Status Code: -235140 Internal Hardware Error occurred in Power_Supply hardware. Please contact National Instruments Support.

[TobiasG]

It would be incredibly useful to know the specific computer model and Thunderbolt chip used.

We have seen that many PC/motherboards have buggy Thunderbolt implementations, and there is a known issue in the Intel Maple Ridge Thunderbolt controller which I understand is currently being investigated.

If you continue to experience this type of issue, I encourage you to open a support ticket.

[cahyo_pri]

I'm using HP 800G9 PC, with 12th Gen Intel(R) Core(TM) i7-12700 2.10GHz.

There's no clear information about which Thunderbolt chip it actually has but the Thunderbolt Control Center application confirmed that it is a Thunderbolt 4 with 1 port..

[TobiasG]

Per https://knowledge.ni.com/KnowledgeArticleDetails?id=kA03q000000oxBkCAI&l=en-US,

---

Our Thunderbolt™ devices are primarily designed to work with third version Thunderbolt™ and/or ports that are built-in on the motherboard. Therefore, please be aware that all third-party PCI/PCIe add-ons are not officially supported and recommended by NI.

---

That said, there are some additional steps you can attempt:

1. Disable Kernel DMA Protection
2. Disable ASPM and USB selective suspension
3. In BIOS Power Options, disable USB Suspension
4. In BIOS Power Options, disable PCI Express Link

If none of those resolve the issue, you may need to switch to use a different host PC.

NI is working to improve the situation with Thunderbolt 4 controllers, but there are challenges due to differing implementations in BIOS and Thunderbolt controllers.

[CFRandall]

Tobias, 

We have been facing similar issues (the same error, in only some scenarios, with multiple cards in multiple slots, and using a Thunderbolt 4 PCIe card in the PC), and have had no luck after following your suggestions. Would there happen to be any updates regarding this topic in general since this thread was last active? Finding further information related to this error code has been difficult. 

Unlike the previous poster, however, we are on Windows 11. 

Thank you very much, 

Charles Randall 

[Yi_Fan]

Charlas,

I believe the TBT controller used on your add-in card is having device ID of 0x1134 or 0x1137. You can check it out in device manager under TBT controller properties. The issue Status Code: -235140 Internal Hardware Error only happens with these 2 specific discrete TBT controllers. Integrated TBT controllers and older TBT3 discrete controller have no issue so far. NI will have to work with Intel to figure out what is going wrong with these 2 TBT4 controller. Given that TBT is propriety interface and very limited info about the controller is being shared, it might take some time to root cause it. 

The advices from me would be 

• If you are using laptop, get a laptop with integrated TBT controller
• If you are using desktop, consider PCIe MXI like PCIe-8398 add-in card instead.

[CFRandall]

Yi, 

Thank you very much! The controller does indeed show up as 1137, unfortunately. 

We have laptops with integrated controllers in the other systems, and have had no issues with them, as you say; we will try to continue that practice for future systems. In the meantime, we have added an Abort and Initialize on either side of every troublesome LabVIEW VI, as a hotfix (based on the post by cahyo_pri), and have not yet had further issues. 

Thank you very much for all of the details and help, 

Charles 

[Blackclaws]

We're using one of the Gigabyte Maple Ridge add-on cards which as others have mentioned shows up as Thunderbolt Controller 1137 in Device Manager:

https://www.gigabyte.com/de/Motherboard/GC-MAPLE-RIDGE-rev-10#kf

It did come as a surprise to us that the PXI Chassis would have problems with Thunderbolt 4 links or specifically add-in cards as we just considered "Thunderbolt" to be similar to USB in that regard where the specific controller shouldn't play a role. It would be great if the issue could be resolved on the NI side of things, is there any sort of plan on when this might possibly be remedied?

[Yi_Fan]

The issue is kind of weird because same NI hardware and drivers with integrated TBT 4 controller has no issue at all. The issue only happens on Maple Ridge discrete Thunderbolt 4 controller, older Alpine Ridge discrete Thunderbolt controller has no issue at all. May need Intel to figure out what is going wrong but due to TBT IP is proprietary, it might take more time due to limited information available.

Thunderbolt is superset of USB4, it is not running USB protocol unfortunately even though they shared similar physical link. It used alternate mode of USB4. PCIe tunneling that NI hardware used is mandatory in Thunderbolt but optional in USB4.

[Blackclaws]

Some further info that maybe helps tracking down the root cause.

There might be more functions that are broken but for us we found one specific case that breaks.

The setting that causes this error for us specifically is this (where dcPower is an NIDCPower object, this is C#, curiously VoltageLimitAutorange is not affected):

dcPower.Outputs["0"].Source.Current.CurrentLevelAutorange         = DCPowerSourceCurrentLevelAutorange.On

Specifically any setting of the CurrentLevelRange whether automatic or manual breaks.