Tektronix Technical Forums are maintained by community involvement. Feel free to post questions or respond to questions by other members. Should you require a time-sensitive answer, please contact your local Tektronix support center here.

2450 SMU - periodic glitching when sourcing voltage to resistor

Models 2450, 2460, 2461, 2470
Post Reply
bwilkins
Posts: 6
Joined: April 16th, 2021, 8:05 pm
Country: United States

2450 SMU - periodic glitching when sourcing voltage to resistor

Post by bwilkins » April 16th, 2021, 8:39 pm

Hello,

I recently started using a 2450 source meter. I've encountered some strange yet consistent results from a basic test and hope someone here can help me.

Test setup:
- The 2450 is set to source 1.1mA through a ~100ohm resistor, and measure voltage.

Meanwhile I have a digitizing multimeter (50kHz sample rate) measuring the voltage across the resistor. Based on Ohms law I expect ~110mV (it is not a precision resistor) + random noise.

The multimeter data (see plot below) shows ~110mV + noise as expected, but also a burst of high-low glitches occurring at 500ms intervals.
- The glitches alternate high then low at exactly 0.5ms intervals.
- Each burst of glitches lasts 16ms.
- There is a 3.5ms delay between each burst of glitches.
- There are 3 such bursts of glitches.
- The pattern repeats precisely every 500ms.

2450_glitches.png
2450_glitches.png (447.79 KiB) Viewed 587 times

Each plot zooms in on the horizontal axis, which is time in milliseconds.

Everything is good except for these glitches, which you can see occur in bursts of 3 groups, and is periodic every 500ms.


While trying to find where this interference is coming from, I found if I set the SMU output to OFF the glitches are still there, however if I turn the SMU itself off, the glitches are gone! Also if I disconnect the cable from the SMU the glitches are gone. So, somehow these glitches seem related to the SMU itself.

The glitches look like it could be a relatively low 1kHz signal that is "clocking" 16-bits x 3 groups, every 500ms. Could it be related to something happening internally of the SMU near the front panel or a sampling circuit?

I wonder if I have something wrong with my setup, or using the instrument incorrectly, but this seems to be an elementary test.
- I've tried sourcing voltage, measuring current and setting a 1.1mA limit and the problem is also observed.

If anyone has ideas what I could do differently or test, I am most willing to try.

Thank you!

bwilkins
Posts: 6
Joined: April 16th, 2021, 8:05 pm
Country: United States

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by bwilkins » April 17th, 2021, 6:19 pm

Addendum: To show this more directly from the setup, and that the digitizing multimeter is itself not the problem, I've tested with both Keysight 34465A and Keithley DMM7510 and took photos.

- The 2450 SMU is sourcing 0.5mA in this case, so the voltage across the resistor is nominally 50mV.

keithley.jpg
keithley.jpg (621.38 KiB) Viewed 578 times
keysight.jpg
keysight.jpg (699.5 KiB) Viewed 578 times

Andrea C
Keithley Applications
Keithley Applications
Posts: 1565
Joined: October 15th, 2010, 10:35 am
Country: United States
Contact:

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by Andrea C » April 20th, 2021, 5:20 pm

We are looking at this.

bwilkins
Posts: 6
Joined: April 16th, 2021, 8:05 pm
Country: United States

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by bwilkins » April 20th, 2021, 5:33 pm

Andrea C wrote:
April 20th, 2021, 5:20 pm
We are looking at this.
Thanks for the reply. Just want to clarify whether your comment means this is a known issue you can reproduce, or this is the first report and you're looking into this as a result?

bwilkins
Posts: 6
Joined: April 16th, 2021, 8:05 pm
Country: United States

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by bwilkins » April 23rd, 2021, 11:11 pm

I did another test:

1. Keithley 2450 SMU source 0.95mA into the 100 ohm resistor.
2. Keithley DMM7510 measures current in series with resistor (digitize mode; 50kHz sample rate) (I connected the +/- wires such that it's showing negative current below.)
3. Keysight 34465A is measuring voltage across the resistor (digitize mode; 50kHz sample rate)

Here are photos from the instrument setups showing the same "spikes" as earlier:

smu.jpg
smu.jpg (556.49 KiB) Viewed 533 times
current.jpg
current.jpg (630.33 KiB) Viewed 533 times
voltage.jpg
voltage.jpg (853.53 KiB) Viewed 533 times

One of the outcomes I found from this test is that the 2450 Source I RANGE setting (Auto, 1mA, 10mA, 100mA, etc.) has an effect on the severity of the spike artifact.

Andrea C
Keithley Applications
Keithley Applications
Posts: 1565
Joined: October 15th, 2010, 10:35 am
Country: United States
Contact:

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by Andrea C » April 26th, 2021, 8:19 am

The periodic nature of it was curious to me. I checked with the factory/design engineering. The periodic nature of the increased source current is due to some internal sensor monitoring that the instrument performs. We must have a little internal coupling into the signal path. They were not alarmed by the observation.

They reminded me to keep in mind: our performance specs are all defined at 1NPLC settings using an integrating A/D converter.
Using a relatively wide bandwidth tool such as the 1MHz digitize feature of DMM7510 may show signal characteristics that are well above the effective bandwidth of any IV data the 2450 will return.

