What is the maximum power available at connector for PXI-6515?

[thisisnotadream]

Trying to power two MAX490 chips using the +5VDC output on the connector of a PXI-6515.  Seems to be OK for a few seconds (regulating 28VDC to 5VDC in the card), maybe upto about a minute, then I get a 67ns sine wave imposed on the line.  Are the chips drawing too much current for the card?  I cannot seem to find the specification for this line in the documentation for the card.  MAX claims their chips draw no more than 1mA.

Dan

[S_Hong]

Hi Dan,

I would like to ask you a couple of questions which will give me a better idea of what might be going on.

Is there anything that is changing in your circuit such as a change in input signal when the sine wave starts to occur?

How large is this sine wave and does it continue indefinitely? What is its approximate frequency?

Does the sine wave appear consistently after about a minute or running your hardware?

Could you tell me exactly how you have your hardware connected on the pins of port 6 & port 7? (I am assuming you are using one of the pins on these ports as your 5V source)

Regards,

[thisisnotadream]

Thanks for following up!

---

S_Hong wrote:

Hi Dan,

I would like to ask you a couple of questions which will give me a better idea of what might be going on.

Is there anything that is changing in your circuit such as a change in input signal when the sine wave starts to occur?

The oscillations begin about a minute (using my best guestimate...I never pulled out my watch) after power up.  However, I can get immediate results by attempting to transmit data through the transceiver at any time prior to the first oscillation.  Also, the power up isn't the cleanest...ie--I can see some ringing temporarily on the scope in the first second of power up.  

How large is this sine wave and does it continue indefinitely? What is its approximate frequency?

The oscillation is not a pure sine wave (sort of a small mountain-big mountain pattern), though it is very periodic and repeating at about a 67ns wavelength from a low of about 3.5V to a high of 7.5V.  It continues until power down, and I have found no way to reset without removing power.

Does the sine wave appear consistently after about a minute or running your hardware?

As described above, I'm not sure how consistently the first unforced oscillations occur, though it doesn't happen immediately and we never have to wait longer than about 1 minute.

Could you tell me exactly how you have your hardware connected on the pins of port 6 & port 7? (I am assuming you are using one of the pins on these ports as your 5V source)

We have proprietary hardware powered by 28VDC supply.  One of the functions of the hardware is to provide 28VDC to electromechanical sensors.  We have tied one of these unswitched 28VDC lines to the VCC of ports 6 and 7.  (I checked the supply voltage which the 5VDC line was oscillating and found it to be rock solid.)  We have +5 of port 6 connected to the VCC of one chip and the +5 of port 7 connected to the VCC of the other.  Do you need all other connections to the chips?  I'd have to have the lab look that up.

Regards,

---

We have moved on to using a 5VDC regulator chip to power the MAX transceivers and are getting better results.  This isn't the first time, though that one of our hardware guys has eyed those +5VDC lines lustily and asked, "Hey, can't I just power this chip with those?"  I have usually been able to dissuade them, though.

Dan

Message Edited by thisisnotadream on 01-13-2009 04:25 PM

[S_Hong]

Hi Dan,

This does seem like a strange behavior. I have discussed about this issue with R&D and it seems like the voltage regulator might be going unstable simply due to the PXI-6515 large amplitude of the observed oscillations. It is possible that it takes some time to reach the point of instability which might then cause the oscillations in the 5V line. Are you connecting anything else on ports 6 & 7 aside from the VCC and +5V lines? Have you tried measuring the VCC line to see if you were still getting a constant 28V when the oscillations occur? Do you also get the same behavior when powering other 5V devices?

[thisisnotadream]

Stupid message board erased my original wonderfully-crafted response.  I don't feel like rewriting it exactly, so excuse my brevity and do not take it for brusqueness.

We have taken the card out of the loop.  We are now relying on a 5VDC regulator chip.  Still seeing problems.  5V steady, then bursts of noise, then 5V steady. 

Answers to questions:  Nothing else on ports 6 and 7 of that card (it's mostly a spare), though 28-volt line is also tied to output ports on another PXI-6515, which normally provides a number of discrete outputs.  However, during hardware testing these lines are unloaded and uncommanded.  28V appeared constant during noise (though in our current environment, we did notice a 1-volt oscillation at same frequency, still investigating supply).  No other devices applied to 5-volt line.

What is the stated current output spec for 5-volt lines?  It isn't listed on the documentation.

[S_Hong]

Hi Dan,

Perhaps we can try using another 28V power supply and see if you still get the behavior. It's also possible that since you said you had the 28Vsupply tied to the VCC of multiple output ports on multiple 6515s, the current requirements from might exceed specifications. Can you try havingonly a single VCC connection and see if this alleviates the problem?

You are correct that the current output specifications for the 5 volt line is not listed in the document. I will take a look into this and see if I can getyou an answer.

[RavensFan]

---

thisisnotadream wrote:

Stupid message board erased my original wonderfully-crafted response.  I don't feel like rewriting it exactly, so excuse my brevity and do not take it for brusqueness.

---

Have you ever noticed it is only the long messages that seem to magically disappear?

[S_Hong]

Hi Dan,

After looking further into the issue, the maximum current which can be drawn form the +5V line will depend on many factors such as ambient temperature, input/output voltage and channel count. Assuming the worst case scenario, with all the DI, DO channels running at 30V, the worst case spec for the +5V line current is rated at 15mA.

Regards, 

[thisisnotadream]

Thanks for checking that out for me.  We tried it with another power supply...with the same results.  We tried breadboarding the circuit, without any connection to anything else, just the chips, load resistors, and supplies (at this point we are using an external regulator chip).  No disturbances were reported.  Before we left for the long weekend we were testing and found that the noise had gone away!

(A little background:  Normally, the COM ports would provide 422 traffic to the MAX chips, which would convert them to TTL-like signals.  These signals would be input to a PXI-7831R.  This FPGA card would also provide TTL outputs which would be converted by the other side of the MAX chips and sent to the COM ports.  We are attempting to emulate SSI encoders.  During test, we disconnect the FPGA card and provide a loopback line to verify the wiring and the other circuitry.  It should appear to the tester that the COM port receives what it transmits.)

Well, we were all excited that the noise went away, so we tried our little loopback test.  The COM port is transmitting fine, but the output at the MAX chips didn't seem to change.  Couldn't figure it out until I noticed that someone had plugged back in the PXI-7831R.  As soon as I unplugged it from the SCB, all hell broke loose on those MAX chips' VCC lines.  At that point I uttered an expletive, said some things that I am not proud of, and left.  Let's just say I'm glad the snow didn't keep me from church on Sunday.  There were some things I needed to get off my chest.

Dan