RockDawg Posted June 12, 2007 Share Posted June 12, 2007 I'm having a hardware problem that is baffling me. I have a GIGABYTE GA-945GZM-S2 mobo and I was using it in a small case to migrate my drives and data from an existing server. It has four onboard SATA ports along with one onboard IDE channel. While it was in the small case, everything worked fine. I had two IDE drives on the lone IDE channel and three SATA drives on the onboard controller and they all were always detected by the BIOS on boot. I never had to alter any BIOS settings. Things just worked. Now, I have transferred the entire system to a large server case and all of the sudden whenever the system is powered down, it will fail to see my two IDE drives. I have changed my IDE cables multiple times. They are short, flat and 80 wire. If I randomly change the "On-Chip SATA Mode" BIOS setting a few times and then return to "Enhanced Mode" (both SATA and IDE) they will be recognized and work as long as I don't power down. Even a reboot that doesn't power down will be alright, but as soon as the power goes off, the drives refuse to be recognized on future bootups. Does anyone have any idea what the problem could be? Why did everything use to work, but not now? Link to comment
parsec Posted June 12, 2007 Share Posted June 12, 2007 Longshot 1: loose/slightly loose/failing battery, setting the bios to default when powered off. Link to comment
Joe L. Posted June 12, 2007 Share Posted June 12, 2007 Longshot #2 (actually, a list) Marginal power supply, different temperature in case, different orientation of drives causes initial spin-up to draw more current on supply rail than it was designed for. Power supply has "split" 12 volt outputs, now all on one and it is overloaded, before, load split between two. Link to comment
RockDawg Posted June 13, 2007 Author Share Posted June 13, 2007 I'll check all those things. Thanks for the input. Link to comment
RockDawg Posted June 14, 2007 Author Share Posted June 14, 2007 Longshot 1: loose/slightly loose/failing battery, setting the bios to default when powered off. BIOS settings are sticking, but I tried a known working battery anyway and nothing changed. Link to comment
RockDawg Posted June 14, 2007 Author Share Posted June 14, 2007 Longshot #2 (actually, a list) Marginal power supply, different temperature in case, different orientation of drives causes initial spin-up to draw more current on supply rail than it was designed for. Power supply has "split" 12 volt outputs, now all on one and it is overloaded, before, load split between two. I tried two different power supplies: a 430W and a 500W. Both are Antecs. Problem still exists. The new case is larger and has cooled drive bays, so drive temps went down 10-15 degrees on average. I tried all kinds of different plug connections to try and change the way the load is distributed, but nothing changed. Throughout my trials I did find a couple oddities. If I disconnect the data cables to all three of my SATA drives, the mobo would find the onboard IDE drives every single time. Also, when I tried the 500W power supply it would power up and then shut off after a second if I had all three SATA drives plugged into it. As soon as I unplug the power cable to any one of them it would power up and stay on. It did this regardless of whether or not all 9 of my IDE drives were plugged in or removed from the power supply. The 400w never showed any problems no matter how many drives were plugged into it. It was what was powering all 11 drives in my previous server (non-raid) for thr last couple years. I'm really at my wits end and I have no idea what to do to get this workng again. Link to comment
limetech Posted June 14, 2007 Share Posted June 14, 2007 1/ Check the mounting of your motherboard in the chassis and of PCI cards (if any) in the motherboard. Make sure everything is well seated, no possible shorts, all connections solid. 2/ You mention the new case has "cooled drive bays". As a test, bypass those bays - just hook up drives directly to power & data cable, letting them hang out the case. Also consider disconnecting all the case fans - but not the cpu fan It should be perfectly fine to run your system like this long enough to figure out if it's now working correctly. If so, start reconnecting stuff one-by-one until you find the component which is the culprit (e.g., a bad fan might be producing all kinds of ripple on +12). Link to comment
RockDawg Posted June 15, 2007 Author Share Posted June 15, 2007 1. I tried that today. I totally removed everything and re-installed it and it did the exact same thing. 2. By cold bays, I mean devices like 4-in-3 adapters with a 120mm fan in front except they are 3-in-3. There are no secondary connections. I have disonnected all but the cpu fan and tried every logical combination. I still get no love. I'm starting to think it's my mobo. Link to comment
RockDawg Posted June 17, 2007 Author Share Posted June 17, 2007 I fianlly decided to buy an additional IDE controller card and disabled the onboard channel. Since doing that, I haven't had any problem. Must just be a funk in the mobo. Link to comment
limetech Posted June 22, 2007 Share Posted June 22, 2007 ... Must just be a funk in the mobo. Yep. If something's not working right, 99% of time it's hardware - problem becomes one of isolation - at least that's what I like to say to keep those engineers busy while I debug the code Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.