USB-6501 power-up state

[IS_Dynex]

Hello,

I have a USB-6501 that is being used to drive the input side of an opto-isolator (DIO->100 Ohm resistor->opto->Ground).

At power-on, the opto is turned on, despite the DIO lines supposedly defaulting to "high-impendance inputs" (quoting the spec sheet for the USB-6501) and appears to be sinking about 0.7mA from the digital input.

Is this because of the internal 4k7 pullup resistor, and if so why is this present on a "high-impedance input"?

Is my USB-6501 faulty? It works exactly as expected once configured by the software, it is just this power-on state that does not seems to match the specs.

Thanks,

Richard.

[RER]

The power-up state of the USB-6501 is always a high impedance input - this is powerup state can not be reconfigured for the 6501.  Each input is tied to a 4.7kohm pull-up resistor and the DAQ  board does not drive the signal high or low (so the inputs are essencially floating). Have you tried taking an multimeter to the digital lines to confirm this?

If it is necessary to set the power-up state of a digital line, other DIO boards do support that option.  For example, the power-up state of all m-series can be set in MAX as discussed in the following knowledge base article.

http://digital.ni.com/public.nsf/allkb/A98392F93E54D07786256E6F004DE4C3?OpenDocument

[IS_Dynex]

Hello Rich,

Thanks for your reply. I have tried measuring its power up state using a multimeter.

USB 5V is actually 4.7V

Pin1 (GND) to Pin14 (P2.2) voltage is 4.69V

Pin2 (5V) to Pin14 (P2.2) voltage is 16mV

Pin2 (5V) to Pin14 (P2.2) resistance is 3.22KOhm (note this is powered on so not truly accurate)

Pin14 (P2.2) to Pin1 (GND) short circuit current 1mA

Pin14 (P2.2) to Pin2 (5V) short circuit current 2uA

If this is intended, then I think describing the input as 'high impedance' is misleading (I'd expect 'high impedance' to be 1MOhm or more to each rail, and less than a few microamps to hold it to either GND or 5V). As for the 4.7KOhm pull-up, I would only expect that to be in-circuit when configured as an output.

I have tested a second USB-6501 which acts in the same way, so this does appear to be how it is designed.

I am aware of the m-series, but I don't actually need to program the power-up state. Defaulting to input would be fine, if only the input couldn't sink or source any appreciable current.

Thanks,

Richard.

[Ben_01880]

In the default unconfigured state does the 6501 "sink" or "source" thru the pin?  If the pin is

tied up with a 4.7K resistor to Vcc, I would expect that it would "source" some current. 

[paulmw]

I'd like to add that I agree that the documentation on the power up state of the USB-6501 is VERY misleading.  How can you state that it is high input impedance if there is a 4.7kOhm pull-up.  Unless NI thinks 4.7kOhm is high impedance

I was mislead and now I am concerned that the safety relay that this is driving will un-knownly switch on if the USB-6501 is unplugged and plugged back on.  I don't know if there is a simple solution because, even though the documentation states a "weak" pull up  (4.7kOhm is not that weak).  If I add a 1Kohm pull down the device doesn't drive.  Adding a 10k pull down doesn't over power the "weak" pull up

[Ben_01880]

What I did was to use a 74LS buffer that had a enable/disable function. Allocated one 6501 I/O pin as the enable for 

all the ports. When the board powers up it sees the pin "high' which disables outputs. Then I configure the board in software, write all zeros to all the pins except the pin used a the master disable. Then I finally write a zero to the master enable pin. This works quite well. I have done many power up/down cycles with no problems. It does cost you time and complexity for

what I feel was a flawed design.

[paulmw]

---

@Ben_01880 wrote:

What I did was to use a 74LS buffer that had a enable/disable function. Allocated one 6501 I/O pin as the enable for 

all the ports. When the board powers up it sees the pin "high' which disables outputs. Then I configure the board in software, write all zeros to all the pins except the pin used a the master disable. Then I finally write a zero to the master enable pin. This works quite well. I have done many power up/down cycles with no problems. It does cost you time and complexity for

what I feel was a flawed design.

---

Couldn't agree more!  Your work around is the same thing I was thinking.  Not sure if I have the time to implement it.

[boydie]

Why has someone not updated the spec yet? We just fell into the same problem.

[Tim_OP]

I think there has been a fundamental misunderstanding of what the "high impedance" means.  The DIO pin on the silicon, which is separate to the pull up resistor, is in a high impedance state and configured as an input.  The pull up resistor is there too, and the specifications show that it is always there (there is even a picture). That is so that it isn't floating.  It is a TTL level, not CMOS, so 4.7k is a reasonable value, otherwise the inputs would need a much larger pulse width to be seen.

It is a 69 quid device and has fixed pull up resistors, I imagine for the very reason that it is only 69 quid.  I wouldn't dream of using a device like this in something safely critical, there are others out there that have programmable power up states, and selectable pull up/down resistors.  Yes you pay for it, but that is what is needed, and you end up paying the nearly, if not the same by making your own circuit.