Water cooler for Monarch Bitminer

[Mumps]

Second, @Mumps: Does the unit power up? If so, how long does it run before it starts throwing errors or shuts down?

In my case, the unit will power up and run for maybe at most 15 minutes before it shuts down. (Actually trips the self-protection on the 1000W PS.)

Of the two "pumps" that are mounted on the ASICs, the one farther from the radiator "growls" and gets progressively louder until the shut down, so I'm sure that's toast. It may also have something to do with the way the radiator (and therefore the tubes) are mounted on the sled. So if there's any leak that leads to air in the system, it's guaranteed to be right there going into the tubes exiting at the top of the radiator to the pumps. No indication of any leak though. The unit ran for months without ever being moved off the shelf it was originally put on and there was never any fluid expelled I ever saw.

If I do bother repairing it, I'd most likely look to add a reservoir to the loop so I could bleed air off and add coolant if it ever became required. But I still don't think it'd be worth more than the $50 I originally paid for the Monarch to repair it.

[John P. Myers]

In my case, the unit will power up and run for maybe at most 15 minutes before it shuts down. (Actually trips the self-protection on the 1000W PS.)

Of the two "pumps" that are mounted on the ASICs, the one farther from the radiator "growls" and gets progressively louder until the shut down, so I'm sure that's toast. It may also have something to do with the way the radiator (and therefore the tubes) are mounted on the sled. So if there's any leak that leads to air in the system, it's guaranteed to be right there going into the tubes exiting at the top of the radiator to the pumps. No indication of any leak though. The unit ran for months without ever being moved off the shelf it was originally put on and there was never any fluid expelled I ever saw.

If I do bother repairing it, I'd most likely look to add a reservoir to the loop so I could bleed air off and add coolant if it ever became required. But I still don't think it'd be worth more than the $50 I originally paid for the Monarch to repair it.

Noisey pumps can be caused by an air bubble getting lodged in the pump chamber and the pump still be perfectly fine. If it's possible to add a few drops of water to the system, might want to try that before spending any money.

[Mumps]

Noisey pumps can be caused by an air bubble getting lodged in the pump chamber and the pump still be perfectly fine. If it's possible to add a few drops of water to the system, might want to try that before spending any money.

Yup. Understood. Cavitating is common on our Glycol loops here at the office as well, so I'm familiar with the general concept. (Just on 5HP pumps, like the pair right above my office, rather than a CPU sized one. )

In this case, the Monarch loops are closed systems, so even to get to the point of trying to add any fluid, it would require breaking the loop. So I might as well insert a reservoir that can get the port for bleeding air or adding fluid. And the fact that the Monarch trips the overload protection on the 1000W PSU, and it's the only thing on it, suggests there's truly something struggling and the heat isn't being removed from the system as needed.