USB-6008 5V External Power Problem

[Jamie Fraser]

Dear NI,

Anything to report on this?

Thanks

Jamie

[stilly32]

Hi All,

This has been reported to R&D under issue #440DPQJ4.  Just talking to the developers, it is a software issue with the board that will be fixed in the next version of DAQmx, but I don't have a concrete timeline for release. As soon as we know for sure when the next version  is going to be released I will post to this forum. Also, you can check the drivers page to see when we post a new version of DAQmx (the current version is 8.3.1).

Untill then, your fix with the transistor is a good work around. If anyone comes across other fixes, I encourage you to post them here. 

Cheers, 

Andrew S.

National Instruments 

[Jamie Fraser]

Andrew,

I wasn't expecting it to be a driver issue, or is it a firmware issue that will get flashed by DAQmx when an update is released. Does DAQmx have the ability to update firmware transparently when it detects an old firmware version? Thanks for the update. Regards.

Jamie

[Jamie Fraser]

Andrew,

The workaround that I thought worked OK (driving transistor for 2.5V line) does not work reliably. It works OK on an old machine (W2k, USB 1.1) but not on newer machines (XP, USB 2.0). I have looked at the way in which the 2.5V and 5V lines come up using a scope and it is not simply a case of one leading the other. The only reliable workaround seems to be to only switch the load on once the 5V line has reached about 4V or so. This ensures that all activity on the 2.5V and 5.0V lines has come to an end before the load is presented.

Seeing as this is causing a lot of problems for me and at least one other customer, can NI not issue a patch or something to fix the problem before the next release of DAQmx?

Jamie Fraser

[Doug Long]

Dear NI,

We too are having issues with reading AI channels.  We have been using the 6009 for awhile (past year) and have just started using the OEM version and are now seeing issues like the ones described in this thread.  We unloaded the USB 5V (used external 5V supply) and it did not improve the problem.  We then went into "work-around" mode and decided to feed the otherwise unused USB 5V output right into an unused AI channel (single ended) to use as a sort of watchdog (if 5 volts go away, try, try, again): 

In our application, we are only using channels 1 and 4 and had all the other unused channels are AI GND'ed.  For reading the USB +5V I chose channel 5 (AI5) because it was convenient.  We then changed our app S/W to read the offsets and if channel 5 did not report something close to 5V we could then run a recovery routine (et al).  But what we observed was when the AI went belly up, we would read the 5V (physically attached to AI5 and verified when working properly) on channel AI 4!  It is like the imbedded 6009 OEM S/W has lost its way while MUXing its AI channels. 

What is the short term fix before NI "gets around" to updating the driver S/W?  The problem appears to be more seroius to me -- If there are users out there using all the channels with similar input levels, they may not even know there is an issue!  The card does not read AI inputs consistantly enough for us to commit using this product in production at the present time.  Is there an easy patch for those of us trying to ship product?

thanx

[MarkLe]

Hi Doug,

That doesn't sound like the same problem as the rest of the thread if you are not trying to use the 5V output on the 6008 to supply significant current and power external devices - the problem above is that the 2.5V reference which is used for analogue input does not start up if the 5V line is under significant load, once the reference HAS started, the 6008 can cope with higher current draw from the 5V line.

It sounds to me more like the 5V supply on the device is being drained somehow else if you find that the supply collapses in use - if this supply isn't maintained properly then you may start to find problems with sampling similar to the ones you are seeing.

Are there any connections on the device which may be shorting the 5V supply to Ground? As you're using the OEM device - could there be any possibility of short circuits on the back of the PCB itself? Does this behaviour happen on several different units, or just the one? If the problem only happens on one unit - it could possibly be a faulty unit.

Is the unit connected to a 'healthy' main USB port - ie: if you're connecting it through a hub you could find that unless you power the hub externally there may not be enough power to service all the devices connected.

If you could post some more details about exactly how the product is connected up (to the PC and externally), and how often and in what circumstances you find the supply fails, we can try and reproduce this here or flag it up as an issue with the device.

Best wishes,

Mark

NIUK and Ireland

[stilly32]

Hey All

DAQmx 8.5 is available, you can download it here. This should resolve the issue with the 5 sourcing on power up and the 2.5 Vref.

Cheers,

Andrew S

[nealpederson]

With the new DAQmx 8.5 the main power-up problem has gone away but now I'm
still seeing what seems to be a related problem. Now, if I increase the
power draw but still keep it within the ratings of the USB6009OEM, I see the
same problem but not always at startup. Sometimes it happens after having
the device on for a while. I then unplug the USB connector and have to wait
a few minutes before I can plug it in again and it will work. Then if I
leave it on for a while it will happen again. Sometimes it does it when I
first plug it in too. When the problem happens my voltage readings stay
close to one level and the +5V that comes from USB6009OEM connector reduces
in voltage significantly. My old solution of delaying power draw when
everything is plugged in does not fix this problem because it happens after
everything has been plugged in for a while. Right now my only work around
is to get my +5 volts directly from the USB connector and this is not an
elegent or easy solution. Can NI fix this?

"stilly32" <x@no.email> wrote in message
news:1173111015621-486570@exchange.ni.com...
> Hey All
> DAQmx 8.5 is available, you can download it <a
> href="http://digital.ni.com/softlib.nsf/websearch/E2C96196892A649D86257291004F8EDC?opendocument&amp;node=132060_US"
> target="_blank">here</a>. This should resolve the issue with the 5
> sourcing on power up and the 2.5 Vref.
> Cheers,
> Andrew S

[ejeklint]

---

@stilly32 wrote:

Hey All

DAQmx 8.5 is available, you can download it here. This should resolve the issue with the 5 sourcing on power up and the 2.5 Vref.

Cheers,

Andrew S

---

Any hint on when an updated Mac OS X driver with this fix will be available?

/Per

Message Edited by ejeklint on 03-19-2007 03:20 AM

[ejeklint]

Bumping this thread a little.

Still no news about when/if a mac version with this bug squished will be available?

I'm beginning to believe that Mac OS X is no longer supported, but I hope I'm wrong...