Multiaxis control using pulse train

I am working on a project which eventually should become an elevator picking up objects from a shelf. I need to control movement X and Y with possibility to expand to Z.

Configuration:
PLC-ABB AC500 with CPU PM571
Servo- Omron R88D-UP03V (5V pulse train, 5V direction)

I need help in any way possible. The Servo system is controlled by a pulls train for positioning and one signal for direction. I am gonna get a fast encoder module CD522, a new product launched by ABB last year. 2 x 300 khz independent counters. It's a little expensive compare to an high counter module DC541, but it's faster, has got two timers and I can get 5V output.

I have many questions.
Servo impulses should be 5V, but I have only 24V output if I chose the high counter, cheaper unit. What is common thing to do to scale down the voltage? I would just use input resistance of the servo and use a resistor, but my supervisor does not like the solution. Is there any fancy way to scale down the digital signals?

I also have problems to se the way to synchronize the impulses for the different engines, but also with other control impulses, if CPU is much slower than the counter. Is there any good source on algorithm for the PLC motion control? Shall I look for an interrupt solutions?

Thx in advance.

Plz, give me anything. At least the part about scaling 24V output from PLC to 5V input to the servo. Why would the extra resistor be a bad idea? What is a common practice?

I might do the motor control in sequences anyway. They don't have to run simultaneously.

It is not uncommon use resistors in your application, it all depends on the output configuration of the ABB AC500 module.

I tried accessing AC500 documentation on the ABB website, but no success.

BITS N BYTES said:

It is not uncommon use resistors in your application, it all depends on the output configuration of the ABB AC500 module.

I tried accessing AC500 documentation on the ABB website, but no success.

Click to expand...

Thx for the response B&B.

ABB? Well, they are a little strange when it comes to manuals and knowledge. It looks like their marketing is a little 60's. "Buy and then we tell u."

But u saying that it should be ok? I mean the problem should be only in Servo part, too much effect, but if I know the input resistance, it should work with splitting the voltage.

Servo input says: TTL, photoisolated, 6 mA at 3V

and ABB module says 100 mA at 24V for high-speed output and 500 mA at 24V for ordinary digital output.

The current rates should not be a problem, right?

Here is an excerpt from the OMROM CJ1M series PLC showing typical connections for PULSE and DIRECTION outputs. This should help you decide what size resistors to use in the pulse and direction signals between ABB and OMRON.

BITS N BYTES said:

Here is an excerpt from the OMROM CJ1M series PLC showing typical connections for PULSE and DIRECTION outputs. This should help you decide what size resistors to use in the pulse and direction signals between ABB and OMRON.

Click to expand...

Thx a lot. So it looks like that is a common thing to do. It helps a lot that u found the file, because now I have evidence that my supervisor wants to bully me. He said that it is not the smartest way, but it looks like it is "the way".

Thx a lot again.

I have a question. Do you really need 300 KHz for your encoder. There must be a formula to work out the maximum frequency required based on speed and resolution. I only ask because I once worked on a pick and place crane that just used a simplle proximity switch and regular PLC input.
BTW if you need a simple bi-directional encoder, look no further than your computer mouse.
Roy

vedano said:

Thx a lot. So it looks like that is a common thing to do. It helps a lot that u found the file, because now I have evidence that my supervisor wants to bully me. He said that it is not the smartest way, but it looks like it is "the way".

Thx a lot again.

Click to expand...

Looks like you need a new supervisor OR he needs to study Electrical Engineering 101 [very basic course].

SUPERVISOR -> <- YOU

What is the encoder for in this application?
How many encoders? One per axis?

Roy Matson said:

I have a question. Do you really need 300 KHz for your encoder. There must be a formula to work out the maximum frequency required based on speed and resolution. I only ask because I once worked on a pick and place crane that just used a simplle proximity switch and regular PLC input.
BTW if you need a simple bi-directional encoder, look no further than your computer mouse.
Roy

Click to expand...

I actually don't need an encoder at all, I just need high-freq output that I can control.
ABB claims 0,3 ms per 1000 statements and 6 ms for floating point. That would be 3,3 kHz divided by 2. Motor can go up to 2048 pulses/rot. Let's say I go down on half, which means twice as much origin search, I would still get only 1,6 rotation per second. Am I right?

I agree that u can make positioning with switches and sensors, but the whole point of the project is to build a system, that could be used for educational porpoises and it should be based on PLC-servo communication. And because it is a school, they want to use the things they already have. Servo that takes pulse train and direction signal.

I don't know what bi-directional encoder does, I have worked only with a rotary encoder. I tried to look it up, but nothing with any sense came up. If u think I could use it with ABB plc, it would be cheaper than PWM/high counter module, could u post a link on it? Just remember that I need fast output, not sampling.
Thx.

Sergei Troizky said:

What is the encoder for in this application?
How many encoders? One per axis?

Click to expand...

There is no encoder, position loop is closed Servo-Motor, PLC just sends reference by a pulse train. But I might actually need one motor running with encoder, because the module I want to use can produce only two high speed signals and I need eventually 3 axis. Maybe not me, as I want to do only X,Y movement, but it should be implemented also with Z. But the Z movement would be actually only in and out of the shelf, so I can do with a switch.

It's actually smart with the pulse train, because the last thing I feel like doing is a closed loop design on PLC. I never tried that before either.