Ryzen Crosshair 6 Hero MB Killing 10Gbe Cards

Hardware Motherboards
1

I Know my post was long, I am sorry just trying to give as much details as possible, but for those who don’t feel like reading the whole thing, I would really like some ideas on this, so here is the short version.

My Crosshair Hero’s PCIEx8 Slot is Killing 10Gbe Cards Specifically. I can put any other cards, from other NIC, To USB Cards, To Graphic and Sound Cards in that slot, and there is NO PROBLEM. But this Specific 10Gbe Card I have now put 2 different ones in, and it has KILLED BOTH. Any Ideas as to why this would be happening?

More Details in the Post Below.

Ryzen 1600x
Crosshair 6 Hero
Corsair RM1000i PSU

2

I apologize that this turned into such a Long Post, I just want to make sure all the info you need is here.

I have a Ryzen 1600x and Crosshair 6 Hero MB. I am NOT Baseclock overclocking, because that has caused issues with my PCIE Cards in the past, so I am just using the Multiply to get me up to 3.97.

I recently Purchased Two 10Gbe Cards, specifically Dell XR997’s, to install in My Ryzen and “Server” to give me a 10Gb direct Link. I have used these specific cards to do the exact same thing for two different clients recently, and it worked great.

However, in this case, for what ever reason, the Card I put into my Ryzen System, upon starting up the PC, I immediately smell that bad Electronic Burning Smell, so I of course Turn Off The PSU, and Double Check everything is connected Properly. I reseated Everything, turned the PC Back on, and this time no Smell, everything boots properly, but the 10Gbe Card is not showing up at all in the system. Nothing in Device Manager, Nothing showing up in that slot in Aida64. This is under Windows 10, so I boot into my Windows 7 Disk, and Still nothing. I boot into Ubuntu, and Still nothing.

I had the thought that maybe it was an IRQ assignment issue, so I disabled the Onboard Lan, along with everything else not being used, Unfortunately this did not help. I decided to use my other PCIE x16 slots, both the bottom one which is x4/x1 and the Top Graphics Card one that is x16/x8 and it didn’t work in either, no matter what PCIE Lanes I select to use in the BIOS.

I Also should mention that the Fan on the 10Gbe does work, and I am getting a Link Light from the Card on my switch, so it is at the very least getting power.

The Other 10Gbe Card I purchased was put directly into my make shift “Server” (A Re-purposed old FX 8370). This card Worked Great and was recognized right away by Windows 10.

I Pulled the Bad card out of the Ryzen build and put it into my Server as a diagnostic step, and the card still was not acknowledged by the System.

So at this point I am thinking I just got one bad card. But I wanted to be certain that if I got a replacement it would work in my Ryzen Build, so I took the Working Card and Seated it in the x8 Slot in my Ryzen Build. Upon Starting the System, again I smelled that Burnt Electronics Smell. I went through the Same Process of reseating the card (even though I know it was already properly seated the first time). When I turned on the Computer again, it did not have the burn smell and booted just fine, but again the system didn’t acknowledge the card.

I Put the card back into my Server, and now the card is not being recognized at all by the Server Either even though it worked before putting it in Ryzen. There is now no doubt in my mind, that my Crosshair killed both of these cards. What I don’t understand is why.

I have Several 4 Port 1Gbe Nics, and a bunch of PCIE Sound Cards, RAID Cards, Video Cards and USB Cards that are just extra cards and I wouldn’t care much if They Died. I took the time to put EVERY SINGLE ONE of those cards into that same exact x8 Slot in my Crosshair, and EVERY SINGLE ONE OF THEM Worked just fine! No Burn smell, and they just work! Some of them are PCIE2, Some are PCIE1, and two are PCIE3, some are x1, some are x4, and a few are x8 and x16. They all work just fine.

So My Question is, why is my Mobo Only Killing these Specific 10Gbe Cards, but every other card I put in it work Fine?

Also I have a Corsair RM1000i Power Supply, it is Gold Rated and provides excellent power with very low Ripple. I have monitored through software my 12v, 5v, and 3.3v rails, and all are extremely tight and spot on at all times. So while I can’t rule it out completely, I tend to discount that its a PSU issue at this point. I am pretty sure it is down to the Motherboard, I am just wondering if anyone else has had really weird issues like this, where a specific card will be killed, but all other cards work great.

Sorry if this was too long, I just wanted to give all the information I have at this time. Thank You Guys for your help, you are ALWAYS ALL AWESOME!

3

It could still be a psu issue verywell.
I mean if other cards in the same slot work fine,
then i highlly doubt that the motherboard is the cullprit.
I dont have specific details of those said Dell Nic´s by hand,
But do those cards need a dedicated powerplug,
Or are those cards powered by the pci-e slot only?

1 Like
4

They are powered by the Motherboard Only. And they are actually the same Cards as a lot of other older 10Gbe Cards on the Market. Specifically they are the Intel EXPX9501AT Chipset. Thank You For Your Help.

