Instrument Control (GPIB, Serial, VISA, IVI)
GPIB-120B causes noises
I have had this problem as well and I don't see a resolution on this thread. I believe NI is aware of this problem but hasn't resolved it. I have used the older GPIB-120A units for years with good results. These not only block DC ground loops between your computer and the instrumentation on a separately grounded AC power but also supress high frequency noise that is usually emitted by the computer.
Hower now that I have some experience with the newer GPIB-120B units I can verify that these suck. Not only do they put out huge (ie in the volt range) 60Hz and harmonics from the switching power supply they also radiate in the Mhz range! This RF immediately gets transmitted to my cryogenic equipment and makes achieving mK temperatures impossible much less making any kind of sensitive measurements possible. The GPIB-120A units had software quirks but at least were good to break ground loops and remove noise. The newer GPIB-120B are so much worse than nothing that it is absurd. It is almost fraud to market these as isolators.
Technically they isolate but the side effects and noise created obviates the advantages of the DC isolation.
The switching power supply in the dongle makes a squarewave from a switcher power supply, of course this has so many harmonics that seem to come straight through. There is no case ground from the power supply dongle to the case. The case is all plastic and so there is no case shield to attach to the power ground.
Installing a secondary shield doesn't help that much since there is obviously coupling between the input power and the shield and pins on the GPIB connector. This coupling is not only to the 60Hz and harmonics but to the MHz clock that the internal CPU runs at. The unit should have been designed with internal shielding and ground planes between elements.
Does anyone have a stash of the old GPIB-120A units? Or better yet, a currently supported and marketed product that can be used for noise isolation and reduction. Other GPIB isolators are IOTech (now owned by NI so suspect) and ICS electroncs and this unit locks up occasionally.
Just to follow up, from the NI AE
I got word from R&D and they are aware of the noise issue in the 120Bs.
Which is support speak for we just realized that we are shipping POS. This is real nasty and a joke for a device that has been on the market for a couple of years now. An isolator to reduce noise is actually the biggest source of noise. I hope I can get my hands on a bunch of the 120A units since it seems like it will be years before this gets fixed.
Well now I can post this since they are safely ensconced in my lab. But it turns out that NI had 5 of the old 120As left over. I managed to immediately purchase the entire lot. These have to last until they trash the model 120B and make a 120C where they pay attention to noise. I was happy to purchase all 5, but I shouldn't be relying on unsupported old equipment.
I don't know if I mentioned this before. In looking at the units they use a cheap cell phone type DC supply. These things are horrible noise sources since they don't have a transformer and chop the incoming voltage down to 5 V. So there is very very high frequency hash on the power lones from these things. Possibly replacing the power unit might solve a lot of problems. Those things radiate all over the lab, so never let people charge their cell phones on the lab bench. Also don't let them use them either since they can be a big RF source as they are designed to be.
Cheers,
Scott