To illustrate, compare the DMM7510 reported current when the DMM7510 is set for DCI using the integrating A/D at 1 NPLC.
Using integrating A/D at 1NPLC on either the 2450 or the DMM7510, both will show current values that are well within the allowed sourcing error of the 10mA range of the 2450.
2450_10mA_SrcCurrent_measured_with_DMM7510_1NPLC.PNG
measure at 1NPLC setting
2450_10mA_SrcCurrent_measured_with_DMM7510_1NPLC.PNG (56.66 KiB) Viewed 523 times

bwilkins
Posts: 6
Joined: April 16th, 2021, 8:05 pm
Country: United States

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by bwilkins » April 28th, 2021, 9:18 am

Hi Andrea, thanks for investigating this issue with the factory/design engineering. I appreciate that and the follow-up here.

I understand that measuring with 1NPLC averaging would yield values within the product specification. However at least in terms of average value, that would also be the case even if the glitching were continuous. (Any periodic signal -- sine wave, glitches, etc. -- that fits an integer number of times within 1NPLC and having an average value over one period matching what is desired, would overall satisfy the spec in terms of the average value.)

Considering your comment "The periodic nature of it was curious to me." it seems you are equally surprised at the finding, and probably can appreciate from a users point of view that this is not anticipated. The average value is great -- no problem at all. On the other hand the glitches are unexpected and easily confound tests and measurements.

I think it's worth highlighting two things:
- The glitches far exceed surrounding levels of noise.
- The period of glitches is 1kHz. This is significantly lower "bandwidth" than the 1MHz digitize feature of DMM7510. The results I posted above were at a 50kHz sample rate. I expect the glitches will be easily observable at 10kHz or even 5kHz.


Where to from here? Is Keithley design/engineering able to consider solutions to reduce the artifact? Is it possible to resolve by FW update at all?

Andrea C
Keithley Applications
Keithley Applications
Posts: 1565
Joined: October 15th, 2010, 10:35 am
Country: United States
Contact:

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by Andrea C » May 12th, 2021, 8:32 am

I did/do share a bit of surprise that the sourced current has the periodic signature.

The 2450 design and performance specs call for use of an integrating A/D at 1 NPLC. This results in a lot of signal averaging and low effective bandwidth relative to the faster tools such as the digitizer in the DMM7510.

Operating at the 1NPLC conditions, I think you will find that the reported V for the DUT that is excited by 1.1mA from the 10mA source range is unaffected by the periodic artifact. When taking this intended use case into consideration, we believe the product is fully meeting it's performance spec.

In my last post, I used the 1NPLC setting to measure the 1.1mA source current and show the stats page. In more than 32K readings, the standard deviation was only 8nA whereas the source spec for the 10mA range allows for a lot more.

Do you have a device for which the IV data reported by the 2450 is not correct because of the observed artifact?

bwilkins
Posts: 6
Joined: April 16th, 2021, 8:05 pm
Country: United States

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by bwilkins » May 23rd, 2021, 7:02 pm

Andrea C wrote:
May 12th, 2021, 8:32 am
I did/do share a bit of surprise that the sourced current has the periodic signature.

<snip>

Do you have a device for which the IV data reported by the 2450 is not correct because of the observed artifact?
Hi Andrea, thank you for the reply. I do appreciate that the 2450 design and specs are at 1 NPLC and under those conditions the instrument is within spec.

For your last question, yes. My situation is I have a DUT that I source a constant current, say 1mA, continuously. My DUT has a sensing element subject to other physical phenomena that assumes the input current is constant+noise. The output of the sensor is a voltage that varies as it is subject to the other phenomena, and superimposed on that signal is the periodic glitching artifact of the 2450.

When analyzing the measured voltage data I would have to average all of the data in order to "smooth out" the glitching, but that means I'm lowering the effective bandwidth of my sensor, which is very undesirable. Other options could be to "remove" the glitches, or gate the recording data, or "cut out" the affected samples in post-processing. These work around seem like they may not be robust. For example, changing the range (1uA, 1mA, etc.) has an impact on the artifact.

What would be really great is if we can join forces and come up with a work around -- perhaps a small inductor at the output -- that can eliminate these current glitches from the source. I would be really grateful of such support.

JethTran
Posts: 1
Joined: May 26th, 2021, 11:28 pm
Country: United States

Re: 2450 SMU - periodic glitching when sourcing voltage to resistor

Post by JethTran » May 26th, 2021, 11:53 pm

Hi...Have you perused the manual for this gadget? Do you realize how to control it utilizing LabVIEW code? Have you any experience composing LabVIEW code? [If along these lines, show us the code you have written to produce the flows you need - we can get a smart thought where you may require help].

Accepting you have a current source that you can handle, how would you intend to transform this current into a voltage that you can quantify? How might you get this voltage (utilizing LabVIEW)? Incorporate a depiction of the equipment you will use for the obtaining and an illustration of the code that you expectation will allow you to make the estimations and chart normal voltage each second.

Do these different strides in isolated bits of code. When each piece is working accurately, consider how to join them into a solitary program.

Post Reply

Return to “2450/2460/2461/2470 Touchscreen SourceMeter”

Who is online

Users browsing this forum: No registered users and 2 guests