Inexpensive Speed Measurement

My question first.

Does anyone have any recommendations for a low-cost incremental shaft encoder and frequency to analog voltage (F/E) converter? If this isn't the best way to go, what else could I use to accurately and repeatably measure speed?

I need to accruately measure the speed of various moving surfaces (derived mostly from motor shafts) in order to characterize and ultimately improve synchronization of a poorly-designed piece of equipment. I plan to use several encoders, each connected to a F/E converter and capture the data with an multi-channel analyzer. I've used the Red Lion IFMA series converter, but I'm looking to save $$. I need at least four sets to get started. Real-time analog information is the only way to go here.

Once I characterize existing performance, I am quite confident I can find some low hanging fruit and quickly eliminate a few problems for my customer. At that point he can decide whether my services warrant further investment in optimizing performance.

Background:

I have been asked to examine a food manufacturing process (let's frozen waffles) that never reached the production levels promised by the vendor. The machine is a one-off deal and the on-site engineer has reached his limits for what he can do to improve things (the vendor is out of the picture now).

There are five transfer stages between the freezer conveyor and the cartoning machine. Waffle positioning, counting, stacking, traansfer and packaging all take place within these five stages. There several belt conveyors, an axis inverter, two bin conveyors and the cartoner that are crudely synchronized together. There are many, many variables at play. Additional operators are deployed to constantly remove jams and compensate for incorrect counts.

A SLC5/05 is running the show (except the cartoner)and there are numerous small VFDs. I do not yet have the drawings or the code.

> I've used the Red Lion IFMA series
> converter, but I'm looking to save $$.

How much are you looking to save?

Have you thought about using mag pickup with high speed counters?

So you want to accurately sychronize shaft & line speeds through a SLC / & a few VFDs cheaply?

I don't think this is going to be cheap or easy.

Cheap encoders


Cheap. Fast. Good.
You can have any two, but not all three.

Lost me with most of that.

There are numerous ways to do it that may be implemented into a digital input. Mount a gear onto a shaft and use a sensor to count the pulses, mag pickup works the same way.

IF the frequency involved are low then an encoder could be coupled to a shaft and use a digital input.

If you need precision accuracy then I would look at getting the HSCE or HSCE2 which you can diretly connect the encoder too.

Since you mentioned Red Lion they offer a multitude of options in this area. I have used pulse generators and their displays for more years then I can remember.

CroCop said:

So you want to accurately sychronize shaft & line speeds through a SLC / & a few VFDs cheaply?

I don't think this is going to be cheap or easy.

Click to expand...

Not sure about the cost, that depends on too many variables not provided, but not sure why it would be hard.

rsdoran said:

Not sure about the cost, that depends on too many variables not provided, but not sure why it would be hard.

Click to expand...

He wants to sync shaft position through a PLC & VFD accurately.
Syncing shaft positions isn't something a SLC series PLC & a few analog outputs & VFDs are designed to do.

Quick update, have to dash out.

It seems I provided too much information. I am not (yet) planning to do anything at all with the SLC & VFDs. All I want to do is get accurate analog signals into the analyzer so I can characterize relationships between physically moving parts and existing sensors and SLC I/O. I'm aware of the difficulty (impossibility) of using a SLC to exercise precise motion control.

Just looking for the least expensive instrumentation to deploy, that's all.

Gotta go for now, will check back tonight.

Thanks for all the input everyone.

All I got was needing options to get encoder or some form of information that will allow him to obtain the RPM of the conveyors. I attempted to show there are numerous options that can provide fairly accurate speed referencing dependent on the resolution (accuracy) needed i.e. that all that signal conversion may not be necessary and analog may not be the best option.

HOW you use that information depends on what is needed.

He mentioned conveyors and a cartoner so think the use of synchronizing "shafts" may have been slightly misleading, in other words just looking for a simple method to try and get the machines to work more efficiently and possibly faster.

Last thought, not sure this application would be considered motion control, more of "speed matching" for lack of a better term. I have worked with motors and generators for 30 years that had to run matched either at the same speed or lead/lag.

If it was done 30 or more years ago, then why would it be hard now?

Ah, well I read too much into it. My bad.
Have you thought about a National Insturements USB card?
For high speed data caputre, you can't beat NI IME. Good USB stuff, VERY reasonable prices, & good support.

So, are these encoders to be a temporary monitoring tool?

What sort of physical limitations do you have?

And, what type of VFDs do you have? You may be able to get a pretty responsive output from them...

Just a thought...
Paul

CeCo3 said:

Quick update, have to dash out.

