Speed Sensing

[rsdoran]

You are welcome. I agree with Alaric, nothing wrong with what you have, especially if it works for you.

I had another idea though, you stated as it is the sensor is on or off half the time, maybe use something smaller to sense. From your description I am envisioning a drill head driving screws. Maybe a small bolt sticking out of the drill head that would only be sensed a tenth or so of the time of travel.

I am not sure if that may require the use of a high speed input but it could decrease the chances of landing near the sensor in a position that would cause the sensor to oscillate. If this is a farily consistent process you may be able to establish where the sensing point will stop in relation to the sensor.

Since it is such a fast process you may be able to count the turns and establish a base stopping point, who knows, lots of ifs out there.

 

[craig_avanzar]

wouldn't a tach work here? positive feed back for motion and some other advantages for the future.

 

[geniusintraining]

This is what I use.. then one input to the PLC, they make a very good product.

http://www.pepperl-fuchs.com/selector/gui/show_product_detail.kly?selected_prod_id=28911&lang=ENG

 

[Big John T]

This is what I use.. then one input to the PLC, they make a very good product.

http://www.pepperl-fuchs.com/selector/gui/show_product_detail.kly?selected_prod_id=28911&lang=ENG

The link does not work. What is the model number of the part you use?

Thanks
John

 

[Big John T]

wouldn't a tach work here? positive feed back for motion and some other advantages for the future.

There is not any way to connect a tach to the air tools.

Thanks for the suggestion.
John

 

[Big John T]

This is a typical "speed sensor" with the prox and a target...

rsdoran that is some interesting things to consider. The smaller the target the less chance of ending up on the edge. I like the idea of counting the number of rotations. Most are 1/4-20 or 5/16-18 screws so the number of turns is low... The RPM of the tools is usually 750 RPM or less... Now my head hurts LOL...

Got to chew on this a bit...

Edit: That is not my welding I do much better than that!

Thanks
John

 

[rsdoran]

There is not any way to connect a tach to the air tools.

Thanks for the suggestion.
John

There are hollow shaft encoders and tachs with optional bore sizes that may fit on the head or the bit, just depends.

Since it is air I wonder if there is a pressure differential between cutting and stopped.

Like I said, lot of ifs.

I missed that pic, the hollow shaft encoder/tach could be mounted in there, I think http://www.beiied.com/incremental_encoders.html

 

[WireGuy1950]

We use a notched feed roll and a prox to determine if rubber strip is being fed into our extruder screw, if the roller stops it will generate an alarm. I am attaching a pdf of my circuit. It too is pretty simple.

 

[Big John T]

rsdorn,

The hollow shaft would work but I barely have 1/2" of lenght to put it so that rules that out as they are 1 3/8" long. Thanks for the idea.

Thanks
John

 

[rsdoran]

It depends on how crucial the issue is. The encoder aspect may work better if doing a count, may offer a tighter range, but may require a HSC.

 

[rsdoran]

Not done yet though, that clamp could be made into a pulley with an encoder/tach mounted where the sensor is.

Lots of ifs, lol.

 

[Big John T]

It depends on how crucial the issue is. The encoder aspect may work better if doing a count, may offer a tighter range, but may require a HSC.

Basicly I just need to know that it is not rotating. The only problem is sometimes the tool will end right on the edge of off and on on the prox and the slight movement of the tool when it is stalled will "look" like it is still running. After 10 seconds the "automatic" will fault out if not complete then someone has to reset the machine. It doesen't happen often but it does happen. Perhaps something with a wide hysteresis zone that once on has to move quite far to be off again...

Thanks
John

 

[Jimmie_Ohio]

Simple is Better - One Timer is Less than Two Timers

When I do speed sensing, I use the motor run contact (PLC input) to drive an on delay timer (TON). Then I simply one-shot the proximity switch to a reset rung of the timer. (Exactly as rsdoran's post #6)

Using separate one shots for "on" and "off" is redundant and accomplishes NOTHING, except for additional programming. If you had a thousand circuits to implement, we're talking some serious rungs if you do this "double-programming".

Bottom Line: If the screw is turning properly, it is making AND breaking. If it didn't "make", it couldn't "break" because of the one-shot, now could it? (and vice versa)

Finally, if this oscillating issue is really critical in your application as a nuisance problem, use TWO proxs aimed at two different points of the same target. There's NO WAY that BOTH will oscillate on-and-off without faulting out the timer circuit proposed above...

 

[PeterW]

regarding rsdoran's post#6, its similar to what I would have, except a N/C of the running would directly reset the timer and then in parallel with that a rising and falling edge detection of the sensor to reset the timer, therefor ethe timer will only time out when the motor is running and no pulse is detected.

The picture shows a very rough method of what i am used to seeing, we would normally have a plate with 4 high points (more if you need it) screwed into the end of the shaft with a sensor detecting the high points.

If you only have half an inch, then the plate could be thick enough for the sensor to pick up from the side and detect there.

 

[Big John T]

Looking at rsdoran's post #6 and it is wonderful by the way. I came up with this rendition that would knock out a lot of code for a machine with 4 heads... Any reason why this would not work? I don't have a machine to try it on at the moment... I should have one powered up in the next day or two to try it out...

Thanks
John