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.

A known bug in Clarius v1.5 burns devices - to update or not?

Post Reply
dndndn
Posts: 4
Joined: July 30th, 2018, 11:23 am
Country: Israel

A known bug in Clarius v1.5 burns devices - to update or not?

Post by dndndn » July 30th, 2018, 11:58 am

Hello,

We have a Keithley 4200A with Clarius version 1.4.1. We wish to update to Clarius 1.5 to use multi segment sweeps (issue numbers SCS-3730, SCS-3803 in the release notes https://www.tek.com/keithley-4200a-scs- ... ease-notes). However, in the list of known problems (see release notes), problem number SCS-3712 / AR60504 reads:
Subsystem: 4200A-CVIV
Symptom: Device damage can occur when SMU is executing a range change. This is due to an SMU loop not
being able to handle CVIV inductive load during one step in range change process. In such instance,
SMU output can slew to power supply rail. If interlock is present, output voltage can be as high as
210V for a short time during range change. This issue affects high current ranges of the SMU,
therefore it is present with or without Preamp.
Workaround: Use OVP (AVM) at a setting just above required test voltage. This will reduce likelihood of
DUT damage.
This seems like a very serious problem. I have a few questions regarding it, and I would appreciate answers to some or all of them:
  1. Was this problem introduced in v1.5 or was it there also in previous versions, just was not discovered? It is not listed in the release notes of v1.4.1 (http://download.tek.com/software/suppor ... V1.4.1.pdf), but maybe it was just not discovered.
  2. When is this serious problem expected to be fixed?
  3. Would changing the range between tests also cause this problem, or only when changing the range during a test?
  4. If this problem was not present in version 1.4.1, I think we would stay with our current version until it is fixed. If this problem was just recently discovered but was always there, we would update. Would you recommend updating from v1.4.1 to v1.5?
Thank you!

Andrea C
Keithley Applications
Keithley Applications
Posts: 1323
Joined: October 15th, 2010, 10:35 am
Country: United States
Contact:

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by Andrea C » July 31st, 2018, 8:43 am

What is the firmware version of your 4200A-CVIV switch accessory?

Brad O
Keithley
Keithley
Posts: 28
Joined: February 3rd, 2018, 10:43 am
Country: United States

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by Brad O » July 31st, 2018, 11:59 am

Thanks for posting, and we apologize if this has effected your workflow at all.
1. Was this problem introduced in v1.5 or was it there also in previous versions, just was not discovered?
The error is caused by a hardware issue in the 4200A-CVIV and so has been present in previous versions of Clarius. It was discovered shortly before the release of v1.5.
2. When is this serious problem expected to be fixed?
The problem has already been fixed in new 4200A-CVIV hardware. Existing units will have the fix applied when they are sent back for calibration.
3. Would changing the range between tests also cause this problem, or only when changing the range during a test?
This issue only occurred while the output was on, so you would not see this behavior outside of a test.
4. Would you recommend updating from v1.4.1 to v1.5?
We recommend updating to v1.5 and using the OVP setting workaround until you send your unit back for calibration, at which time you should be sure to send back your 4200A-CVIV as well. We know of only 1 customer who's seen this issue and using OVP prevented it in every test case we used.

dndndn
Posts: 4
Joined: July 30th, 2018, 11:23 am
Country: Israel

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by dndndn » July 31st, 2018, 2:45 pm

Thanks for your comprehensive replies, I appreciate it.
  • I assume the CVIV switch accessory is called "CVIV1" in the firmware upgrade window. Our firmware version is 1.04.
  • Should this or any of the other firmwares need upgrading in order to install Clarius 1.5, would such firmware change affect the likelihood of this problem happening?
  • We received our Keithley 4200A a few months ago. Is there a way to check whether our unit is already fixed, or you did not ship fixed units without Clarius 1.5?
  • We also have another instrument, a Keithley 4200 upgraded to 4200A, and I believe it does not have the CVIV switch accessory (but I am not sure). Would this problem affect an instrument without the CVIV switch accessory?
  • To make sure, the range change which causes this problem refers to a range change in the force section, when the SMU is sourcing either voltage or current, and does not refer to range changes in the sense section, correct?
  • This issue only occurred while the output was on, so you would not see this behavior outside of a test.
    When two subsequent tests have different ranges, the range change occurs while the output is off between the tests, so this problem is not affecting this scenario, correct?
Our firmwares' versions are as follows:
  • SMU1 to SMU5: M31
  • CVU1: 2.13
  • TUM1: 1.0.0
  • CVIV1: 1.04
Thank you!

Brad O
Keithley
Keithley
Posts: 28
Joined: February 3rd, 2018, 10:43 am
Country: United States

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by Brad O » August 1st, 2018, 8:14 am

[Would a] firmware change affect the likelihood of this problem happening?
A firmware update wouldn't be able to fix this particular issue, but of course you should always be running the latest firmware.
Is there a way to check whether our unit is already fixed, or you did not ship fixed units without Clarius 1.5?
4200A-CVIVs that have the fix will have a Hardware Version of 2.1 or later. You can find the hardware version listed directly below the firmware version in KCON. The fix has only been applied to units that shipped with Clarius v1.5.
Would this problem affect an instrument without the CVIV switch accessory?
Not at all, this problem would only appear when a SMU autoranges its output while going through a 4200A-CVIV while also connected to a complex device under test.
The range change which causes this problem refers to a range change in the force section, when the SMU is sourcing either voltage or current, and does not refer to range changes in the sense section, correct?
It might be possible that a current measurement autorange change would cause the problem, but we have never seen that happen. A voltage autorange change (either sourcing or measuring) would not cause the issue.
When two subsequent tests have different ranges, the range change occurs while the output is off between the tests, so this problem is not affecting this scenario, correct?
Correct, you would only see the problem while autoranging while the output is on.

