High speed dynamic frequency output

[jholm90]

Looking for suggestions on a pathway for a high speed, variable frequency output for a slitter line. There is a corporate data collection system that is some knockoff brand and is expecting some high speed pulse input to measuring production speed.

We cannot use standard logic with timers as this system is expecting a 10-1500hz *variable* signal and scan time can not flick the standard outputs off/on fast enough.

We tried to implement a PWM output from the 1734-VHSC ouptut module, however it has a fixed frequency but variable duty (length of time output is high). I have found lots of plc input cards for variable frequency inputs, but nothing to compare for a output. This VHSC module can have clock speed changed in the module properties, however it requires disable/re-enable module and will not be dynamic with process flow while that change is being made.

This system needs to be dynamic so when the line is creeping at startup for 10ft/min we get a pulse frequency of 10hz and when it is at max speed of 900ft/min we get a frequency of 900hz. These are generic numbers and we can scale down on the data collection side, however it cannot be scaled up (we can make 10 pulses=1ft/min, but we can't make 1 pulse=10ft/min)

The end of the day our result might be hooking an encoder to the input for direct line speed, however I would prefer not to install another mechanical device that can get bypassed or broken when we have a very accurate speed already in the plc itself. It's too bad this knockoff data collection system will not allow us to message a value directly...

 

[RET]

Give this a look

https://www.phoenixcontact.com/en-us/products/frequency-transducer-mini-mcr-sl-ui-f-2864082

Phoenix makes a lot of other different signal transducers.

 

[BryanG]

There are two related by different pulse output methods. You found PWM (Pulse Width Modification) where the Frequency is fixed but the ratio of the On and Off time is altered, this is often used for Heating control. Then there is PTO (Pulse Train Output) where the On and Off times are a fixed ratio (1:1), but the frequency can be altered. This is often used in motion control, I use PTO on a Siemens PLC to simulate Encoders and Flowmeters when I am testing panels. Hopefully your Rockwell thingy will have something similar to PTO but I don't know that they will call it PTO, Wickipedia doesn't seem to have heard of PTO.

 

[drbitboy]

Building on what @BrianG wrote, the MicroLogix and Micro800 lines have PTO, a description is here.

 

[Peter Nachtwey]

This sounds like an encoder that needs some scaling to me.
I fail to see when 10 pulses for 1 ft/min won't work. Now you have more resolution.
Back in the dark ages we would use hardware to generate an interrupt pulse every so many encode pulses. This effectively divided down the encoder.

 

[widelto]

@Jholm90:
When I was working I designed several slitting lines (I am retired now).
I have two question:
Are you measuring linear speed at the slitter knifes or at the recoiler ?
Do you have any kind of looper located between slitting knifes and the recoiler. ?
Normally the knifes motor has a encoder in order to control its speed, you can use this encoder to measure motor speed and thru a ratio (gear box) you can calculate slitting knifes speed and by knowing diameter linear speed.
Normally what I used to do was to enter knifes diameter (HMI) and this way I obtained linear speed.
Try to get more info from your knockoff electronic device, then use your VFD or VSD in order to generate pulses from it and then these pulses can be measured by your unknown device and this way you can get linear speed.
If you have a looper then you can use a pinch roll before the recoiler and with an encoder you can measure linear speed.
Let me clarify that all the slitting lines I designed were used to cut steel to form pipes.

 

[joseph_e2]

Could you use something like this?
https://www.redlion.net/product/analog-frequency-converter-module
Use a PLC analog output tied to the line speed that it already knows and drive the input of the converter module, whose output will feed the SCADA system.

 

[jholm90]

There are two related by different pulse output methods. You found PWM (Pulse Width Modification) where the Frequency is fixed but the ratio of the On and Off time is altered, this is often used for Heating control. Then there is PTO (Pulse Train Output) where the On and Off times are a fixed ratio (1:1), but the frequency can be altered. This is often used in motion control, I use PTO on a Siemens PLC to simulate Encoders and Flowmeters when I am testing panels. Hopefully your Rockwell thingy will have something similar to PTO but I don't know that they will call it PTO, Wickipedia doesn't seem to have heard of PTO.

Thank you very much for the PTO / pulse train nomenclature- it significantly helped my google searches. Rockwell does not have any product for the compactLogix/controlLogix line that has this output solution, unless using an AMCI stepper driver card is used as an axis

 

[jholm90]

Give this a look

https://www.phoenixcontact.com/en-us/products/frequency-transducer-mini-mcr-sl-ui-f-2864082

Phoenix makes a lot of other different signal transducers.

Thank you for the suggestion, these items are on order now as they suit the spare analog output we already have and make the 1:1 translation without having to add in stepper drivers

 

[jholm90]

@Jholm90:
When I was working I designed several slitting lines (I am retired now).
I have two question:
Are you measuring linear speed at the slitter knifes or at the recoiler ?
Do you have any kind of looper located between slitting knifes and the recoiler. ?
Normally the knifes motor has a encoder in order to control its speed, you can use this encoder to measure motor speed and thru a ratio (gear box) you can calculate slitting knifes speed and by knowing diameter linear speed.
Normally what I used to do was to enter knifes diameter (HMI) and this way I obtained linear speed.
Try to get more info from your knockoff electronic device, then use your VFD or VSD in order to generate pulses from it and then these pulses can be measured by your unknown device and this way you can get linear speed.
If you have a looper then you can use a pinch roll before the recoiler and with an encoder you can measure linear speed.
Let me clarify that all the slitting lines I designed were used to cut steel to form pipes.

This application only has one encoder wired to the plc on the take-up drum and it dynamically changed speed based on how many times the take-up drum has indexed. Every rotation adds the material thickness *2 to the diameter, so takeup rpm is much faster at the beginning of the coil than the end of the coil. We tried duplicating the singal to the data collection device, however the scaling factor would change over the process cycle and will not work.

Unfortunately they do not want a line speed encoder mounted on the material so we will end up sending out the analog speed reference through a frequency transducer (thanks @RET) to convert to the PTO signal the knockoff data system is expecting.

 

[Ken Roach]

one encoder wired to the plc on the take-up drum

How, exactly, is it wired to the PLC ?

How is the takeup drum driven; VFD, servo, etc ?

You mentioned 1734-VHSC so I assume this is an A-B control system. Do you have a CIP Motion capable ControlLogix or CompactLogix controller (with an -M in the part number if CompactLogix) ?

If you can calculate the output speed as a gearing or scaling factor inside the controller and express that in a virtual axis, then maybe the 2198-ABQE module is appropriate for this application. It can create an quadrature encoder output that is proportional to a real or virtual CIP Motion axis.