New AMD Ryzen Build Failing to POST. Any thoughts?

B

BeerEngineer

n00b
Joined
Feb 22, 2017
Messages
55
AMD
1700x Asus CrossHair IV Hero
Corsair Dominator 16gb 3200mhz (on QVL)
Corsair hx850
Ekwb p280
Gigabyte gtx 1080
My digital sad 480gb

When psu is turned on, the lights on the motherboard illuminate.
The CPU ready light is green.
No q or error codes displayed.
Trying to power on the system by case, start button directly in mobo, and by shorting the power btn header fails to spin up fans or turn on the water pump.

Power supply is from a known good system.
Removing the atx cable and using a psu bypass atx connection turns on the power supply.

I tried replacing my gtx1080 with a known good gtx970.

I've disconnect all of the peripherals.

Verified a pwm connection is plugged into the CPU fan header.

Verified the 8 pin atx CPU fan is plugged in.

Any quick hit items that you all can share with me?

Otherwise I think I have to go though the pain in the butt of setting up the motherboard outside of the case to rule out shorts and then RMA.

That's a huge pain. Hopefully it doesn't come to that.

Thanks in advance for any thoughts you can share.
 
Does the BIOS flashback work? I would try following the directions and updating to the latest .CAP file and see if it helps.
 
Looking online, it appears that bios flashback is supported on this mobo. Since it refuses to even power up the power supply on hitting the power button, its anybodies guess on if trying to flash the bios is possible.

I'll try that.
 
I had a similar issue with my build, turns out I had a bad case reset switch, it was grounding out on the case and caused the MOBD to basically freeze, what I would do is unplug any IO that you have plug in and try to post, and if still nothing try pulling the MOBD out there could be a stand off that is touching the back of the MOBD that you did not see when mounting, (I had that happen when building a athlon system back in the day) if anything else you will eliminate the possibility of a bad IO or a loose power supply connection,

EDIT: forgot to mention if you do pull the MOBD out try to get it to post on top of the box, with only the CPU, RAM, PSU and Vid card connected you can basically eliminate everything else either way, as in my case I did pull my MOBD out adn then worked my way back until i found the bad reset switch
 
Krato basically conveyed the exact sentiments I was going to: Your system has a ground short somewhere in it. Though, I hadn't considered a reset switch. I was actually going to say likely a bad motherboard. All it takes is one single itty-bitty component to be faulty, like a transistor or a resistor, or even a glob of solder that somehow went astray and bridged another pad... and your system will do exactly that, because the PSU is sensing a ground fault.
 
You stated the PSU is from a known working system and seems to power up when testing, but have you tried another PSU anyway?

I've seen PSUs test just fine right up until you put a load on it.

I know I'm SWAG'ing but I've learned the hard way always test the *EASIEST* things first, not the most likely.
 
Disconnecting all switches and trying to use the start switch on mobo failed.

Disconnecting all fans, USB, audio etc also didn't remedy the issue. pulling the board to rule out ground fault is going to be a pain in the rear because I have a custom water loop. This would be far less of a pain if I bought the noctua d15. Ugh!
 
rtangwai said:
You stated the PSU is from a known working system and seems to power up when testing, but have you tried another PSU anyway?

I've seen PSUs test just fine right up until you put a load on it.

I know I'm SWAG'ing but I've learned the hard way always test the *EASIEST* things first, not the most likely.
Click to expand...
I pulled the psu from my trusty 2500k system. It's never been flakey. Very solid performer.
 
the CPU mobo power pins plugged in? I have seen failure to plug these in will allow the system to boot but not post bios.
 
When I unplug the CPU 8pin, the motherboard identified they were missing and displayed an led next to them. When I reseated the cable the light went off.

I've done a fair bit of troubleshooting. The two identified next steps are basically to pull the motherboard and attempt to post on cardboard and to flash the bios using a USB drive.
 
Cleared CMOS? Some people had to clear CMOS multiple times or jumpered for a long time to get a default reset.
 
