Analog Input Noise

[rmayer]

Have 2Vac riding on a 6Vdc 0-10V analog input line driving the program bonkers. I suspect a dying rectifyer. I already have series resistors in line that scale the signal to the 0-10V range. The rectifier has a smoothing cap with bleed resistor. Grounds are all at a single point on the chasis but the PLC and control has a seperate power supply from this recifier. The ripple is making all of the inputs on this IO block fluctuate.

I'm working on ruling out ground loop.
I am wondering if I could use an AC choke on this line but suspect the plc wouldn't be enough of a load for it to work.
I have ordered a signal conditioner that will isolate and transform to 4-20mA but would prefer a 'quicker than UPS' fix.
I put a 22000uF cap across the input but it didn't do much to steady the signal. To top it all the scope is down so I can't completely quantify what I am seeing/not-seeing.

I'm sure you are all better at this than I am so how bout showin off a bit of that brain power and helping a brother out?

 

[itsmoked]

Is this a new problem? Or a start up problem? Was it working fine for any length of time previously?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[rmayer]

This is a new setup. We were simulating this voltage feedback with a pot in test. Now that we have switched to pull the voltage feedback off of the rectifier output we are getting the fluctuations.

 

[roydm]

So you have this large rectifier feeding some sort of DC load
The rectifier has an analog output 0 - 10 VDC
When you connect the rectifier analog output to PLC the signal is very noisy.
Certainly sounds like a ground loop to me. Here are a few simple tests you can do.
Try disconnecting the positive wire at the rectifier and substitute a small battery between negative and the lead going to PLC (leave the negative connected to the rectifier).
If the PLC is still noisy remove the negative wire from the rectifier so you now have just the battery connected to the PLC, nothing else.
Do you have a shielded pair Rectifier - PLC? The shield should be grounded at one end only, usually the PLC end.
If the rectifier's signal is referenced to ground (measure from ground to negative terminal with Ohmmeter) you will likely need a signal isolator, if so you might be better off with a 4-20 mA version assuming you have 4-20 mA PLC input option

How large is your rectifier?

In the absence of a scope you could substitute a small speaker and series capacitor to track down the AC noise.
Hope this helps
Roy