Ammonia Posted November 26, 2020 Share Posted November 26, 2020 My Helios64 was running fine with 5 HDDs for a couple of weeks, then I suddenly couldn't access the shares anymore and I noticed only two HDDs were running (the rest having no indicator lights). HDD slot 3, 4 and 5 appears "dead", no power up on boot, no indicator lights. I have checked both +12V and +5V rails and they are present, everything is plugged in. I can move the HDDs from slot 3, 4 and 5 to slot 1 and 2 and they will power up just fine. The hard drives also work in other machines. I'm a bit lost as to what might be wrong, any ideas? 0 Quote Link to comment Share on other sites More sharing options...
gprovost Posted November 27, 2020 Share Posted November 27, 2020 How did you check +12V and +5V ? on the Molex J8 connector directly ? HDD Slot 3, 4 and 5 are on the HDD rail A. So if you get the expected voltage on Pin 1 and 3, then it could an issue with the harness. But it's a bit strange that it stopped suddenly. 0 Quote Link to comment Share on other sites More sharing options...
Ammonia Posted November 27, 2020 Author Share Posted November 27, 2020 I checked the +12V and +5V on the 15-pin power connector directly (that the hard drives plug into). All measure fine, but of course that is without a load connected. In my case only the two hard drives connected to 12V HDD Rail B start up. Something wrong with the power staggering approach in bootloader? dmesg output immediately after booting: ata1: SATA link up 6.0 Gbps (SStatus 133 SControl 300) ata2: SATA link up 6.0 Gbps (SStatus 133 SControl 300) ata3: SATA link down (SStatus 0 SControl 300) ata4: SATA link down (SStatus 0 SControl 300) ata5: SATA link down (SStatus 0 SControl 300) 0 Quote Link to comment Share on other sites More sharing options...
Ammonia Posted November 29, 2020 Author Share Posted November 29, 2020 I did some more troubleshooting. The +5V HDD Rail A is present when no hard drives are connected, but drops to 0.8 V when the the drives are connected. It looks like the rail is enabled by a TDM3421 P-channel MOSFET? Measuring the MOSFET with no load: Gate = 3V, Source = 5V, Drain = 5V. With the rail loaded: Gate = 3V, Source = 5V, Drain = 0.85V Is it implemented as a high-side switch? Any plans of releasing the schematic for the Helios64? I'm no electrical engineer but I suspect the MOSFET is faulty? 0 Quote Link to comment Share on other sites More sharing options...
aprayoga Posted November 30, 2020 Share Posted November 30, 2020 yes, TDM3421 is implemented as high-side switch. According to your measurement, indeed, the MOSFET seems faulty 0 Quote Link to comment Share on other sites More sharing options...
gprovost Posted November 30, 2020 Share Posted November 30, 2020 @Ammonia We will disclose the schematics soon, but here the section you are looking for 0 Quote Link to comment Share on other sites More sharing options...
gprovost Posted December 14, 2020 Share Posted December 14, 2020 @Ammonia Have you managed to fix it yourself ? If not please contact us to support@kobol.io 0 Quote Link to comment Share on other sites More sharing options...
Ammonia Posted December 14, 2020 Author Share Posted December 14, 2020 2 hours ago, gprovost said: @Ammonia Have you managed to fix it yourself ? If not please contact us to support@kobol.io I ordered a replacement MOSFET (same model), including a drop-in replacement from vishay with better power ratings (just in case it should happen again). Unfortunately shipping is taking longer than it should. I temporarily bypassed the MOSFET to get all the hard drives to boot up so I could transfer the more important data. I was hoping fixing it would be faster than receiving a replacement board - and for me dealing with UPS is playing Russian roulette. Fixing it by replacing a 13-14 cent component is the better option for us both. Hopefully I won't ruin the board in the process! 4 Quote Link to comment Share on other sites More sharing options...
SIGSEGV Posted December 14, 2020 Share Posted December 14, 2020 @Ammonia Do you have some pictures of the board with the bypass and link to the replacement MOSFET with better power ratings? 0 Quote Link to comment Share on other sites More sharing options...
TRS-80 Posted December 14, 2020 Share Posted December 14, 2020 (edited) 11 hours ago, Ammonia said: Hopefully I won't ruin the board in the process! Just take your time... steady hand... Edited December 14, 2020 by TRS-80 0 Quote Link to comment Share on other sites More sharing options...
Ammonia Posted December 20, 2020 Author Share Posted December 20, 2020 The MOSFET has now been replaced with a new one of the same model and all hard drives are now booting up as normal. It didn't go completely without any casualties, but shouldn't be an issue. As I often ask myself, is that dust or some 0201s? 0 Quote Link to comment Share on other sites More sharing options...
gprovost Posted December 21, 2020 Share Posted December 21, 2020 @Ammonia Nice job :-) What was the casualties ? Should have told you that better unsolder first the Molex connectors since they easily start to melt when expose to close to heatgun. 0 Quote Link to comment Share on other sites More sharing options...
Ammonia Posted December 21, 2020 Author Share Posted December 21, 2020 16 hours ago, gprovost said: @Ammonia Nice job :-) What was the casualties ? Should have told you that better unsolder first the Molex connectors since they easily start to melt when expose to close to heatgun. R144 and C129 disappeared into the 4th dimensions, as SMDs of that size has a tendency to do. I protected the connectors and the rest with some kapton tape, that worked well. I will put replacements of those passives back after the holidays, just need to get myself a smaller tip for the iron. Doubt that missing soft start will be an issue though. 0 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.