PIC to monitor multiple ammeters

[cosmiccloth99]

I was wondering if it is possible to use a PIC to monitor ammeters that are 480v and read up to 300 amps? I would need to use the values from the ammeters as variables and if any of them were spiking past a certain amperage I would need the pic to adjust the value of one of them. That probably makes no sense so I will try to explain what I am attempting to do. The first ammeter only reads to 40 amps it monitors a motor that is turning a conveyor and the value is relative to the amount of weight on the conveyor. I would like to be able to set a target value and have the pic maintain it. There will be 5 ammeters that need to be monitored that are separate from the first one. If any of them go over a set value the first would need to be lowered to accommodate. The problem is I don't have any idea how to monitor an analog ammeter or if it's even possible. If not what could I convert the ammeters to that would allow for accurate readings?

[enzine]

I think you have to monitor alternating currents.

I think the best, simplest and safest way to do this is to insert CTs (amperometric transformers) and  close the secondary winding with a low value resistor to convert current to voltage across the resistor.

It may be that CTs these are already present inside the ammeters .

In this way you have an AC voltage signal proportional to the value of the current to be measured and ISOLATED from mains.

This signal therefore need to be processed (ac / dc conversion or RMS ).
There are several ICs that do this or you may acquire the  signal directely in the A/D input pin of your PIC and process by SW .
Remember, however, that CTs are expensive .

Also remember to NEVER leave the secondary of a CT without a load resistance of a few ohms

I hope this information can be useful to direct you to the solution to your problem.

Ciao
Enzo

[Gallymimu]

it sounds like you expect to have closed loop control when you say a reading from one ammeter will force you to change settings on another.

If you are READING from an ammeter, what exactly are you ADJUSTING?  I wouldn't think you are not actually adjusting the ammeter but something else which affects the current in the motors.

[cosmiccloth99]

This is an idea to automate the feeding of a rock crushing plant. When feeding the plant the material drops onto a conveyor that is what the first ammeter is reading. depending on the amount of material dropped onto the belt the amperage increases or decreases. the other ammeters are monitoring the crushers in the plant. If one of them is getting too much material then the amount of amperage it draws will spike up that is what will need to be monitored and the feed would need to be slowed down until the crusher is back in its operating range. the feed is controlled by a gate that is opened or closed to control the flow of material.

[Gallymimu]

you can do what you wan't, but a microprocessor isn't really the right tool for the job.

This should really be done with PLCs and industrial ammeters suitable for high reliability and maintainability.

If this is just an exercise for fun we can all certainly help. 

You should decide, for the ammeters, if you want an industrial ammeter that will produce dc output, a current transformer and IC that will produce an isolated AC signal and then RMS to DC conversion, current transformer being directly read at modest sample rate by the PIC ADC, or designing your own AC monitoring circuit.

Then, once you get the data into the PIC, you'll need to determine a suitable control loop (PI controller at low speed will likely work fine).  That will depend on your response speed and system gain (I presume there are many seconds or more between when you detect high current and when you need to slow the rock flow).

Finally you need an output mechanism i.e. a DAC that feeds the controller of the gate motors or something similar i.e. if it is a simple open and close that you need to control with on/off for a period of time (bang bang control)

[solutions]

Retain someone who knows what they are doing in heavy industrial closed loop control systems and make them responsible for it. There's likely a lot more to it than "ammeters".

Screw this up and the damage to the crusher could hire 1,000 consultants. Do you really want to be "that guy?"

[Gallymimu]

Retain someone who knows what they are doing in heavy industrial closed loop control systems and make them responsible for it. There's likely a lot more to it than "ammeters".

Screw this up and the damage to the crusher could hire 1,000 consultants. Do you really want to be "that guy?"

How is he going to learn if he doesn't blow up a few million dollars worth of equipment?  

We've ALL done it!  Well, I've never made a mistake but I am sure other might have.

In any case we should try to teach him, explain the risks, but not just tell him to hire someone and give up.

