Floating Power Supply Grounded Through Signal Negative Terminal

[fifty60]

I have a humidity sensor that I am connecting to a 24VDC Class II power supply. I leave the Class II power supply floating because of instrumentation connected to the power supply. For example, there are sensors that have 4-20mA outputs, as well as 0-5V outputs powered from the power supply.

I also leave the power supply floating to prevent the unintentional grounding of ground referenced test equipment.

When I check for continuity between the ground ring on the Humidity sensor board, and the - com terminal on humidity board, I get no continuity. But, when I power up the sensor, i have approximately 22Volts between the humidity boards -com terminal and the ground washer. My guess is that this means they are coupled together internally by a diode.

The -Com terminal on the humidity sensor is also connected to the - terminal of the 24VDC 60W Class II power supply. This means that my entire Class II 24V circuit is now ground referenced.

Why would the Humidity sensor want to force the circuit to be ground referenced? It has a 0-5V output. The wiring from the sensor to the PID controller it is connected to is about 10+ feet routed through the machinery. Having the signal referenced to ground at two seperate places on the machinery cannot be good for the 0-5V signal...

My question is, why would they have it internally connected to ground like this? The humidity sensor can run off of 24VAC or 24VDC...would the ground ring be for an earth ground if the power supply is a 24VAC transformer? To me, the ground lug on 24VDC Class II power supply powered humidity sensor serves no purpose...:thumbsdown:

 

[hbiss]

Not knowing anything about your particular humidity sensor and how it works I have no idea. Your question is probably best directed to the manufacturer.

-Hal

 

[fifty60]

It is a solid state sensor with a 10-30VAC/DC power supply input, and a 0-5V output signal.

I plan on asking the Manufacturer, but am trying to understand better first. It just does not make any sense to me why they would internally connect the signal common to the ground ring on the rh sensor printed circuit board. I do not see any advantages of this.

It can't be for safety since the sensor is powered from a low power/low energy 24VDC Class II supply. It definitely does not help the 0-5V signal output out when compared to a floating supply.

 

[Sahib]

One reason may be: Higher voltage line when accidentally contacts lower voltage grounded supply, its voltage is nearly brought down to that of lower voltage instead of lower voltage supply voltage rising nearly to that of higher voltage!

 

[fifty60]

Understood. The RH sensor is separated from the higher voltage lines as a matter of good wiring practice. I do not see any functional reason to ground the sensor.

My main concern is the 0 to 5V output signal. The RH sensor is powered from a 24VDC circuit, and the PID loop controller it is feeding its 0-5V signal to is powered from the secondary of a 110V transformer, which just so happens to be the same transformer that powers the 24VDC power supply.

There is a signal - on the RH sensor, and a signal - on the loop controller. If these are physically located in different locations, then wouldn't having them ground referenced compromise the 0-5V signal? Even if the ground potentials were off by a 1/4 of a Volt, this would be pretty huge discrepancy in the sensor reading...

 

[fifty60]

I understand how a grounded supply can HURT the 0-5V output signal, but I do not understand how they can positively help the 0-5V signal...can anyone please explain?

 

[Sahib]

I understand how a grounded supply can HURT the 0-5V output signal, but I do not understand how they can positively help the 0-5V signal...can anyone please explain?

No positive about a ground loop current. Try to break the ground loop at least at the controller side, if possible, by disconnecting the grounding there if no safety issue. However I think unless there is any malfunctioning in your case (no sensitive measuring instrumentation), you need not bother.