Daqcard 6036e affected by laptop power supply?

[cogswell]

Hello NI types;

I have a problem that seems pretty bizarre to me, so I thought I would share it with you to see if anyone has any suggestions.

I have a brand new Daqcard-6036E (pcmcia card), with shielded cable and the CB-68LPR terminal block, and I'm trying to use it with a Sony Vaio laptop (PCG-955A) with Windows XP professional. Labview 7, full installation with all the recommended drivers from the Labview 7 cdroms (didn't install anything from the 6036E cdrom, since I figured the stuff on the LV7 disc was probably more current anyway).

The card is recognized by MAX, Labview, etc. I can bring up the test panels no problem. I measure in differential mode, -10v to 10v range, I'm using the proper terminal pairs for the differential mode (aich0 + aich8 for diff1, etc). Problem is, regardless of what I put in for an input (I've used a little dc voltage supply, and bridging the terminals with a wire to enforce 0V) I get garbage not even resembling the proper input. It does random fluctuations all the way up to the input limits (-10/10 V). I measure with a meter and everything looks great at the terminals. I don't have any other terminal block connections (no connections from AIGND's, AO, timers, etc).

Here's the weirdness: the setup works just perfect... as long as I'm running the laptop off the battery and not using the power supply plugged into the wall. This was really, really annoying to try and find (pulling out the power supply to the laptop was the last thing I tried after fussing with it for a couple of hours). Battery only: reads correct voltages, nice and stable. Laptop power supply: reads completely incorrect fluctuating voltages.

I checked that I was trying to read differential mode (in MAX, and my labview stuff I've written), and I have no idea what to try next. I've used NI stuff for years (LV2,4,5, Nubus cards on macs, pci cards on pcs and macs) but I've never had an opportunity to use a pcmcia card before. I've never seen this behaviour on any other setup I've worked on.

I have a few suspicions: laptop power brick (which plugs in on the same side of the laptop as the pcmcia cage) isn't really nice 19.6 V dc @ 3A like the back of the brick says (have not verified this), maybe bizarre ground loop problems (I get this problem even with no external sources hooked to the CB-68LPR, just the differential inputs connected together on the channel). Maybe the power supply brick makes rf interference that annoys the card (I tried moving the brick farther away with no difference). Maybe Sony just doesn't like me.

As a dumb test, I ran the laptop's power brick plugged into a 120v inverter running off a 12V car battery (ie - 12Vdc -> 120Vac -> 19.6 V dc -> laptop) and that made no difference (still didn't wdfdfork). My other thought is to try and run the laptop directly off a regulated dc supply fed by a 24V battery setup, but that'll take me a little while to assemble to test.

Any thoughts on this? Sadly I do not have a second laptop to try this all in to see if that rectifies things. Has anyone else encountered this?

I hope I haven't left out a salient detail. If I forgot to mention something relevant please ask.

Best regards,

Steven Cogswell

[Ben]

I think you are the right track with your ground loop thought.

Check your outlet wiring.

Try one one those old adapters that let you plug three wire plugs into 2 wire outlets.

Most Laptops will float relative to ground unless you take action otherwise.

Ben

[cogswell]

Hello Ben and others;

Indeed, you hit the nail on the head, except it was the other way around. I hadn't taken notice of it before, but this laptop (which isn't mine) doesn't even have a 3-prong plug on the adapter in the first place, only 2-prong (the cord/brick have no facility for a 3rd either, and it's the original sony adapter). I ran a connection from the AIGND (AIGND/AOGND/DGND on this card are all the same point according to the manual) to the 120v ground (3rd prong) and sure enough, everything cleans up and voltages are correct again.

And a small note for those taking notes, although I said the laptop is a "PCG-955A" (that's what's on the bottom), it's actually a Vaio FX120 (written in tiny print on the screen bezel).

Thanks for all you
r help,
Steve

[Ben]

I have told people for years, "Don't ask me what the problem IS, let me work it some more and maybe I'll be able to tell you what it WAS."

You were probably exceeding the CMR spec of that device.

Thanks for the feedback, its nice to know what the final answer was.

Ben