5 VDC high speed counter input

I have an AB MicroLogix 1100 that I want to use to count pulses from a motor mounted encoder. The encoder is supplied with 12 VDC and the ML1100 takes 12-24 VDC inputs. However, the actual voltage from the encoder is 11.7 VDC when the encoder is not turning.

The PLC will count pulses when the motor is moving VERY slowly but as soon as it picks up speed it sees nothing. This version of the ML1100 can handle inputs up to 40 kHz and I am using the Z channel so speed should not be an issue. I think the problem is voltage.

Is there a way to build an interface to "bump up" the pulse train voltage so the PLC will see it? I can purhcase a device that will do this but lead time might be an issue.

If your encoder is just outputting a square wave, have you considered a solid state relay? I'm sure they are available with switching frequencies well beyond 40kHz.

Hi cjvanrem, I just finish an order of a MicroLogix 1100 and a flowmeter which will give pulse too but the power supply will be 24VDC, so maybe there would not be a problem like yours, but I am very interested in it, please let me know how you will solve this problem. Thanks.

With high speed counters, it is sometimes desirable to use a "pull up" or "pull down" resistor in order to improve the waveshape that the counter input sees. My practice is to examine the pulse train at the counter input with an oscilloscope to determine exactly what the signal looks like at the full speed condition.

To say this another way, with solid state inputs to the counter, you may be going back and forth between a defined input condition when the encoder output is true and an undefined counter input condition when the encoder output is false. The purpose of the resistor is to pull the input to the condition that is opposite to the direction when the encoder is true. Therefore, the counter input is always defined, one way when the encoder output is true and the other way by the resistor when the encoder is false.

This is a common issue with solid state logic and in this case, it is possible to have the same issues.

Best Regards,

Bob A.

My thinking would be opposite; if the MicroLogix "sees" the inputs at low speeds but not at high speeds I would not assume voltage is the problem. Of course, like Bob A. I consider an oscilloscope the only way to tell for certain.

The MicroLogix family fast inputs can be used as standard DC inputs to a counter or other ordinary logic instruction, or you can configure them to act as interrupts, or you can configure them to route the signals to a separate hardware High Speed Counter chip onboard the MicroLogix that has presets and outputs that are programmatically loaded and handled by the main controller.

The User Manual is, I think, very useful for understanding this, and there is a good Quick Start in the Knowledgebase.

The #1 novice mistake I see made with the MicroLogix 1100 is that the default Input filter is 8 milliseconds; be sure to configure it for a time that is more in line with the speed you expect from your encoder pulses.

Brilliant!!! Filter was set to default. Changing it to the smallest allowed (12.5 uS) makes everything work!!!

Thanks a bunch, Eddie. I'm still going to throw a scope on it just to see what it looks like.

I never argue with success, but you may want to be alert to the possibility of electrical noise giving you artifically high counts, with the filter set to the highest bandwidth.

Having thought about your original post over night, my thinking now is that your encoder output is probably "open collector" and likely pulling your input to very near the DC neutral. If you determined that 11.7 should be higher, you could test a pull up resistor in connected at the counter input to a 15 - 24 VDC supply. The normal range of resistance would be starting at about 10K down to maybe 2200 ohms.

It is important not to be lower than necessary so that you don't load down the output on the encoder because when it is on, it is now conducting for the counter input and the resistor and you don't want to "cook" it. This can happen on some encoders if they stop in the on state and everyone goes home for Christmas vacation! (It would not be too difficult to figure out how I know this).

If noise becomes a problem, I would combine this technique with some filter tuning from Eddie W, and that would give you improved resistance to electrical noise. One last tidbit, if the input to your counter is hovering around the logic threshhold where the counter actually counts transistions, it is possible to get a huge number of counts in one blast of noise. The job of the input filtering and the resistors if necessary, is to eliminate that dwell by the input signal, near the counter's logic threshhold.

Best Regards,

Bob A.

Originally posted by Eddie Willers:

The #1 novice mistake I see made with the MicroLogix 1100 is that the default Input filter is 8 milliseconds; be sure to configure it for a time that is more in line with the speed you expect from your encoder pulses.

Click to expand...

I thought the filter setting might be a problem but I couldn't find where 'default' means 8 msec. Where didi you find that little nugget?

Keith

kamenges said:

I thought the filter setting might be a problem but I couldn't find where 'default' means 8 msec. Where didi you find that little nugget?

Keith

Click to expand...

Thanks, Mickey. I missed that.Keith