I'd love if some knowledgeable folks were able to tell me if I'm on the right track with my understanding here:
Usually with regard to the V-Guard on a 5700 series calibrator, I usually only do what the K-pro calls for. I've never had a trainer who can explain exactly what it does, and the best explanation that I have found in the manuals is that it's a shield that goes around the active circuits, and between the secondary and primary winding on the power transformer. To me, that sounds like it acts as a faraday cage, shielding the circuits and test set-up from external noise.
I might never exactly understand the theory behind it's use, but I wanted to at least have an understanding on HOW and WHEN to use it. From the operation manuals, diagrams, and k-pros, I've interpreted the following pointers. When I say UUT, I'm really talking about anything you're connecting the 5700 to. I'd appreciate any input on the following:
- Use of the guard is REQUIRED when the UUT's Low input is internally (or externally) connected to the chassis ground (regardless if that chassis ground is connected to earth ground).
- Also when the UUT's Low input is connected to chassis or earth ground, the "EX GRD" button should be pushed so that it's illuminated. (which disconnects the guard internally from the Low on the 5700). This is the only time the button is used.
- The Guard-to-Ground shorting bar on the 5700 should be disconnected if the guard wire is connected to earth ground on the UUT. This prevents ground loops. (In any test set-up, there should only be one common earth ground connection with the guard.)
- The 5700's guard should be connected to the UUT guard, if it has one.
- If the UUT does not have a guard connection, but has a ground, it can be beneficial to connect the 5700's guard to the UUT's ground.
- With DC and low frequencies, use of Fluke's cables that incorporate a blue guard wire are optimal. Higher frequencies use a triaxial cable, with guard connected to the outer sleeve.
- Guard can be used when calibrating handhelds to shield the cables. Connect the guard wire at the 5700, and leave the wire at the handheld disconnected.
Is that everything? Did I get it all straight?
I think you have it, mostly. Your third to last one would make sense if you don't short the guard and ground terminal together on the calibrator. As you probably know the 5700 shorts the low to guard terminal when in OPER mode unless EX GUARD is enabled. So if the guard terminal is linked to the ground terminal, you could potentially create a ground loop if your UUT's ground is connect to the calibrator's guard.
A guard circuit is typically used for reducing parasitic capacitance and leakage currents. So if you connect DC to a high impedance load (say a voltmeter), you have an error due to leakage currents and the guard, especially if "driven" (placed at HI potential, shorted to HI terminal), helps prevent this. When sourcing high-freq signals, parasitic capacitance can also be reduced with the guard.
To add, the best thing to do is play around with the different configurations until you get the measurement you want!
You've obviously read the manual. After 30+ years of this that is what I do even when I think I know the answer.
An interesting side note: Quite a few years back at the GAFB PMEL I took over the DC standards section and the procedure for verifying the working standard cell called for using shielded ESI leads and we had the exact leads called for in the procedure. Excecuted exactly as the procedure called out, it simply did not work with those leads. The only thing that worked was using plain solid Copper wire and standing outside the room while the measurement was being made. (Also know as playing around with different configurations until you get the measurement you want)
Quote from: N79 on 09-01-2016 -- 19:17:03
To add, the best thing to do is play around with the different configurations until you get the measurement you want!
I heard this once from a guy doing torque wrenches... Let's say he didn't work out. :-o