your normal running situation due to what has been remembered by the plc on boot.  If those results are carried over somehow the upset occurs.

2).  Bearing in mind point 1, do you have an inbuilt time delay on all operations starting?  Or does it all just bang in straight away?  I believe from your description that you likely don't have one and you really need to!   I would be putting in a start delay that doesn't let anything happen on power up until, say, 10 seconds has gone by and the system has stabilised and accurately recorded all I/Os.  There are various way to do this, but all relate to the use of one "master timer" that is triggered on power up.

3).  Lastly, do you know about power up values?  On certain operands you can set the values that will be put in place on power up.  Look up "Power-up" in the Help files.  You can try these out to see if it helps, but in your case you may need to add a little further complexity by having all your I/Os going through MBs , which are initially referred to until my suggested start-up delay has finished and the system switches over to direct connection.  This shouldn't be needed if the start delay in 2). is used, but sometimes odd  things occur that need odd  solutions!

4).  Lastly, is the battery OK?  Check this by looking at SB8.  Also have a look at Power-up in Help again, and click the Battery Backup link at the bottom.   This will run you to another necessary piece of knowledge as to how things are retained (Battery Back-up values), and you may not have set this properly either.

cheers, Aus

[kratmel]

In Jazz PLC  not isolated input used.

Then PNP and NPN changeover input n/p used  for selected INPUT PINS  and it connect some number of pull up (down) resistors to GND or +24VDC.

When PNP or NPN mode selected - some cirquit connected to n/p PIN on board send to PLC MCU "0" or "1" for invert or not  MCU input mode.

If user do not connect n/p input to GND or  +24VDC -    PNP mode selected as "default" but it is not real PNP mode.

If user send PNP signal to one or more  inputs - pull up (down) resistors translate +24VDC to n/p pin and MCU can invert  input mode.

This MCU invert mode change is happend if  "1"  PNP signals number is much more then "0" PNP signals number.

Then.... PLC MCU input mode can be changed randomly via input signals.

7 hours ago, EJfr said:

The machine is working well, but I have a recurring random issue.

Please check n/p pins  connection and maybe it solve your problem. 

 

[Ausman]

Thanks Kratmel, good to know.

cheers, Aus

[EJfr]

Thanks for your help ! I will check this this week on the machine and give you a feedback.

 

[EJfr]

Finally, I can work this week on the Jazz issue.

I've checked connections, and your remarks get me on the right track.

Connection of the 24 VDC was coming directly from power supply, and

Pin n/p I16-I17 and pin n/p I0/I15 were fed by the same power supply in +24 V but not directly because going through a contactor that can be switched off with the emergency stop circuit (I hope my explanation is good enough).

I simplify the connection and did a direct feeding of the three pins of the inputs by +24V.

I seems it works well now : after modification, no inversion occurs, I cross my fingers.

(One remark about battery check : before this I try to check the battery with SB8, but this function seems to be only available for M91 device.)

Thank you :  your advices helped me 🙂

 

[Ausman]

That's great, EJfr. 

Are you saying that ONLY the PNP/NPN control inputs go through the contactor, whilst the PLC itself does not?   If so, I feel sure that Kratmel's great advice and the use of the contactor is the main culprit.  It is normally imperative that all power supply to anything directly connected to the PLC, like expansions for instance, is on the same direct connection so that the PLC reacts to everything and "what-to-subsequently-do" correctly on power-up.

As an aside, I would also look at your emergency stop system.  Sometimes it is useful to have manually controlled function available from an emergency stop, depending on how a machine works and potential for human entrapment and need for the ability to easily release said entrapment.  A changed power up sequence that only comes on when the Emergency stop is activated can sometimes be useful.

@Joe Tauser    I thought SB8 was universal across all devices that use U90.  Is this incorrect?

cheers, Aus

[Joe Tauser]

Here's what the official descriptor says:

I honestly don't know on the JZ10.  I don't use the Jazz series very often.

Joe T.

[Flex727]

15 hours ago, Ausman said:

I thought SB8 was universal across all devices that use U90.  Is this incorrect?

 

10 hours ago, Joe Tauser said:

I honestly don't know on the JZ10.  I don't use the Jazz series very often.

I happen to have a JZ10 sitting on the shelf. I hooked it up and went online. SB 8 was zero, so that really doesn't tell me anything. So I opened it up with the intention to remove the battery to see if SB 8 goes high, but found that the battery is soldered in.

[EJfr]

For safety, I kept my initial wiring for the output side : the feeding of the outputs is going through a contactor linked to emergency stop.
Consequently, if emergency stop is activated, all the outputs are no  more feeded and the actuators, electric motor, etc stop. With this, safety of the machine is correct.

For the battery, as my initial issue seems to be solved, I will let the Jazz as it is and hope the battery will be OK for a long time 😉