[zielpunkt]

Without entering into the question about the security of what you intend to do, I believe you should note the Texas AMC1100 chip and its application notes. There are also variations of this device and it may be that you find in them what you are looking for in converting high values of voltage and currents with relatively low costs and good accuracy. I hope I helped you. Good luck.

[cosmiccloth99]

I would like to thank you all for taking the time to advise me on this topic. I am just getting into pics and their potential applications. I spend a great deal of my time at work(the rock crushing plant) so I started looking around the plant at our current systems to see if there were any that could be improved upon by implementing a microcontroller or two.I was using this as a learning exercise to try and better understand what types of systems one can interface with and control. I can see this idea goes way beyond my current ability but I am thankful for the knowledge that you all have shared with me.

[CocaCola]

I am just getting into pics and their potential applications.

so I started looking around the plant at our current systems to see if there were any that could be improved upon by implementing a microcontroller or two.

IMO, leave large and dangerous industrial machine control of this caliber that can easily kill, maim or cost huge amounts to repair if something goes wrong to those much, much, much more experience...  And even then I (as others) would argue that something like a PIC just isn't the right tool for the job...

Stick with small scale, if you want to pursue a proof of concept on this, since this concept intrigues you, build a small scale low voltage mock-up of the conveyor belt system and simulate your design on this mock scaled down build, just as a proof of concept as you learn...  But, don't get in over your head believing that if you can get the small-scale mock-up to work that you are ready to build one for the real beast, as that is a huge leap...

[cosmiccloth99]

IMO, leave large and dangerous industrial machine control of this caliber that can easily kill, maim or cost huge amounts to repair if something goes wrong to those much, much, much more experience...  And even then I (as others) would argue that something like a PIC just isn't the right tool for the job...

Stick with small scale, if you want to pursue a proof of concept on this, since this concept intrigues you, build a small scale low voltage mock-up of the conveyor belt system and simulate your design on this mock scaled down build, just as a proof of concept as you learn...  But, don't get in over your head believing that if you can get the small-scale mock-up to work that you are ready to build one for the real beast, as that is a huge leap...

I was using this as a learning exercise to try and better understand what types of systems one can interface with and control. I can see this idea goes way beyond my current ability but I am thankful for the knowledge that you all have shared with me.

Thank you for restating what I had already said. To be clear I understand this project is way above and beyond what I am capable of, I actually realized that before I had ever even posted my original question. However, I was curious about the possibility of it and also the idea of using an ammeter or any meter for that matter as an input device.

[CocaCola]

However, I was curious about the possibility of it and also the idea of using an ammeter or any meter for that matter as an input device.

Did you fail to read the part where I encouraged you to explore your ideas of using an ammeter on a mock build of the concept?  You can certainly use an ammeter or another meter as a feedback input...

[Gallymimu]

if part of the task it to approximate how many rocks are on the conveyor, it might be interesting to set up a thru beam type sensor on the conveyors to approximate rock load.

[witsanukai]

some study i think you may try to use CT with your system clamp to motor cable and try to mornitor the curent and obeserve what happen in your system first, without touching the previous control system, then you will have more idea what you want to do, remember safety first rule too.
 

[thunderer]

I know this is a kind of old thread, but FYI, such crusher, fed with rocks by a conveyor (this is how they are used in oil sands in West of Canada), has motors rated anywhere between 500hp to 5000hp (rated in MV). See this link for an idea of the size: http://www.fam.de/english/Products/Mineral%2520processing/Double%2520roller%2520crushers/index.html.

Such a crusher control system I made for FAM included a real DCS system (not only a PLC). And the motors were all protected by Multilin 369 relays from GE. Such protective relay handles lots of protective functions and communicates lots of info.

FYI, it was not the current that provided info on the loading of the crusher. Because that crusher was really crushing everything. There were lots of sensors, and so on...

If your employer asks you to sketch-up a control system, do not propose the PIC solution. First, you should get the P&ID diagram. You will see all motors, pumps, limit switches, and all sorts of sensors. After you see the complete image of the system... then decide if you can do anything.