Corsair H115i Platinum doesn't show in HWiNFO64 6.02

Remanent

Active Member
I have a Corsair H115i RGB Platinum connected to the usb2 port on a Corsair Commander Pro, the sensors in the h115i don't show up at all inside HWiNFO64, but the sensors on the commander pro show, as do the sensors for the h850i power supply that is connected to the other usb port on the commander pro.

I have attached the report and debug files in the zip archive below.

USB configuration for h115i:

h115i to usb2 header on commander pro -----> Commander  pro ---> usb2 header on motherboard.
h850i to usb2 header on commander pro --------^

I tested both HWiNFO64 versions 6.02 and the beta build of 6.03 build 3680.  Same issue in both builds
 

Attachments

  • HWiNFO64-6.03-3680.zip
    338.6 KB · Views: 2
We do not have information about the H115i Platinum protocol yet, so don't know how to read its values.
The author of SIV will ask you for some reports, which will allow us to investigate this in more detail and hopefully determine how to support it.
 
It should be possible to add support for the H115i/H100i Platinum, but we need to figure out the protocol to use. I added initial support to my SIV utility (http://rh-software.com/) and if you can post what it reports then Martin and I may well be able to get it reporting quite quickly.

To do this using SIV 5.36 or SIV 5.37 Beta-20 (the beta has better Gigabyte Z390 M Gaming motherboard support) have a look at Menu->Tools->Link Status to see if the H115i is there. Either way do Menu->File->Save Local and post the generated files. I suspect a protocol trace should also help and to generate this from a command window do:

SIV64X -DBGAIO -SAVE=[link-status][link-dump][mutex][handle]=stdout > SIV_DBGOUT.log | more

and also post SIV_DBGOUT.log.

Update: Please use SIV 5.37 Beta-21 as I have added some more H115i Platinum code, do Menu->File->Download->SIV Beta to get it.
 
5.36 did show the cooler, albiet it showed as a h100i plat, updating to the 5.37 beta showed it as a h115i plat, but no data was reported.  i've run the file generation, and the protocol trace, and placed them all into a zip file attached to this post.  let me know if there's anything else you need from me.
 

Attachments

  • siv_reports.zip
    431.6 KB · Views: 1
Thank you, what you did was ideal. The H115i vs. H100i was down to a simple typo so easily fixed, but looking at the log it contains:

Code:
CID 12      @ 14:29:49.731 seq 02 Open( \\?\hid#vid_1b1c&pid_0c17#7&183a92cc&0&0000#{4d1e55b2-f16f-11cf-88cb-001111000030} )
CID 12      @ 14:29:49.732 seq 02 -> 00 02
CID 12        14:29:49.886 seq 02 <- 00 FF F8 10 21
CID 12      @ 14:29:49.888 seq 02 - Protocol Status 0xFF returned for command 0x02
CID 12      @ 14:29:49.889 seq 03 -> 00 06
CID 12      @ 14:29:50.241 seq 03 258 = ReadReply( 000000000D892E30, 65, 350 ) WAIT_TIMEOUT

Which tells me that the Platinum coolers use a different protocol to the CLCP which is what I was hoping they used, so adding support is going to be far trickier to the extent that I will need to have a [font=Tahoma,Verdana,Arial,Sans-Serif]Platinum [/font] cooler here so I can reverse engineer the protocol from a USB trace.

From my perspective the good news is that Beta-21 did a pretty good job of reporting your Gigabyte Z390 M Gaming motherboard sensors.

I will talk with Martin and decide what to do next.
 
So good and bad news. If there’s anything else I can provide to help you, short of buying you the cooler (lol) let me know.
 
Do you need me to rerun the reports and resubmit the results without incur and the Corsair services disabled?
 
Looks like we're making good progress in understanding the protocol ;)
We might have some test version soon...
 
Here's a copy of the reports without icue running, and the corsair services disabled.  Also, i noticed in SIV 5.37 beta 21 that there was an error in the link status page for the h115, one section is still reporting the h100i, i grabbed a screen cap of the window, and have it included here.

siv_link_status.png
 

Attachments

  • siv_reports-noicue.zip
    429.1 KB · Views: 1
Remanent said:
Do you need me to rerun the reports and resubmit the results without incur and the Corsair services disabled?

No, but thank you for doing this anyway. It was more to let you know that if you do this things will be slightly wonky.

I have already fixed the name, it was an error in usbdevs.txt.

What does iCUE report as the H115i Platinum firmware version please?
 
red-ray said:
Remanent said:
Do you need me to rerun the reports and resubmit the results without incur and the Corsair services disabled?

No, but thank you for doing this anyway. It was more to let you know that if you do this things will be slightly wonky.

I have already fixed the name, it was an error in usbdevs.txt.

What does iCUE report as the H115i Platinum firmware version please?

V. 1.00.33
 
I have added some code so please will you try Beta-22 and generate a new protocol trace and save files?

Adding a new protocol without having test hardware here is tricky, so I suspect it may take a few attempts before it's 100%.