BeerEngineer said:
Looking online, it appears that bios flashback is supported on this mobo. Since it refuses to even power up the power supply on hitting the power button, its anybodies guess on if trying to flash the bios is possible.

I'll try that.
Click to expand...

Flashback works without the CPU even installed. Last resort though. I agree with the reseating CPU option, my Ryzen chip "hangs up" on the socket and I have to gently push some so it pops in. Doesn't drop right down like the old AMD chips did.
 
Possibly, but if that's the case.. the green led that says CPU ready from asus is poorly implemented.
 
I have not read through this whole thread, but here is what I do with the Crosshair VI I have worked with.

Turn off PSU. Click power on button on mobo until PSU caps are drained (no lights on mobo). Press and hold clear CMOS button on IO panel for 10 seconds. Turn on PSU. Start motherboard. Go into UEFI. Set Optimized Defaults. F10 and save and reset. See if that gets you back to POST/OS.
 
I'm curious, When people have had to clear cmos, possibly even multiple times on a new board, are they seeing the same behavior with lights on the unit, but no response to the power button? in my past experiences if a board has an unstable bios configuration, powering up the system results in power supply and case fans turning on.

I disassembled the system in preparation for the next block of time I have to work on this project.

To do List:
1. Flash new bios
2. Attempt to clear cmos
3. Remove water block, and pull board looking for mobo mount standoffs. I used the default standoff configuration for Phantek Evolv AtX if anyone happens to remember if they had to remove standoffs?
4. If none of this works I'll have to attempt to boot from the board sitting on a cardboard box. This might be tough to get my water block to reach. I don't have a lot of extra tubing length. I may need to order an inexpensive air cooler for testing. While I'm swapping, I would pull the CPU and reseat it. If I'm still having trouble, rma the board.


Thanks everyone for the suggestions. It's not easy to find time for this hobby with a job at a software startup, an infant, 1 year old, and 3 year old.
 
The system would still not turn on if it is the ram right ? Maybe try one stick at a time (both)
 
BeerEngineer, dam dude that's alot going on, hey if you have a old air CPU cooler laying around you might be able to use some elastic bands (I know this sounds dumb but hear me out ) to temporary hold a cooler down on the CPU so you can post at the very least, this might be better then trying to re run your custom water loop over to the cardboard box with the MOBD on top as that can get messy if you are pulling everything over to work, even just having the block on the CPU with water in it just for a post and as long as you shut it off right away you should be good, just do't let it run till steam rises out of the open water block if your going to do it that way :p
 
Kato1144 said:
BeerEngineer, dam dude that's alot going on, hey if you have a old air CPU cooler laying around you might be able to use some elastic bands (I know this sounds dumb but hear me out ) to temporary hold a cooler down on the CPU so you can post at the very least, this might be better then trying to re run your custom water loop over to the cardboard box with the MOBD on top as that can get messy if you are pulling everything over to work, even just having the block on the CPU with water in it just for a post and as long as you shut it off right away you should be good, just do't let it run till steam rises out of the open water block if your going to do it that way :p
Click to expand...
Any kind of metal container with a flat bottom and some water in it would work long enough to see if it posts.
 
i know someday when I look back at this as a conquered challenge I will laugh, but for now I just want to smash something. :dead:
 
Tried pulling each ram stick.
Tried resetting cmos multiple times with power on and off
Verified that psu would still power on with the psu bypass plug.

Still need to try:
bios flash
Pulling board and looking for rogue standoffs
Booting board on cardboard box outside of pc
 
Flashed the bios. that process appeared to work normally.

The system still will not boot though. Water pump and fans do not start.
 
We have fans spinning
Removing the motherboard and booting from cardboard box is a good sign. Now to find the ground short..
 

Attachments

  • IMG_0336.JPG
    IMG_0336.JPG
    222.2 KB · Views: 56
Put motherboard back into case with same configuration with the only exception being that I reinstalled the water block instead of the air cooler. I tried plugging the air cooler in instead of the pump, didn't boot.

