My experience with large channel counts is that you should separate the shield ground from the data acquisition system. The current from the shields can be very high especially for large channel counts. What would be NI's take on this? I think its difficult to guarantee that the ground used by the chassis is a true earth ground.
Solved! Go to Solution.
What hardware are you using in this system? Are you already using a shielded cable to connect to your data acquisition system?
As far as your concerns with the chassis ground not being a true earth ground, this is correct, and if you need to make special considerations for your system to connect the grounds together to an earth ground then you could tie the chassis to a true earth ground or separate shields from the data acquistion system. Since the specifics for the system will vary depending on the environment and setup, testing must be done to determine if there is any additional steps to take on setting up the system. Generally, the current will be very low in the shielding and have negligible effect on the system.
I think I can agree with low channel count systems. However for large count systems (say 100 or more) could mean a hight current. Each shield is parallel which means the shield current is added to each other. If you had 1000 channels would that not mean you could have at 100 mA per shield, 100 amps running to the ground?
I forgot to ask about the effect of having the current running through the chassis along with the digital ground. Would this interfere with the operation of you chassis?
I forgot a detail that one of the chassis will be located remotely and therefore will not share the same power outlet.
For specific systems, we cannot recommend how to set up shielding due to the complexities that lie within the application. What I did though was found an article here describing some shielding practices and considerations for large channel count and high shielding current applications.
I did check and even if there is high current flowing through our chassis, and it will not affect the operation since the current will be distributed over several chassis. The current I am assuming you are working with is around 30A max since 300 channels at 100mA max.
I understand. It would be better to be conservative in this case. Its impossible to know how much noise will be induced in our environment. The shields will go to a large plate (true earth ground) and not go through the NI equipment.
Perhaps NI will try an experiment with 1000 or so channels in a noisy environment to test out the PXI chassis.