I would like to check the H115i Platinum USB descriptors, but it's in power saving mode so [USB Bus] reports Status 31 A device attached to the system is not functioning so will you use [Link Devices] to disable EnhancedPowerManagementEnabled by unchecking it and then pressing [Restart] for both the H115i Platinum + HX850i before doing the save?

attachment.php
 

Attachments

  • [Link Devices].png
    [Link Devices].png
    67.2 KB · Views: 35
Here's the reports, and a screenshot of the link_status screen.  I get that this make take some (or a lot) of trial and error with not having the hardware physically on hand, but I don't mind helping at all.  I'll try to reply as soon as I'm able, but I also have a family and job (I was off today, but work the next two days 6am-2:30pm EDT).

siv_link_status2.png


so on the h115i


the fan rpms aren't in the correct order.

temp1 and fan 2 are the two fan rpms, and fan 3 is the pump rpm.
temp for temp1 appears correct.
 

Attachments

  • siv_reports.zip
    434.1 KB · Views: 1
Thank you for testing Beta-22 and the fan RPMs look to be correct to me. There is only room for one name, so SIV uses the name of the left most item. Look at [Status] or [Link Fans] and the names will make more sense. You can also press [ H115iPt ] to see all the names and can choose you own names. The [Link Status] panel is mainly there so I can check things are getting reported correctly and there are no transfer errors or timeouts.

Without the hardware it's quite frustrating as if I make a small error it takes a disproportionate time to fix it, but with the H115iPt so far I have done OK. In general getting reporting to work is the easy part, control is much harder and I will need a [font=Tahoma,Verdana,Arial,Sans-Serif]H100iPt here before I even try. Another tricky aspect is recovering from protocol errors.[/font]

I have made some more changes and released Beta-23 which should change the default name from Fan 2 to Pump and report the firmware version.

As things are basically working we now need to see how well so from a command window start SIV using:

siv64x -dbgaio | find "H115iPt" > SIV_DBGOUT.log | more

This will just save the [font=Tahoma,Verdana,Arial,Sans-Serif]H115iPt[/font] trace information. Leave SIV running for 5 minutes then generate the save files and exit SIV.

After doing this if HWiNFO looks to also be working it would be good to run both utilities and check "[font=Tahoma,Verdana,Arial,Sans-Serif]they [/font]play nicely together", they should. If you watch Menu->Help->Lock Handle you should see both programs acquiring the locks from time to time. Also check [Link Status] for any error reports, hover the mouse over Corsair H115i Platinum USB Cooler Device (#1) to see the detailed status information.

I just noticed there were about 200 instances of C:\Program Files (x86)\EVGA\Precision XOC\PrecisionXServer.exe active on your system :s , which is 199 more than I would expect.
 
red-ray said:
Thank you for testing Beta-22 and the fan RPMs look to be correct to me. There is only room for one name, so SIV uses the name of the left most item. Look at [Status] or [Link Fans] and the names will make more sense. You can also press [ H115iPt ] to see all the names and can choose you own names. The [Link Status] panel is mainly there so I can check things are getting reported correctly and there are no transfer errors or timeouts.

Without the hardware it's quite frustrating as if I make a small error it takes a disproportionate time to fix it, but with the H115iPt so far I have done OK. In general getting reporting to work is the easy part, control is much harder and I will need a [font=Tahoma,Verdana,Arial,Sans-Serif]H100iPt here before I even try. Another tricky aspect is recovering from protocol errors.[/font]

I have made some more changes and released Beta-23 which should change the default name from Fan 2 to Pump and report the firmware version.

As things are basically working we now need to see how well so from a command window start SIV using:

siv64x -dbgaio | find "H115iPt" > SIV_DBGOUT.log | more

This will just save the [font=Tahoma,Verdana,Arial,Sans-Serif]H115iPt[/font] trace information. Leave SIV running for 5 minutes then generate the save files and exit SIV.

After doing this if HWiNFO looks to also be working it would be good to run both utilities and check "[font=Tahoma,Verdana,Arial,Sans-Serif]they [/font]play nicely together", they should. If you watch Menu->Help->Lock Handle you should see both programs acquiring the locks from time to time. Also check [Link Status] for any error reports, hover the mouse over Corsair H115i Platinum USB Cooler Device (#1) to see the detailed status information.

I just noticed there were about 200 instances of C:\Program Files (x86)\EVGA\Precision XOC\PrecisionXServer.exe active on your system :s , which is 199 more than I would expect.

Just got home, I'll get at this in a moment.

As far as the evga program, i saw that as well, shut down the main program and the dozens of servers shut down.  I have no idea why that happened, and haven't seen that before.
 
Here are the reports, and a screenshot of the lock handles.

[attachment=3293]

HWiNFO showed the pump, and its' data.
SIV showed the firmware, and the fan relabeled as a pump.
 

Attachments

  • siv-reports.zip
    417.8 KB · Views: 1
  • hwi64_603_3687-reports.zip
    416.1 KB · Views: 2
  • siv_lock_handles.png
    siv_lock_handles.png
    53.5 KB · Views: 4
Thanks for the feedback !
Looks good to me in HWiNFO, doesn't it? Let me know if you see any issues.
 
I didn’t have a lot of time to look around, just to verify it showed up in the sensors list, I was receiving wife agro (lol) and had someplace to be, but didn’t want to keep the two of you waiting for the reports. I will experiment with it more this evening. It’s looking good so far.

Thank you to the two of you for how quickly you’ve been able to implement fixes / workarounds within such a short amount of time.
 
Back
Top