Pulled the ssd mount and fan controller off the back to check for clearance. Wouldn't boot. No stand offs are in the wrong places. No random screws or anything stuck behind the board.

Nothing I see is an obvious issue.
 
Here's what I'd do, since your system isn't easy to take apart with the H2O setup.

Method one: Remove the PSU from where it sits in the case and isolate it from the chassis. If it's a bottom-mount PSU, let it sit on the floor/table next to the case, just so long as it isn't touching.
Try turning it on.
Works? Seems like something is funky with the PSU or Mobo, passing some amount of voltage somewhere through a ground, tripping the PSU somehow.
Doesn't work? See below...

Method two and/or step two: Get a big enough piece of cardboard that will cover every motherboard standoff. Remove every screw connecting the motherboard to the chassis (including the GPU's bracket), and lift the board up enough to slide that cardboard in, then set it back down. Make sure no metal from the motherboard is touching metal on the case, the rear-IO panel included.
With the graphics card, use electrical tape and tape along all contact points (basically the entire bracket, the bottom tab included), and don't screw it in. Again, we're isolating EVERYTHING from the chassis.
See if it turns on.
Works? Same as above, something funky is going on with the ground.
Doesn't work? I'll have to think about something more...


EDIT: Is your water block's backplate somehow causing a short...? Perhaps try putting tape along every part that contacts the board to... once again... totally isolate it. heh
 
This is one of those love/hate problems. You hate the problem and all the troubleshooting when it happens, then love the feeling you get when you figure it out.

So something with the case or the CPU block is causing the issue. Could you possibly be over-tightening the stand off screws or the block mount? I know this is tough in most cases but can you see if the board is warping or bending anywhere? Could there be a standoff that is shorter or taller than the rest causing the board to flex? Could some of the EMI contact points on the I/O shield be bent into one of the USB ports? I actually had that happen once and caused a similar problem.

This is a weird one for sure!
 
RedWagnum said:
Could some of the EMI contact points on the I/O shield be bent into one of the USB ports?
Click to expand...
I dig the theory. I was racking my brain on how the shield (beauty plate, if we're honest lol) could be causing the issue, but I'd say that's plausible. If not in a port, then perhaps it's suck under the board and making contact with the 5V solder point.
 
Are you using the AM3 mounting holes with an AM3 block on this build? Only asking because I think I remember reading that the CPU are different height so the block may be putting too much pressure on it? Just a thought I had while looking at your update.
 
LigTasm said:
Are you using the AM3 mounting holes with an AM3 block on this build? Only asking because I think I remember reading that the CPU are different height so the block may be putting too much pressure on it? Just a thought I had while looking at your update.
Click to expand...
I can confirm it's indeed taller. That's why Asetek had to redesign their AM4 AOI adapter, and why all the companies aren't getting stock untill just NOW for those.
 
BeerEngineer said:
Put motherboard back into case with same configuration with the only exception being that I reinstalled the water block instead of the air cooler. I tried plugging the air cooler in instead of the pump, didn't boot.

Pulled the ssd mount and fan controller off the back to check for clearance. Wouldn't boot. No stand offs are in the wrong places. No random screws or anything stuck behind the board.

Nothing I see is an obvious issue.
Click to expand...
There was a series of EK backplates that the rubber was conductive enough for the CH6 not to work.
http://www.overclock.net/t/1624603/rog-crosshair-vi-overclocking-thread/5120#post_25961369
Link above is the massive of massive threads for the CH6 - on top of page just type in Search in this thread "EK backplate" to find the appropriate cure for your issue. This is known issue with EK water blocks on the CH6.

elmor at OCN who works for Asus stated this:
EK Predator & Supremacy backplate issue

Using the EK Predator or Supremacy rubber gasket causes the board not to turn on or become unstable. Contact EK directly for a replacement.
Click to expand...

http://www.overclock.net/t/1624603/rog-crosshair-vi-overclocking-thread#post_25889679
 
Last edited:
Thank you for the input! I definitely have the ekwb supremacy evo block, and the kit included a am4 bracket. I read the description from the link to over at the oc net and the description sounds similar. Thank you for posting that. Despite my searching the web for answers that didn't come up previously.