For a little more background, we have only observed this issue with multiple SMUs autoranging at the same time while going through the CVIV while also connected to a specific, complex device, along with all the other requirements discussed. Even then, the problem only occurred during our testing 1/50 times or so. So even if your workflow meets all the prerequisites for this problem, it's likely you would never see the issue. We still recommend using the OVP setting until the hardware fix is applied to your unit though.

dndndn
Posts: 4
Joined: July 30th, 2018, 11:23 am
Country: Israel

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by dndndn » August 1st, 2018, 9:48 am

Thanks for your answers. If you do not mind, a few follow-up questions:
  • It might be possible that a current measurement autorange change would cause the problem, but we have never seen that happen. A voltage autorange change (either sourcing or measuring) would not cause the issue.
    To make sure I understand, the problem might seldom happen when the instrument changes source current range while in source current mode. It might also happen while the instrument changes current sense range while in source current, source voltage or any other mode.
  • Would using a "best fixed" range setting cause the range to change while the output is on?
  • I met today with a local representative and upgraded the software and firmware. He said that in order to keep the instrument operating while the CVIV switch accessory is shipped for the fix, it is possible to disconnect the preamplifiers from the CVIV switch accessory and to connect them to the main instrument. This way we would be able to use the SMU's just the same, but to switch to the CVU one would need to disconnect and connect cables. Correct?
  • Is it true that you do not need the preamplifiers to fix the CVIV switch accessory?
Thank you!

Brad O
Keithley
Keithley
Posts: 28
Joined: February 3rd, 2018, 10:43 am
Country: United States

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by Brad O » August 1st, 2018, 11:20 am

The problem might seldom happen when the instrument changes source current range while in source current mode. It might also happen while the instrument changes current sense range while in source current, source voltage or any other mode.
That's right.
Would using a "best fixed" range setting cause the range to change while the output is on?
No, so that is another effective way to avoid the problem, another would be to use any fixed range, e.g. 10nA. "Best Fixed" automatically selects a fixed range before the test starts based on either the values in the sweep (when sourcing) or the compliance level (when measuring).
It is possible to disconnect the preamplifiers from the CVIV switch accessory and to connect them to the main instrument.This way we would be able to use the SMU's just the same, but to switch to the CVU one would need to disconnect and connect cables. Correct?
Yes, the preamplifiers can even simply be removed from the 4200A-CVIV and used directly without any recabeling to the mainframe (after updating your configuration in KCON). I'd direct you to page 21 / section 3-7 of the CVIV User's Manual (https://www.tek.com/default-accessory-s ... s-manual-1) for specific instructions, but of course your local representative is the best resource since he's physically there. Without the CVIV switch, yes, you would need to manually change your cables to make measurements with the CVU.
Is it true that you do not need the preamplifiers to fix the CVIV switch accessory?
Correct, the hardware fix only affects the 4200A-CVIV switch itself, not any instruments, pass-through modules, or preamplifiers.

I'm glad you're in contact with your local representative, and I hope these answers help!

dndndn
Posts: 4
Joined: July 30th, 2018, 11:23 am
Country: Israel

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by dndndn » August 1st, 2018, 1:56 pm

Thank you for your thorough replies! Might I just say that the description of this bug in the release notes was lacking the details you provided above, or even a link for a more detailed description. I accept that you might not have known everything by the time Clarius v1.5 was released, a linked resource could be updated.
On a more general level, and although you only know of one customer who noticed it, I would have liked to be informed about this (by a push rather than pull), rather than serendipitously finding out there is a new version, and pedantically reading through its release notes.
Nevertheless, thanks for your responsive technical support on the forum!

Brad O
Keithley
Keithley
Posts: 28
Joined: February 3rd, 2018, 10:43 am
Country: United States

Re: A known bug in Clarius v1.5 burns devices - to update or not?

Post by Brad O » August 1st, 2018, 2:47 pm

Yes, it was a rather complex bug to discover and we didn't have all the details by the time v1.5 was released so the release notes were left somewhat vague. Unfortunately for this case, the release notes is a tightly controlled document that fulfills some legal requirements, so it's not very easy to update. But you're right that there should be some sort of update on this issue sine so many more details have been found. We are looking into adding an addendum on the Clarius v1.5 page.
We are starting to experiment with a sort of push notification system to the 4200A-SCS, so it's good to know that you'd like something like that! It certainly would be helpful in situations like this.
And thank you for posting your question! Hopefully the replies answer some other people's questions.

Post Reply

Return to “4200-SCS Semiconductor Parameter Analyzer”

Who is online

Users browsing this forum: No registered users and 1 guest