Discussion:
How many currents can 655x sink?
(too old to reply)
AhGang
2008-04-01 09:10:07 UTC
Permalink
Hi,
 
    When PXI 655x receives data from DUT, how many currents can it sink?
 
    If we check the channel electronics, it seems that it can sink maximum current of 0.5mA (10kohm load with 5v maximum voltage).
 
&nbsp;&nbsp;&nbsp;If <a href="mailto:***@VOH" target="_blank">***@VOH</a>&nbsp;of DUT &gt; 0.5mA and VOH &lt;5v, what will happen?&nbsp;
&nbsp;
&nbsp;
&nbsp;&nbsp; Thanks.
&nbsp;
&nbsp;&nbsp;&nbsp;BR
&nbsp;&nbsp; Message Edited by AhGang on 04-01-2008 03:51 AMMessage Edited by AhGang on 04-01-2008 03:55 AM
Joe F.
2008-04-02 15:40:14 UTC
Permalink
Hi AhGang,The sink current for the 655x series can be calculated from the device specifications [1]:Max sink current = Vsource,max / (Rsource + Rterm) [2]It sounds like you've chosen the termination resistance to be 10kOhm for your application, so add your DUT's output impedance to that value and divide the maximum voltage you plan to send to the board by that total. The theoretical maximum sink current can be calculated by using the maximum input voltage for the board instead of the voltage you plan to send to the board.Please keep in mind that the boards have input voltage limits and while there are protection diodes for each channel [3], going overvoltage can still cause serious damage to the board.If you'd like more clarification, please let me know. In that case, please provide more information about your DUT's output characteristics: what voltage does it generate and what is its output impedance?[1] 655x Specifications<a href="http://digital.ni.com/manuals.nsf/websearch/E4C93B141B71ED93862573CC005E8EA1" target="_blank">http://digital.ni.com/manuals.nsf/websearch/E4C93B141B71ED93862573CC005E8EA1</a> [2] HSDIO Help: Input ImpedanceNI Digital Waveform Generator/Analyzer Help » Devices » HI 655x » Hardware Architecture » Channel Electronics » Input Impedance[3] HSDIO Help: Input ProtectionNI Digital Waveform Generator/Analyzer Help » Devices » HI 655x » Hardware Architecture » Input ProtectionMessage Edited by Joe F. on 04-02-2008 10:19 AM
AhGang
2008-04-03 07:40:22 UTC
Permalink
Hi Joe,
&nbsp;&nbsp;&nbsp; The DUT is characterize by output voltage VOH and output current IOH.
&nbsp;&nbsp;&nbsp; VOH = 2.4V and IOH =5mA. It means that the DUT will source 5mA current to HSDIO when it outputs 2.4v.
&nbsp;&nbsp;&nbsp; Then it 5mA * 10kohm = 50V &gt; 5.5v. So external circuit shall be added between DUT and HSDIO.
&nbsp;&nbsp;&nbsp;&nbsp; My solution is to add are resistor between the DUT's channel (also corresponding HSDIO channel) and GND.
&nbsp;&nbsp;&nbsp;&nbsp; The resistance can be calculated as : 2.4/(0.005-(2.4/10000)) = 500ohm.
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Does that make sense?
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Thanks.
Ryan M
2008-04-03 15:10:10 UTC
Permalink
AhGang, Generally, a DUT is characterized as either a current or a voltage source, but not both (assuming a constant load).&nbsp; That is, a DUT can force a constant current and let the voltage depend on the load, or drive a constant voltage and let the current depend on the load.&nbsp; V=IR, if we hold R constant and change V or I, then I or V (respectively) has to change as a result.&nbsp; In order to keep V or I constant, we have to change the R component which is usually not under our control.Most digital buffers have a VOH spec with an IOH "condition".&nbsp; When the DUT generates that voltage, there is a spec that says how accurate that voltage is under a certain load.&nbsp; The driver may not necessarily be driving that much current, just that it is capable of driving it.&nbsp; For example, if you look at the datasheet for a standard digital buffer (http://focus.ti.com/lit/ds/symlink/sn74lvch162244a.pdf), you'll see IOH specs and VCC "operating conditions".&nbsp; What this means is that with a VCC of 3V, for example, the buffer can provide source up to 12mA.&nbsp; However, if my load is 10k then it will only pull 300uA.