sure for that, I've bought mine on january this year ... problably that's why they didn't argue.
Marco
H12SSL-I Stuck at "bmc initiating"
- Thread starter Dick Lee
- Start date
-
- Tags
- bmc boot failure supermicro
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Marco
Update: The pre-heater took its sweet time to get here, but it's a game changer! The lead free solder flows so much easier. Unfortunately I had botched things pretty badly running with just hot air, so was getting no voltage from U4/U5/U6 regulators. Remove/replace them all: Now U4 is 2.5V, U5 is .61V and U6 is zero volts. Let's assume for now that U4 and U5 are functioning correctly... 2/3 is an improvement!
I tried reseating U6, but still no luck. Upon inspection I have clearly lifted the "power good" pad, but it doesn't appear connected to anything? I'm out of components, but placed another digi-key order. If I haven't damaged anything else, this might actually work!
Don't get me wrong, this is the most challenging solder repair I've ever attempted, but it's within reach. If I had access to magnification during the solder process I think I could probably perform the operation reliably. (My current process of nudge, inspect, nudge, inspect, hope isn't very efficient.)
You should give it a go. What do you have to lose?
ref: center left pad should not look like that :/
pretty possible its not connected. microprocessors have internal voltage supervisor/reset units.
the TPS62088 deads very easy on too much too long heat. the factory uses IR soldering(preheated)
0.61 Volt ? the regulation range of the TPS62088 is 0.6 - 4 V.
can't imagen any unit which requires 0.6V but it is not impossible.
plugged ? I thought they didn't do that anymore today, galvanic process instead.
As far I know the via would still have to be filled to be plated over. I don't think mass-produced MBs can afford such an expensive process, especially with so much space around.
the galvanic process builds the via by coating copper in the hole.
edit:
LPKF Contac S4
Through-hole plating for the lab with the LPKF Contac S4 - ✓optional tinning ✓simple operation - Find out more here ⇒
all are manufactured that way. it is lower cost because the only way to proper create 0.3mm via's
Well well.. Shuffled PCIE cards around on H12SSL-i (Rev1.01 from china), afterwards no BMC/IPMI. VGA is dead. BMC NIC led stays green but doesn't do anything. Heartbeat led blinks fast for ~10 secs and then goes off while power is off. With power on, heartbeat stays constant green. Otherwise the system works and I can boot to OS fine.
edit: Added info about VGA.
edit: Added info about VGA.
Code:
ipmitool
Could not open device at /dev/ipmi0 or /dev/ipmi/0 or /dev/ipmidev/0: No such file or directory
Last edited:
@RolloZ170 did say that the symptoms are different depending on which of the three chips is affected and for how long. If your motherboard is booting - awesome, but you should check the chips carefully and maybe get the board repaired while Supermicro is still handling returns
Yeah. The server was removed from the rack and unplugged. Hindsight 20/20 I propably did touch the MB a bit while shuffling cards/cables. No dropping / big force, nothing I would normally worry about while building stuff on 4 decades. Last time I broke something like this I was 10 years old.
Have to be extra careful in the future. Ordered H12SSL-NT for replacement, will try to RMA this after it arrives.I'm here to attest that it is indeed possible to repair one of these motherboards yourself! The BMC heartbeat is back!!
Rollo, you were right to be suspicious of the 0.6v reading. The voltages on my now functioning motherboard: U4 2.5V, U5 1.20V, U6 1.15V. I only removed R57/R63 for out of circuit measurements, but I've probed voltage drops enough to be confident in these output voltages. The attachment has a few notes for anyone else who's interested in tinkering. It's interesting to note that the U5 "power good" signal feeds the U6 enable pin. (U6 won't come alive if U5 is dead)
Definitely reach out to SM if that's an option, but maybe this will give others some hope if not
I ended up having to replace the RAM chip right next to the voltage regulators. It was getting HOT for no reason. Unknown if that was from the initial failure, or if I broke it mid-repair. In either case, it was SO easy to replace compared to these god forsaken voltage regulators!
Rollo, you were right to be suspicious of the 0.6v reading. The voltages on my now functioning motherboard: U4 2.5V, U5 1.20V, U6 1.15V. I only removed R57/R63 for out of circuit measurements, but I've probed voltage drops enough to be confident in these output voltages. The attachment has a few notes for anyone else who's interested in tinkering. It's interesting to note that the U5 "power good" signal feeds the U6 enable pin. (U6 won't come alive if U5 is dead)
Definitely reach out to SM if that's an option, but maybe this will give others some hope if not
I ended up having to replace the RAM chip right next to the voltage regulators. It was getting HOT for no reason. Unknown if that was from the initial failure, or if I broke it mid-repair. In either case, it was SO easy to replace compared to these god forsaken voltage regulators!
Attachments
SuperMicro Europe basically told me to F- myself when trying to init RMA. I cannot see any physical damage to U4,5,6 on my board. I guess the options are try to RMA thru China or sell as-is without BMC (or with BMC-addon card) with considerable loss.
Just want to share that I have a H12SSL-CT rev 1.02
It suddenly died, no BMC, no lights, no response nothing. PSU is fine and all power is regulated through a clean UPS.
No hardware was removed/installed during the time of the failure. It just suddenly turned off and won't turn on again.
SM agreed to RMA. But this sudden failure with no wanring doesn't give me much confidence.
It suddenly died, no BMC, no lights, no response nothing. PSU is fine and all power is regulated through a clean UPS.
No hardware was removed/installed during the time of the failure. It just suddenly turned off and won't turn on again.
SM agreed to RMA. But this sudden failure with no wanring doesn't give me much confidence.
Reports like this is making me consider just ordering cheap 2nd hand Tyan board from eBay to just sit in the shelf as a spare, when (not if) something happens to my H12SSL.
I've been swapping cards quite often, it's probably just matter of time when my luck runs out.
I've been swapping cards quite often, it's probably just matter of time when my luck runs out.
Hi, I have exactly the same problem as in this discussion thread.
My PC turns on but the BMC does not initialize.
If this can be fixed through the programmer its good news,so that I don't make mistakes, I want to clarify
I looked carefully through a magnifying glass at the chip markings on my H12SSL-I board
16-SOIC: Macronix-MX25L51245GMI-08G
16-SOIC: Winbond-25M02GVSFIG
8-SOIC: as far as I understand it does not apply to the BIOS
Can you please clarify where the BIOS chip is and where the BMC chip is?
