Plug in power conditioners, thoughts?

[JCH13]

Pretty sure this is the most appropriate forum for this.

I've got some very sensitive DAQ equipment at work that's seeing tons of 60hz noise, no doubt from the factory equipment where the DAQ equipment must be. I have a strong suspicion that some or all of the interference is coming from the equipments power supply (i.e. the wall).

Does anyone have any suggestions for the absolute best, and best bang-for-the-buck, plug in power conditioners? Thanks in advance.

 

[Mark R]

The optimal solution differs depending on the nature of the interference.

Most electronic equipment has extremely strong rejection of noise, harmonics or interference at their power inlet, which makes me wonder if the problem really is in the electricity supply. If mains hum is being picked up, it could be a grounding problem - e.g. leakage current, ground loops, excessive ground impedance, etc. These are much more difficult to deal with.

Similarly, if this is a heavy industrial unit, large currents flowing in distribution wiring (especially if 3 phase cables are laid seperately or through individual conduits rather than the recommended arrangement of being clamped into a trefoil arrangement), or nearby transformers leak magnetic flux, which can induce 60Hz hum in signal cabling.

If you have something like a UPS, try powering the DAQ from the UPS, but keeping everything else the same (ensuring that you keep the ground connected when the UPS is running on battery e.g. use a switched power strip, rather than simulate a power failure by pulling the plug).

 

[bryanl]

A simple line filter can cut higher frequency noise by a factor of 10-100 (20-40 dB) and be very effective against motors turning on and off but are totally ineffective against brief brownouts that last even a few tenths of a second. The better plug-in surge protectors contain good line filters.

Doesn't mains hum power all of the equipment? On the other hand, ground loops can be a major problem in some industrial environments.

 

[Concillian]

If it's seeing 60Hz noise, that's likely seeing the AC.

This sounds like an internal design issue. A power conditioner is still going to have 60Hz AC. Sounds like your internal AC to DC transformers are not suppressing this well enough.

Have you tried different circuits and such? I've had cases at work where lines had a poor connection somewhere in them. They'd spec fine at no load, but pull 10 amps and they look like garbage.

In your case, you may have some success with a REAL UPC that runs with a battery always in the circuit. Most UPCs will run from the wall until there's an issue and switch to the battery very quickly in the event of an issue. The expensive ones always have the battery in the circuit. This can provide some inherent noise filtering. However, I still think if you're seeing 60Hz ripple, that it's likely to be coming from the internal power supply for the DAQ.

 

[JCH13]

Thanks for the responses so far.

I haven't been able to do any debugging personally, I'm flying to the client's lab in about a month, so I need all my solutions ready before the trip.

The primary issue is that I can't just low-pass filter everything because we need a 1khz sampling rate and we're seeing 60hz noise. A simple line filter won't do it. It could very well be a ground leakage issue, in which case a power line filter would help.

One of the backup plans is a different DAQ system with (hopefully) a better power supply. However, I did a thorough shake-down of the system in my lab, literally 20 yards from a power line transformer, and didn't get any 60hz noise, so power supply issues seem unlikely.

 

[Modelworks]

It would take a ton of EMI to make it through the typical shielding of the metal case of a lab type DAQ.
The very first thing I would do is connect a scope to see exactly what the waveform looked like and the strength of it and then try to locate the cause.
If I couldn't determine the source , I would isolate the device with an isolation transformer, breaking any ground loops and possibly removing all the EMI. If the EMI still persist it may be necessary to place more filters in the line, although I haven't seen much that an isolation transformer can't stop.
http://www.tripplite.com/en/products/model.cfm?txtSeriesID=839&txtModelID=227

 

[Rubycon]

When shopping for "best" and "bang for the buck" you may only choose one, not both.

A ferroresonant transformer based line conditioner will provide outstanding isolation.

 

[JCH13]

It would take a ton of EMI to make it through the typical shielding of the metal case of a lab type DAQ.
The very first thing I would do is connect a scope to see exactly what the waveform looked like and the strength of it and then try to locate the cause.
If I couldn't determine the source , I would isolate the device with an isolation transformer, breaking any ground loops and possibly removing all the EMI. If the EMI still persist it may be necessary to place more filters in the line, although I haven't seen much that an isolation transformer can't stop.
http://www.tripplite.com/en/products/model.cfm?txtSeriesID=839&txtModelID=227

I'm pretty sure the noise (definitely 60hz, micro/milivolt range) is getting into the thermocouple wires connected to the DAQ unit, or a ground leakage to the DAQ unit directly. As I said, I am offsite now, and will be flying in to do the debug and fix, so I need to have every option and fix with me when I go.

When shopping for "best" and "bang for the buck" you may only choose one, not both.

A ferroresonant transformer based line conditioner will provide outstanding isolation.

I was hoping for two options

Any examples of a ferroresonant transformer based line conditioner?

 

[Rubycon]

Any examples of a ferroresonant transformer based line conditioner?

http://www.emersonindustrial.com/en...-conditioning/mcr-portable/Pages/default.aspx