MicroLogix 1000 and Accelerometers

If i were to use this accelerometer http://www.pcb.com/spec_sheet.asp?model=352A73&item_id=5963 , what all would I need to do to give the MicroLogix 1000 a readable signal? I am interested in determing if a machine exceed a certain vibration, and the machine's process I will be reading occurs in less than a second. I have talked to a few people who have said that PLCs typically cannot respond fast enough to accurately read the signal from an accelerometer, and they have suggested building a comparator and one-shot with a 555 IC. Is that the case? Or, is there simply some type of signal conditioner that would help?

Thanks

pjones39,

1. is this a homework assignment?

2. we have done this before, with a plc, but we had to do a lot of research to find the unit that best suited our needs and figuring the process out.

please detail your application better.

regards,
james

James,

It is part of my senior design project in mechanical engineering.

Our application involves the monitoring of a high voltage circuit breaker to determine when it is begining to trip slowly, an occurence that can destroy other costly equipment. The breaker trips extremely fast (breaker must trip current in 30-50 ms once fault is recognized), and in the process creates significant vibrations (400 g's) on some of its components. We want to be able to remotely recognize when these vibrations become smaller than usual, indicating the process is slowing down.

Plans are to possibly use a piezoelectric accelerometer and have a signal chain as such:

accelerometer -> signal condintioner (if needed) -> comparator circuit (for dialing in a threshold) -> one-shot circuit (to lengthen voltage signal) -> Micrologix 1000

My concern is if all of this is neccesary?

Thanks in advance.

The micrologix 1000, running just a couple of lines of code can scan it's inputs in just a couple of milliseconds. A longer program with math and compare (16 bit) instructions may typically run a scan time of around 5-20ms. The pulse stretcher is still probably a good idea. I did not follow your link, so I can't say for sure about the other components you listed.

OkiePC,

Thank you for your response. I am moreso concerned about the 1000's sampling rates at its input and not so much about how long it takes to run the logic, though, that may be what you are saying, I am not sure. Essentially my concern comes from the fact that the acclerometer will have high, but very short max voltage peaks, and will the MicroLogix be able to read it without aliasing - recording half way up a spike as opposed to the max.

Are you planning to connect the accelerometer to the ML1000's analog input ( you do have a analog ML1000?). Then see manual below chapter 5. I'm not sure it will be fast enough to do the job.

http://literature.rockwellautomation.com/idc/groups/literature/documents/um/1761-um003_-en-p.pdf

You are about to enter hacker ( in the 1970s sense ) heaven
http://www.parallax.com/
http://www.parallax.com/Store/Sensors/tabid/86/Default.aspx
http://www.parallax.com/tabid/407/Default.aspx
http://forums.parallax.com/forum.php

A PLC is not fast or deterministic enough.

Mickey,

Thanks. I do have the analog ML1000. I believe it has 2 analog voltage inputs and 2 analog current inputs. It does appear that the sampling rate of the ML1000 will be too slow for this application.

Does anyone have any insight on the signal chain I am thinking of doing:

accelerometer -> signal condintioner -> comparator circuit -> one-shot circuit (to lengthen voltage signal) -> Micrologix 1000

The idea behind this signal chain is I can set the vibration threshold via the comparator. When the threshold is broken, the comparator will send a short pulse signal to the one-shot, which will then be enlongated. The ML1000 will then be able to recognize if that threshold has been broken. Do I have faulty logic going on here?

pjones39 said:

Does anyone have any insight on the signal chain I am thinking of doing:

Click to expand...

Yes

I am interested in determing if a machine exceed a certain vibration

Click to expand...

based on peak acceleration only.

accelerometer -> signal condintioner -> comparator circuit -> one-shot circuit (to lengthen voltage signal) -> Micrologix 1000

Click to expand...

You don't need a PLC for that and it is too slow. You don't need a micro controller either. Some op-amps will do. You need a op-amp cook book.

Do a search for op-amp peak detect.

Peter,

Thanks for your reply. After doing some research on the op-amp peak detect, it seems to be what I am looking for. I really only want to know the maximum acceleration during a breaker trip and the rest of the data is useless.

Next step. Since I only care about the peak acceleration, and the op-amp peak detect will give the ML1000 a readable signal of the voltage, is it necessary to use a signal conditioner in front of the op-amp peak detect? In other words, would this signal chain work for measuring the peak acceleration?

accelerometer -> op-amp peak detect -> ML1000

Do I need anything else for this to work at least decently?

The output of the peak detect op-amp may not have enough power to drive the ML1000's input circuit. You are getting into electrical engineering here. You must look at the ML1000's input specifications and make sure your circuit can sink or source the current it needs to raise or lower the voltage above or below the detection threshold. The circuit that does this is usually called a buffer. There are inverting and non-inverting buffers.

Another site that has Op-Amp basics.
http://en.wikipedia.org/wiki/Operational_amplifier_applications

Peter,

I plan on running the circuit output to the analog inputs of the ML1000. There shouldn't be a threshold there, right?