Question Odd Problem of a Mobo Drive Controller LED

[BonzaiDuck]

Last spring, I built a system with some dated parts, including a (new) ASUS workstation motherboard. I had an NVME drive installed on the motherboard with two SATA ports cabled to 2.5" HDDs, and a PCIE dual-NVME (bifurcated) card with two drives. I connected a red LED light to the motherboard controller pins, and yellow LEDs connected to the PCIE card.

I've suddenly noticed, after less than a year's operation, that the red LED never goes dark: the red light still remains on, and drive activity is just obvious because it flashes more intensely.

Since these are motherboard pinouts in the two rows of pins for "power on", "reset", "piezo speaker" etc. -- I'm wondering if I just shouldn't disconnect the lead for my red LED. The "power on" light is a green LED. It's not necessary so much for me to visually see drive activity with the red LED.

This just mystifies me. And it wasn't an observable feature when I first put the system together.

Any thoughts?

 

[mikeymikec]

I haven't seen that before. You've definitely got the orientation correct for the LED plugs? What happens if you switch the power LED with the HDD LED, same issue?

When you say "flashes more intensely" you mean that it remains on but glows brighter to denote activity?

 

[BonzaiDuck]

I haven't seen that before. You've definitely got the orientation correct for the LED plugs? What happens if you switch the power LED with the HDD LED, same issue?

When you say "flashes more intensely" you mean that it remains on but glows brighter to denote activity?

Yes. There's no mistake about the orientation with the pinouts. The power and HDD connections are proper.

This was a system built with a Stacker case that had damaged switches because of some liquid spill -- some mechanical reason that caused the power switch to fail. I had posted a thread showing my solution: new PWR and RESET switches installed on a PCIE backplate and connected to the appropriate pinouts. I think I will bring up my Z170-WS manual to see if there's an alternate HDD pinout, but I vaguely recollect that the only pair with an alternate was the power-on/off switch. I need to refresh myself.

There's no indication of drive or controller failure. I had really taken pains with the wiring: I had to extend the wires, because the LEDs are in the top of the case. Wiring was all soldered carefully, and insulated carefully. I used the proper gauge wire, identical to the original or any such wiring in LED and switch kits. The same misbehaving red LED used in this system is identical to an HDD LED used for the earlier twin system I'd built with a Sabertooth Z170S board.

Not being an electronics expert, I'm wondering if it couldn't just be a flaw in the LED itself. I wouldn't know. But if I have to live without a boot-drive activity light, that's not a big problem. The original system has an NVME boot drive in a PCIE slot, and it's not connected to the HDD LED pinouts. So the HDD LED for that one flashes only when the SATA devices have activity.

One or both of these systems has limited longevity. I was planning to build an i7/i9-12700K system by end of the year with a Z690 motherboard. But other than being a problem for Windows 11 (without using the hack), these are otherwise stellar systems. Last year and starting in 2021, I went through all this trouble with the original Skylake system because of a static-charge disaster that borked the motherboard USB controller. I had started a thread about it in mid-2021. If anyone remembers, a loose mobo or drive power connector was causing occasional random reboots after mobo replacement, and it took several months to resolve it. I acquired the extra Z170-WS when I was collecting parts (thus, I built the twin with a Kaby Lake). I've got spare 2x8GB RAM to put in the original unit for a total of 48GB. The one with the fucky LED already has 64GB.

It's probably just a day for pulling cases apart and attending to these things. I'm waiting for my car-repair shop to call me this afternoon for delivery of my SUV after the starter-motor/solenoid died. What would we do in life without little problems to solve? Watch bad television?

 

[BonzaiDuck]

Opened the Z170-WS manual and there's no "spare HDD LED" pinouts. It will be easy to disconnect: it's on the end of the pin array.

I'd still like to know why it's misbehaving. But like I said, as long as the controller itself is working properly, it's just a frill that I can do without. Never saw something like this before!

 

[WilliamM2]

Opened the Z170-WS manual and there's no "spare HDD LED" pinouts. It will be easy to disconnect: it's on the end of the pin array.

I'd still like to know why it's misbehaving. But like I said, as long as the controller itself is working properly, it's just a frill that I can do without. Never saw something like this before!

Are you sure it's misbehaving? Maybe you do have constant drive activity, I've seen it before.

 

[BonzaiDuck]

Are you sure it's misbehaving? Maybe you do have constant drive activity, I've seen it before.

Task Manager doesn't show anything in that regard. All the drives all show "0%", while the CPU activity is between 2% and 3% -- essentially idle . . .

I don't think I've ever seen anything like this before.

Being old and F***ed up, I sort of struggled to pull my systems apart this afternoon and do what I planned. I just yanked the HDD-LED wires from the mobo pinouts. And like I'd said, I don't show anything else that suggests something wrong with the onboard Intel SATA controller. Otherwise, I think the NVME drive socketed to the motherboard is a PCIE device. But I have two other NVME drives on a PCIE x8 card, and they're not showing any of the same behavior.

The other thing I did with the twin pc was to socket in some identical G.SKILL 2x8 TridentZ RAM of the GTZ series. There was already a 2x16 kit of the GTZR (R denoting LED bling), and both kits were DDR4 3200 14-14-14. Identical specs. The system wouldn't boot. I'm pretty sure this was due to failing to reset the board to default 2133 speed, and then after booting to enter BIOS, change the speed back to 3200. With all four sockets, it would probably have needed a boost to the VCCIO voltage anyway.

So I just pulled out the 2x8 kit and had everything working soon with the original 32GB GTZR kit. I was just trying to make use of some spare, known-good RAM.

But these operations were on two different PCs with the same chipset and CPU-family. I'm just surprised at how this little bit of work makes me slightly tired.

Try and avoid aging if you can help it . . .

Just to add this: before disconnecting the LED, the HDD-LED light goes on as soon as the system is started -- before you could even have a chance to enter BIOS and certainly before posting to the monitor. Otherwise, and as I said already, 0% activity on all drives in the system per Task Manager.