NI USB 6009 & 6002

[Nghtcrwlr]

Hi all

I am having an application that uses NI USB 6009. Due to the offset problem with 6009 , i was planning to swap it with 6002. In program itself  there isnt anything that needs to be adjusted for the new 6002. In NI Max  the device name is changed like that of NI 6009 and test panel runs as wished. But when i run the exe file, error occurs at DIO Port P0.6.

I was wondering whether 6009 and 6002 have any differences in DIO resolution.

[GerdW]

Hi Nightcrawler,

I was wondering whether 6009 and 6002 have any differences in DIO resolution.

NI provides manuals for all their devices - just read them: 6002, 6009!

The 6002 has one more DIO channel (P2.1) when compared with this 6009 device.

But when i run the exe file, error occurs at DIO Port P0.6.

Which error does occur?

Which VI is used to read/write to your DIO channels? It's hard to debug when you don't provide any VIs…

[Bob_Schor]

---

@Nghtcrwlr wrote:

Due to the offset problem with 6009

---

What Offset Problem?  There are a lot of 6009's out there, and if there is an Offset Problem, you would be doing me (and potentially the LabVIEW Community) a Service by explaining what you mean.

Bob Schor

[johnsold]

Bob,

That one is easy. The USB-6008 and -6009 have the incredibly unconventional, annoying, and error-inducing analog input circuit with a Thevenin equivalent of 144000 ohms terminated in 1.4 V.  See Figure 7 on page 16 of the User Guide.  Any $10 DMM has an input circuit with >1E7 ohms and zero volts.  I find it hard to imagine that NI actually saved any money by using this input circuit.

For the USB-6008 a signal with a source impedance >70 ohms will have 1 LSB error due to the low input impedance. For the -6009 the limit is below 18 ohms.

A 50 ohm resistor to ground will have ~0.5 mV across it. At 1000 ohms the offset is 10 mV, which is 1 LSB on the -6008 (single ended).

If you always use sources with adequate signal conditioning so that teh source impedance is very low, it will not be a problem. But if you have the resources to do that, you would not be using the usB-6009 anyway.

Lynn

[Bob_Schor]

Thanks, Lynn.  I'm embarassed to say I didn't realize this silliness.  I've generally used the 6009 as a "demo" unit to teach principles of data acquisition and sampling -- not being an EE, I've often made (erroneous) assumptions about circuitry ...

Bob Schor

[JÞB]

Thanks for asking Bob.  I got to learn something new today!  Nice reply Lynn. IDNKT

[altenbach]

---

@johnsold wrote:

That one is easy. The USB-6008 and -6009 have the incredibly unconventional, annoying, and error-inducing analog input circuit with a Thevenin equivalent of 144000 ohms terminated in 1.4 V.

---

Thanks Lynn.

The 6009 is a legacy product and no longer recommended for new designs. It seems the newer series (6001-6003) have a different setup. Can you have  look at the manual to see if this is true? Just wondering.

The 6002 is listed as a replacement for the 6009, and looks faster, better and cheaper (AI: 50kHz@16bit vs. 48KHz@14bit, AO: 5kHz@16bit vs "static"@12bit, 13DIO vs 12DIO, $329 vs. $369 list price).

Now we still need to find out the cause for the DIO error is (DIO Port P0.6). Since the 6002 is a newer product, have you upgraded e.g. DAQmx, etc.

[JÞB]

The OP did say. Executable.  Hey, how about attaching the nidaq config file?  Appdir/data/ni-daq/*.ini. Or very close that name.  File paths on mobile

[johnsold]

I looked at the USB-6002 specifications (in a different document). I can understand publishing the specifications in a separate document because many times that is the only information someone wants to look at. However, the User Guide should also include the specifications or at least have a prominent link to the specifications document.  It has neither.

The AI input impedance is >1E9 ohms with a bias current of <200 pA so these devices should be OK.

The 6002 has hardware timed analog outputs while the 6009 has only software timed AO. Both devices have only software timed DIO. In both the counters are only configured for edge counting, not pulse generation or period counting.

Lynn