Encoder losing position

So,, Why not figure out EXACTLY how many pulses per rev you are getting. Use that value times the 104 and then you should have your SCP input upper limit and the HSC reset Value...

Does the encoder actually say 1024 pulse/rev on the label? Silly question I know but it seems like something isn't right with the amount of pulses your getting vs the math and values you are using.

PLC Pie Guy said:

So,, Why not figure out EXACTLY how many pulses per rev you are getting. Use that value times the 104 and then you should have your SCP input upper limit and the HSC reset Value...

Does the encoder actually say 1024 pulse/rev on the label? Silly question I know but it seems like something isn't right with the amount of pulses your getting vs the math and values you are using.

Click to expand...

The encoder does say 1024 on the label. I'm not sure how I can turn the encoder exactly 360 degrees. I turned it as close as I could by eye and I got 1000 counts - so I have to assume that 1024 is accurate?

That's your problem. The pullup resistors are mandatory. Without them, the faster you go, the more pulses you miss. Figure 15 in the 1769-HSC manual shows correct wiring for Open Collector. Add 2700 Ohm, 2 Watt resistors.
It would be better if you could replace the encoders with TRD-S1024-VD.
You would need a 5 Volt power supply. Use Figure 14.1769-HSC link.

keithkyll said:

That's your problem. The pullup resistors are mandatory. Without them, the faster you go, the more pulses you miss. Figure 15 in the 1769-HSC manual shows correct wiring for Open Collector. Add 2700 Ohm, 2 Watt resistors.
It would be better if you could replace the encoders with TRD-S1024-VD.
You would need a 5 Volt power supply. Use Figure 14.1769-HSC link.

Click to expand...

You are the man. I'll try this soon!

As a note - by 'quadrature counting' I was just referring to the use of the A and B signals which are 90 electrical degrees out of phase. The diagram and your 'X1' setup shows that you are doing this.

if it is really wires as shown on the print then there may be a possible problem
I see 2 load resistors on channel A to common and none on channel B
I think it's just drawn wrong but worth checking

I see the setup in pulse and direction I don't think that's right
also the setup shows encoder X1 I think it should be X2
the HSC reset should be a digital output to the HSC these are edge triggered
(Reset the HSC counter to 0 )
the min max should work for you depending the application

Here is my take on the situation. I am assuming this machine was running fine with the encoder before from what has been previously posted:

Encoder and linkage has been ruled out. At this point there is no need to beat a dead horse.

Check the grounding on the machine to a nearby earth ground. Run a temp wire across the floor if you need to to test. Test for both potential and resistance to verify that both machines are properly earthed. Less than .1VAC and as close to 0 Ohm is what you want to see here. If it isn't, this may not be the issue, but then again it may and should probably be addressed.

Verify all encoder wire terminals are free of corrosion and connections are tight. I like to pop them off of the terminal blocks and make sure there are no strands possibly shorting out, all wires are making a good connection and everything is good. The last time I checked this on a machine I found an input that had a single strand connected and the rest had broken off.

You may also have a pull-up resistor or such in the encoder circuit. These are cheap. Try swapping with one from a known-good machine or with a new resistor or reading the resistance of said resistor and seeing if it is within the tolerance, if there is one.

The next step can be either of the following, whichever is cheaper or more easily available:

Grab another known-good/new encoder cable and drape it over to the cabinet and install it. See if this corrects the issue.

Take another PLC unit/card and do a quick swap-out to see if this resolves the issue (the last time I dealt with this it actually ended up being an intermittent PLC high speed counter input issue. Everyone told me it probably wasn't the PLC, but when I replace the brick the issue went away and is still working fine to this day.)

Post back and let us know. Having tried new a new encoder and then going back later to verify the coupling is tight, and having the machine run just fine before tells me that something else has went wrong here. Post back and let us know what you find.

Is the error constant or does it vary with each test?

If the error is constant then it is probably a fault with your mechanical conversion.
If it varies then it is probably an electrical issue.

Does the error get worse if you go faster?

Unless the gearbox is planetary then the ratio is extraordinarily unlikely to be an integer. Get the tooth numbers for each stage of the gearbox and get yourself true ratio.

I'm getting ready to try the resistor as shown in the HSC manual. I believe our drawing has two in parallel because the designer assumed we would use an 1/4 watt resistor.

I don't think the machine ever worked properly - or we just didn't know it. I've been here 5 years and it's never worked properly. We've gone from R&D mode to production and now this problem has come to light.

I'll post what I find shortly.

I have two encoders attached to a Flex IO HSC card for a flying splice/cutoff operation. They both worked flawlessly for a long time, but then the one started intermittently getting off.

I finally found that the HSC card had filtering options for each channel. The encoder that always worked had the filtering enabled. The filtering had been turned off on the other.

Don't know if your card has filtering?

Ok so here's what's up now. The resistor didn't work. However, so now I noticed if I jog this wheel (not using auto mode) the encoder is dead on.

I can't put my counter into X2 mode because it crashes the modules. Not sure why. Filtering does the same thing. The HSC won't come online.

Do your counter probes have an X10 mode like 'scope probes? You need high impedance. Put your DVM on the outputs and measure voltage as you slowly rotate. Are we switching close to 0-24V?
I looked closely at the specs for your encoder and HSC module. If those are correct, then I stand by my statement that pullups are mandatory. Your schematic shows a lower resistance than what I proposed, but I can't make it out. Looks like 1.8K. 2.7K might be too weak. The voltage swing will tell us.

They called out (2) 2.2k in parallel. I used what you suggested at 2.7k.

FYI the jog mode is exactly the same speed as auto mode (5 hertz); so it's relatively slow to begin with. I'm not sure why it's happening in one and not the other.

I have X1, X2 and X4 modes available. Again the x2 mode crashed the modules and wouldn't come online.

I will still measure the switching voltage as you suggested.

1100 Ohms gives 20mA. 2700 Ohms is ~8mA. You can use the two 2.2K parallel, or a 1K or 1.2K. The total wattage should still be 2 Watts. For parallel resistors, they can be 1 Watt each.
Take a look at return paths and grounding. Why does X2 crash? What voltage is being pulled down?
Jog and Auto same speed, so it's something else. What else turns on in Auto? How is this thing turning on affecting your encoder? How is it able to knock HSC module offline? Monitor voltages. There's a weird circuit here somewhere. Hope looking at voltages/ return paths will give a clue.

20mA gives better noise immunity. 1/4 Watt is fine. (I was 1 decimal place off in my head. Calculator corrected me).

I measured the on/off voltage on the encoder side of the resistor to ground. I'm showing roughly 5.5v to 12v when turning the encoder very slowly.

In auto there's a another motor with an encoder that turns on located about 6 feet away. There's also a traverse motor that intermittently turns on about 3 feet away from the encoder. The power lines for these motors are close to the encoder cable, but it's shielded and grounded so I assumed that noise was not an issue.