&nbsp; If I understand your last post, are you trying to validate the VOH level while providing a known current load?With the drivers tristated, the input impedance of the 655x is 10k (or 50k depending on your device).&nbsp; If you drive a 2.4V signal, then the 655x will pull 2.4V / (10k + source impedance) = 240uA.&nbsp; Though you DUT is capable of driving 5mA at 2.4V, the 655x will only pull 240uA.&nbsp; Unless your DUT is a current mode driver (constant current instead of constant voltage) then your analysis is correct and the voltage will float to 10k * 5mA.&nbsp; If your DUT is voltage mode (constant voltage) then the DUT will drive 2.4V and the 655x will pull 240uA.If you want to force a load current, you can turn the drivers on.&nbsp; The output impedance of the drivers is nominally 50ohms.&nbsp; If you have to have 5mA of load at 2.4V, you can drive a static voltage of 2.4 - [5mA * (50ohm + DUT source impedance)].&nbsp; However, if you don't need to force the 5mA, just guarantee 2.4V switching, then save some power and leave the inputs in high impedance mode.
AhGang
2008-04-04 04:40:09 UTC
Permalink
Ryan,
&nbsp; Thanks.
&nbsp;&nbsp;My testing purpose is to draw as much as current and see if the DUT can keep voltage at least VOH.
&nbsp;Of course I can use SMU from NI. But it can only test one pin at a time. I am struggling to seek a method of using DIO to test it.
It seems impossible to test this by DIO due to lack of current source and diode bridge.
Is possible to add pin electronics such as programmble load/per pin PMU&nbsp; on DIO from NI?
BR
Ryan M
2008-04-04 13:10:08 UTC
Permalink
AhGang, Thanks for the clarification.&nbsp; The SMU is the best instrument to take that measurement.&nbsp; the 655x can take it but with much less accuracy.For example, if your DUT has 50ohms of output impedance, then you can drive a static voltage of 1.9V (static HIGH, VOH = 1.9V).&nbsp; This will force a 0.5V drop across the DUT's 50ohm and the 655x 50ohm (=5mA).&nbsp; You can then take your voltage measurements.&nbsp; If you take it at the pin of the DUT then you should see the 2.4V.&nbsp; If you take it at the pin of the 655x then you will see the voltage divider effect (2.15V).&nbsp; this would be the case if you use the acquisition engine to capture the inputs.&nbsp; You can take a voltage measurement at the pin of the 655x by sweeping VIH/VIL on the 655x and post process the state of the lines.&nbsp; When the state toggles from low to high, the VIH/VIL level has swept by the input voltage.&nbsp; The inputs are simple comparators where the reference voltage is your VIH and VIL setting.&nbsp; The 655x is flexible enough to allow you to generate the 1.9 V and sweep the input voltage at the same time.&nbsp; the caveat being that all channels will have the same voltages applied to them at the same time (if enabled).
AhGang
2008-04-07 02:10:09 UTC
Permalink
Hi Ryan,
&nbsp;
&nbsp;&nbsp;&nbsp; Good idea. Thanks.
&nbsp;
&nbsp;&nbsp;&nbsp;&nbsp;So the generation part&nbsp;of 655x (output) can sink almost 5mA current? The current flow into aquisition will be small because high resistance comparing with&nbsp;50ohm at generator.
&nbsp;
&nbsp;&nbsp;&nbsp; Due to different character of distinct DUTs, what is the maximun sink current of generation of 655x?
&nbsp;
&nbsp;&nbsp;&nbsp; BR
Ryan M
2008-04-07 14:10:10 UTC
Permalink
AhGang, The Generation specifications for the 655x call out +-50mA per channel, not to exceed 600mA for the entire board (page 3).&nbsp; If you use the trick I mention above then this will be your limit.&nbsp; If you are not using a driver to sink the load then the inputs can sink 5.5V / 10k = 550uA where 5.5V is the maximum acquisition voltage (before protection kicks in; page 4 of specs) and the 10k is the input impedance.&nbsp; If you force a current larger than 550uA, then the resulting voltage will be outside of the working voltage range.
AhGang
2008-04-08 01:40:06 UTC
Permalink
Thanks. I got it.
&nbsp;
BR

Continue reading on narkive:
Loading...