problem with pt100 signal on generator

[dcset]

I have two generators 1,5 mw on a vessel, and they´re are used to supply the generation plant for the vessel and two syncronous motor used for propulsion, the level on harmonics on voltage is not to big, ok, the problem is that with more than 50% of the load, the signal that comes from pt100 are getting a lot of noise and peak that makes the monitoring system to give a signal of error, i have connected other pt100 on the connections on the generator, and the measure is right... this make me think that i have some problem with the pt100 onto generator slot... but this problem occurs to all the pt100 and in both generators, maybe somebody could helps me in order to know why this is going on. thanks

 

[dpc]

Have you tried looking at the signal right at the generator? This would help verify if the problem is in the RTD itself or in the external wiring.



David Castor

http://www.cvoes.com

 

[dcset]

yes, the problem is still if you measure directly on the pt100 connectors on generator.

 

[jimbofitz]

We have had problems at our facility with thermocouples having bad readings in a high magnetic field. It was assumed that the magnetic field was inducing voltage into the sensor that swamped out the low level signal generated by the element. The sensor was placed next to the core gap of a 3-phase inductor (the hot spot) during a prototype run.

Regards,
Jim

 

[dcset]

this is pt100 sensor, and if they´re into the slot, into hot spot, i think they must support such magnetic force without distortion....
and the most important thing, how to solve? cause i can´t install a new sonda onto a wind, is it dependent on harmonics or just about magnetic field.

thanks

 

[ScottyUK]

RTDs are used all the time in this application, so the problem is likely in either the grounding of the cabling or the RTD signal conditioner. My guess would be grounding: either lack of a ground or a ground loop. As it tests bad at the generator, a ground loop seems the less likely of the two.

What happens if you use a multimeter measuring direct at the RTD - noisy or stable?


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[dcset]

hi, if i read directly on rtd the value isn´t stable when you over 50% load, and this is why a think about harmonics or something like that cause the noise increase when you start propulsion motors.

 

[ScottyUK]

Are your RTDs connected using a twisted pair of conductors? How tightly twisted? If you have a lot of HF stuff then you might want to try a tighter twist to minimise the loop area available for coupling.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[dcset]

no they are not twisted and come from inside of generator with no twisted connection, but the influence is on cable or maybe on the pt100 sensor?? i don´t understand, and normally the cable from pt100 sensor is not twisted anymore

 

[itsmoked]

You can end up with common mode noise coupling onto the RTD that actually takes your measurement system's inputs out of their allowed operating regions. This seem to fit your case in that it all works fine up to a certain level then goes awry - just as you exceed the common mode noise limits.

You should consider using an RTD isolator that will remove all of the common mode noise signal so your measurement system remains within its operating limits.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ScottyUK]

Twisted conductors are the most economically viable solution to reducing inductively coupled noise. The key is to reduce the area of the loop enclosed by the conductor to a minimum, and after each twist the loop polarity reverses. Run twisted pair cable right up to the RTD element.

If you have a lot of harmonics then capacitive coupling may also be problematic. Can you get a 'scope to look at the input to the RTD signal conditioner? Be careful not to introduce an additional earth from the scope into the loop. If your 'scope has screen capture post an image or two.



----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[roydm]

Is the RTD grounded?
You say it's unstable when you measure directly at the generator does this mean the resistance reading is erratic?
If the RTD is grounded a 4-20 mA transmitter right beside the generator might help.
The wires should be twisted where under the influence of magnetic field for sure.
I haven't seen 100 Ohm Pt used in motors very often, usually it's 10 Ohm Cu with lower impedance.

 

[dcset]

no i connect directly a device to measure the resistence, have somebody some literature about this concept... and about that pt 10 ohms with lower impedance???
Now the ship is sailing.. but i want to think about

 

[roydm]

Sorry I was unable to find anything to back up my statement that the 10 Ohm Cu is better because of lower impedance but they do seem more common for motor windings. !0 Ohm copper are cheaper perhaps thats the reason.
If you can twist the leads as ScottyUK says I'm sure you will see a vast improvement.
Roy