It seems I provided too much information. I am not (yet) planning to do anything at all with the SLC & VFDs. All I want to do is get accurate analog signals into the analyzer so I can characterize relationships between physically moving parts and existing sensors and SLC I/O. I'm aware of the difficulty (impossibility) of using a SLC to exercise precise motion control.

Just looking for the least expensive instrumentation to deploy, that's all.

Click to expand...

Sounds like the same problem we have. Multi stage process where each stage feeds the next. None of them talk to each other so it is tweak and tweak. Basically we run each stage a little faster than previous.

Cheap instrumentation
handheld tach
stopwatch
handheld counter.

Stopwatch and counter are to count parts (waffles) per minute
OR am I waffling here??

Dan Bentler

Hi there

Well I have Used Pepperl+Fuchs (Model:KFU8-FSSP-1.D), a device for monitoring of periodic signals, both frequency and speed into voltage or current output.It can can count upto 1200 pulses per rotation.
Further information can be found at www.pepperl-fuchs.com

Sorry everyone, I've wanted to post this topic for a week, so out of frustration (never enough time) I just popped it in. I think I provided too much information, but hey, why stop now? Here is some more:

1. The budget is tight.

2. Some costs will be absorbed by the client and some by me (I need to increase the number of sensors and gadgets in my “toy box”). I am now looking at ~$350 per analog channel (encoder, F/E converter, adapter hardware).

3. The intention of my post was to see if there was a less expensive method for getting real-time analog information into the analyzer. Keep in mind, I will also be looking at VFD I/O, SLC I/O, pressure transducer outputs, prox’s, beams and more.

How much are you looking to save?

Click to expand...

Mike, I absolutely did not want to imply that Red Lion Controls are unreasonably priced or undesirable. The IMFA0035 I used at work was very easy to use and reliable. At ~$200 a pop, they’re a great deal, truly. But I’m a little guy and that adds up. There may be less expensive alternatives, hence my post.

So, are these encoders to be a temporary monitoring tool?

Click to expand...

Paul, very good to chat with you again. Yes, initially I plan to attach the encoders in some temporary fashion. I have done this multiple times in the past using a bracket, spring and clamps. It was especially effective for evaluating indexing and continuous motion webs. If constant feedback somewhere proves beneficial and can be integrated, I'll make it permanent.

What sort of physical limitations do you have?

Click to expand...

I expect there will be challenges getting access to certain areas of the machine. The site is some distance away and I've only had one opportunity to kick the tires on this thing. I did not get pictures or video.

And, what type of VFDs do you have? You may be able to get a pretty responsive output from them...

Click to expand...

Good idea. AC-Tech (Lenze) single phase input. No part numbers yet, but as indicated above, I do plan to analyze VFD analog I/O.

Have you thought about a National Instruments USB card?

Click to expand...

CroCop, I already have a multi-channel USB analyzer of my own and a laptop. NI has some very cool acquisition products, both hardware & software, and you're right, their prices and support are reasonable. Thanks for the suggestion.

...in other words just looking for a simple method to try and get the machines to work more efficiently and possibly faster. Last thought, not sure this application would be considered motion control, more of "speed matching" for lack of a better term.

Click to expand...

Nails on the head! The machine runs, but suffers from numerous mispicks, miscounts, jams, dropped pieces, etc. Timing is critical, especially at the transfer points. And yes, the customer wants it to run faster. Thanks for the lengthy posts, I appreciate your time, RS.

If it was done 30 or more years ago, then why would it be hard now?

Click to expand...

Servo drives and motion controllers certainly offer unprecedented ease and precision, but we did get by without them in the past. Who knows, I may come full circle and convince the customer that is the way to go. I plan to lead them to a “data driven decision”.

Cheap. Fast. Good.
You can have any two, but not all three.

Click to expand...

Rats! I want all three! Alaric, thank you for the suggestion! I checked your link and they are definitely inexpensive, most are less than $100. That helps the budget. Any feedback on the reliability and accuracy? (I am very much aware of you getting what you pay for.)

Sounds like the same problem we have. Multi stage process where each stage feeds the next. None of them talk to each other so it is tweak and tweak. Basically we run each stage a little faster than previous.

Cheap instrumentation:
handheld tach, stopwatch, handheld counter.

Stopwatch and counter are to count parts (waffles) per minute
OR am I waffling here??

Click to expand...

Dan, that is essentially what they are doing here and constant tweaking makes it very difficult at times. I thought if I could characterize speed relationships such as the belt conveyors, I may be able to devise a simple feedback system that eliminates tweaking of individual stages without a lot of $$. Too early to tell, I don't have the code or drawings yet.

manmeetvirdi, Thanks for the suggestion. I will check out the Pepperl+Fuchs converters.