I left the backplate and gasket on when the system booted outside of the case. I had the standoffs installed which retained the rubber gasket and the metal plate.

I opened a Technical Support Case with EKWB.

If I'm able to fix my issue by slapping on a new gasket that would be a positive development.
If the water block is too tight, the directions are incorrect. The directions call for hand tightening until reaching the end of the thread.

I attempted to slightly loosen the water block thumb nuts to test if reducing potential board flex mattered. It did not.

Next step, remove the water block and set it aside, and then retest with the air cooler inside the case.

This should isolate the cause if the EKWB Supremacy EVO is the root cause.

You guys are spectacular.
:)
 
Last edited:
Very interesting that the air cooler, using the EK backplate and gaskets, did not cause the fault to occur... :wtf: I mean, nevermind the fact that the rubber gasket has enough additives (impurities?) to cause it to become conductive! Even if it's only conductive in the slightest of amounts.

Happy to see we're getting somewhere, mostly in that it isn't the motherboard that was faulty.

I'd love to say you're welcome, but I'm not going to pretend I managed to help any LOL Kyle's 'outside the case' suggestion and Noko's legwork tracking down that EK issue, they're the real heroes! *wipes tear* haha
Nevertheless, [H] as a whole is pretty spectacular! :smug:
 
Couple of updates.
1. Attempting to reinstall water block with tips from EKWB were not successful.
2. Attempted fan cooler inside of case again, failure this time!
3. Attempted to remove the center of the rubber gasket, and then the air cooler worked multiple times.
4. Attempted to reinstall water block, success! I booted and got to setup. I have a LOT of building to do now... but I'm seeing progress.
5. When the replacement gasket from ekwb arrives, I'll swap out what I have for the new gasket.

I will let everyone know how things are going once I get moving on the install. Until then

"Don't be a Pussy" - Kyle Bennett
 

Attachments

  • IMG_0337.JPG
    IMG_0337.JPG
    130.1 KB · Views: 56
  • IMG_0338.JPG
    IMG_0338.JPG
    246.4 KB · Views: 49
BeerEngineer said:
Couple of updates.
1. Attempting to reinstall water block with tips from EKWB were not successful.
2. Attempted fan cooler inside of case again, failure this time!
3. Attempted to remove the center of the rubber gasket, and then the air cooler worked multiple times.
4. Attempted to reinstall water block, success! I booted and got to setup. I have a LOT of building to do now... but I'm seeing progress.
5. When the replacement gasket from ekwb arrives, I'll swap out what I have for the new gasket.

I will let everyone know how things are going once I get moving on the install. Until then

"Don't be a Pussy" - Kyle Bennett
Click to expand...

Excellent, now for the fun part. (y)
 
BeerEngineer said:
Couple of updates.
1. Attempting to reinstall water block with tips from EKWB were not successful.
2. Attempted fan cooler inside of case again, failure this time!
3. Attempted to remove the center of the rubber gasket, and then the air cooler worked multiple times.
4. Attempted to reinstall water block, success! I booted and got to setup. I have a LOT of building to do now... but I'm seeing progress.
5. When the replacement gasket from ekwb arrives, I'll swap out what I have for the new gasket.

I will let everyone know how things are going once I get moving on the install. Until then

"Don't be a Pussy" - Kyle Bennett
Click to expand...
I was wondering what that noise was.... It was your epic sigh of relief! lol

Glad to see progress was made :D
 
Great News! Hopefully everything else comes easier.

Besides being possibly conductive, the piece of rubber could be acting like a capacitor holding a charge and discharging it. When dealing with high frequency items a lot of strange stuff can happen. This looks like one of them where a piece of rubber stops booting.
 
In-freaking-credible! I would have never guessed that a rubber gasket would cause such problems. Color me amazed! :jawdrop: Kudos to noko for finding that information. (y)

Definitely something to file in the "weird stuff" folder!
 
You must log in or register to reply here.
Back
Top