5

Okay if they are powered by the board only, then the psu can be ruled out probablly yes.
Did you allready physically check the pci-e slot for any minor damages?

1 Like
6

I mean it would be helpful to know if it was the PSU, but I kinda hope its not, I just bought it in June, and Bought it specifically to not have power issues, it has such great praise from the Likes of Johnny Guru, so I hope I don’t have to replace it, but Like you said, it is completely possible. Just hoping maybe someone knows something about these Ryzen Boards that I might have over looked, or something else I may have overlooked.

7

I have the same thought. Just wonder what else it could be…

8

Yes, thoroughly, and I don’t see any, even got in there with a magnifying glass, I don’t see any signs of damage. I do however notice slight scorching on both of the now bad 10Gbe Cards, but only on the 4th Pin,

9

Its very faint though, possibly was already there so I can’t be certain. But it would be weird that BOTH cards have the SAME Behavior, and have the MB Not be the cause.

But at the same token, its incredibly weird how both of these cards have the same problem in this board, but literally EVERY OTHER CARD I have works fine no problems at all, and I have tested exactly 12 other cards, all of different varietys.

Its a weird one huh? Sorry for posting so much, I will slow down and wait for a response. I am anxious for other ideas. I do really appreciate having another brain on this.

10

Yeah well it could be that those said cards pull too much power from the slot.
Like the reference RX480 cards that went outside pci-e slot power specs,
for older motherboards drama from last year.
However if that was the case, then it would be more likelly that the board would die.
And motherboards nowdays shouldnt have those said issues anymore.

1 Like
11

Nothing wrong with providing as much info as possible.

1 Like
12

That is REALLY Interesting! I wonder if I can find some specs on how much power these cards pull. And then the other weird thing would be, does my FX Board have more tolerance for PCIE Power load then my Ryzen Board, that would just be funny. But I like where your head is at, I am going to do some research in that direction.

So far I have done a LOT of searching the net, and haven’t found one other person with my Exact issue.

13

You could try using HWiNFO to see if the board reports power figures.

1 Like
14

It does report Power Figures, I actually do use HWinfo, and everything on the Powerside seems in spec, will post a pic.

15

you mentioned scorching (on the 10gb3 cards), do you have a pic of that?

1 Like
16

Here is HWinfo. This is just now, so this isn’t with the card in the system, but I can put it in the system and get a screenshot of it, but this is how I monitor my system 24/7 and was one of the thoughts I had while the card was in the system, and the Power figures never drifted far from what you see in this screen shot. (Working on Getting Photo Of The Card for you right now, but like I said, it is very faint)

[Album] Imgur: The most awesome images on the Internet
17

Ok, did my best to get some decent pics of the card for you from different angles. Its very hard to see, but its in the very first finger, in the second to last contact, the only pin in that finger that has no gold contact. Also in one of the pics you can see what looks like slight scorching on the very Last Pin, in the last Finger, I just noticed it in that picture. Uploading the Pics now… (Sorry My Google Drive Wasn’t Letting Me Share, took a second)

[Album] Imgur: The most awesome images on the Internet
[Album] Imgur: The most awesome images on the Internet
[Album] Imgur: The most awesome images on the Internet
[Album] Imgur: The most awesome images on the Internet
18

Sorry, That took a second, Google Drive was being Dumb. But There you go. Its really hard to see, but its there, just ever so slightly, also if you smell it, it still has that burnt smell. I however can not smell the burnt smell in the PCIE Slot itself.

Also, I Just noticed, on the back of the card, there is no scorching, but, That first finger that has the missing gold contact on the front, its not missing on the back. I am wondering if this is some kind of issue. I unfortunately don’t have as deep of a background into electrical engineering as I’d like, or a first hand knowledge of the PCIE Spec.

19

not seeing anything on the power connector but that last one does looked messed up.

maybe is a combination of card/case/mobo positioning that is making these things go out of wack. maybe if you’re willing to risk another card try not securing it to the case or take it out all together.

1 Like
20

Yeah, I also had the same thought, So Right now my case is set up like an open test bed. And I took the bracket off of the Card all together because I thought the lip might not go deep enough to allow the card to seat properly all the way, so I took the Bracket off, and thats how I was so certain the second card was seated properly. Didn’t seem to make a difference as that second card I also lost. So I tried one with the Bracket, one without.

I unfortunately do not have another card to test, but I am wondering. The Slot that I put it in is Wired for x8, which is what this card is, but on Ryzen those Lanes go directly into the CPU. The Bottom Slot is a x4 and instead goes into the Chipset, I am wondering if that has anything to with it. Maybe these 10Gbe Ethernet cards are designed in such a way that there is some kind of power conflict when the Lanes are routed directly to the CPU?

If I had another card I would put it in the Bottom PCIE x4 Lane first to see if it makes difference…

Also I noticed that pin with the missing gold contact that looks like its slightly scorched to me, on the back there is no scorching at all, but the contact is not missing, is this an issue to have a missing contact in the front but not in the back?