1520 is supposed to have differential inst. amp. Then why we need to refrence floating voltage signals to measure correctly ?

[Rajendu Choubisa]

Hii,

1520 Revision B can take floating voltage sources and give you correct readings. While in Revision C, referencing the floating source seems to be necessary to get good readings. If not referenced, the readings start drifting.

If 1520 has differential inst. amp, why in first place we need to reference the floating signal ? What changes have been made in revision C that makes it different from revision B in terms of measurement of floating voltage source ?

Need a quick answer .. .

Regards ..
Rajendu Choubisa

[Logan_K]

Rajendu,

I am sorry I didn't get back to your email yesterday. The 1520 does not support measuring high common mode signals, and if the customer wants to measure high common mode signals then he needs an isolated SCXI module, which the SCXI-1520 is not. The 1520 can measure a maximum working voltage (common mode signal + the differential signal) of 10V.

Floating voltage sources are susceptible to inducing a common mode voltages that could exceed the maximum working voltage of 10V on this module. Bias currents and noise can both play a large part in producing a common mode signal, and even if certain amplifiers come from the same manufacturer, different batches may have different bias current levels. My guess is that this is what the customer is seeing.

There isn't much we can do for this customer since it is well documented that floating signals need to be referenced to ground if you are going to use them with a non-isolated board.

-Logan

[Rajendu Choubisa]

Hey Logan,

Thanks for your response. But I still have some reservations in accepting your reply.

We were using a 1.2V battery as floating source. Can this small floating voltage input also have such a high common mode voltage ? Also, our observations were that first few thousand readings were fine, only after that the readings started to drift randomly. Does that mean that sampling the channel also increase or affect common mode voltage ? Can this also be an issue of actually 1520’s mux getting affected with floating signals ?

Also, I would like to know more about actual referencing signals. Is excitation reference same as chassis potential ? And is chassis shorted with input power common ?

Your help is really appreciated …
- Rajendu Choubisa

[Rajendu Choubisa]

Logan,

Just one more thought. How can I measure the common mode voltage ? Can I just take a DMM and measure it ? What would be the reference (absolute?) for that ?

-Rajendu Choubisa

[Logan_K]

Hi Rajendu,

You can measure the common mode signal by connecting the DMM to the minus terminal and referencing it to the chassis ground. You said that if the customer grounds his source he doesn't have any problems like what you just described? If that is the case, then I am pretty sure that the problem has to do with bias currents not having a path to ground that are causing the common mode voltage to build up. Is the customer scanning multiple channels at a time when he sees the issue?

If this problem shows up regardless of whether or not the signal is grounded, then since Rev. C are the modules that he is having issues with - I would advise you to get the serial numbers of those modules, because we had a recall on some of the Rev. C SCXI modules.

-Logan

[Rajendu Choubisa]

Logan,

The serial number of 1520 modules showing this kind of behavior is as follows:
1. D705C6
2. D705CC
Part number for both is 186338C

As you suggested, the problem might be with the bias current. We observed the problem occurring when we had not referenced the floating signal with chassis. The problem was NOT reported when the input floating signal was referenced with chassis, even while the chassis was not grounded, i.e. not externally shorted with power supply’s common. BTW is chassis and power common the same ? I could not read exact (~µV) Zero but some mV between chassis and power common.

We were trying to read from only one channel. Still, I was just wondering if mux can still be a possible cause ! This is because I believe even if we are scanning single channel, the scan clock will actually go through all the channels. (Atleast in our DAQ boards) Please correct me if my understanding is wrong.

-Rajendu Choubisa

[Logan_K]

Those 1520 modules were not affected by the recall - so that isn't the issue. If you are only scanning one channel, then the mux won't move. The chassis and power common are connected together at some level, but depending on where you make your measurement it is natural for different ground references to have a small potential between them.

-Logan