[Geoman]

This will be my last post in this thread I guess - I succeeded at last! Did a fresh IRIX 5.3 install and all the demos run flawlessly and very fast I must say. For this giant PSU I had to do some slight case modding. (reversible, one could put in an original PSU if needed)

[Irinikus]

Very Nice!!!

[jan-jaap]

Good to see it's alive!

So, do you have to switch it on using the breaker at the back now? Normally, in the 'single tower' PowerSeries/Crimson systems (Diehard2 chassis), the breaker will switch power to the PSU, the fan inside the PSU will run, and the on/off switch at the front will lift the 'inhibit' signal to the PSU and the systems comes to life. In the Professional IRIS (12 slot Twin Tower), the power switch at the front is in the AC input cable, and the LH Reasearch PSU comes to life as soon as AC is applied. There is a breaker at the back and either over current or the 'open chassis' alarm will trigger the breaker, removing power from the system.

[Geoman]

Good to see it's alive!

So, do you have to switch it on using the breaker at the back now? Normally, in the 'single tower' PowerSeries/Crimson systems (Diehard2 chassis), the breaker will switch power to the PSU, the fan inside the PSU will run, and the on/off switch at the front will lift the 'inhibit' signal to the PSU and the systems comes to life. In the Professional IRIS (12 slot Twin Tower), the power switch at the front is in the AC input cable, and the LH Reasearch PSU comes to life as soon as AC is applied. There is a breaker at the back and either over current or the 'open chassis' alarm will trigger the breaker, removing power from the system.

Yes, the Mighty-MITE is in an "always on"- mode. So the System starts with switching the breaker. The module at the front with the circuit bracker does its thing nonetheless, with showing "F", then "1","2" in PROM and I think "1" while running IRIX. no unter/overvoltage signals.

The PSU has a silkscreen on the outside "external (fuse-symbol)" required. I hope that this circuit braker at the back counts as one, since I don't know how to cram in an external 250V fuse.

and one disadvantage is, that Crimsons DC-circuit braker has no effect. So if this PSU has a catastrophic failure, it might kill the Crimson. (well it's like brand new, "should" not happen....)

[jan-jaap]

The PSU has a silkscreen on the outside "external (fuse-symbol)" required. I hope that this circuit braker at the back counts as one, since I don't know how to cram in an external 250V fuse.

The PowerOne PSU has a bunch of 30A LittleFuses in the rectifier section, the LH research doesn't so it depends on an an external breaker. I think the breaker on the back is a 10 or 15A device. It might save your house from burning down but I'm not sure what would be left of the Crimson if it ever exceeds 2300W of power...

and one disadvantage is, that Crimsons DC-circuit braker has no effect. So if this PSU has a catastrophic failure, it might kill the Crimson. (well it's like brand new, "should" not happen....)

I guess you mean the circuit breaker on the PS3 status panel PCB?



The Crimson Field Service Handbook refers to it as the over temperature breaker. On the back side of the system, above the backplane, there's a small PCB with a blue cooling fin on it. It's attached to a temperature sensor located above the card cage. I had that circuit go bad in a 4D PowerSeries years ago and it tripped the breaker. I guess it's supposed to save you system if too many fans in the card cage fail, but I wouldn't count on it.

There's a bunch of voltage status LEDs and test points on the PS3, but if they trip the breaker then at least the terminology in the Field Service Handbook is off. It might light the fault LED. I don't have the pinout for the flatcable so I can't test PS3 functionality outside the system, and I'm not going to subject my Crimson to voltages far enough out of spec hoping to trigger it and see what happens...

[Geoman]

The PowerOne PSU has a bunch of 30A LittleFuses in the rectifier section, the LH research doesn't so it depends on an an external breaker. I think the breaker on the back is a 10 or 15A device. It might save your house from burning down but I'm not sure what would be left of the Crimson if it ever exceeds 2300W of power...

Yes, I bought a seperate overvoltage protection adapter for the wall socket to deal with this. (I hope my thinking is right here)

I guess you mean the circuit breaker on the PS3 status panel PCB?

The Crimson Field Service Handbook refers to it as the over temperature breaker. On the back side of the system, above the backplane, there's a small PCB with a blue cooling fin on it. It's attached to a temperature sensor located above the card cage. I had that circuit go bad in a 4D PowerSeries years ago and it tripped the breaker. I guess it's supposed to save you system if too many fans in the card cage fail, but I wouldn't count on it.

Yes, the PS3-board (status panel assy, I forgot the name).

Crimson's "power delivery circuitry" only works correctly with boards installed inside the cardcage: The ciruit braker also tripped while I tested the PSU with all boards removed (didn't want to fry them on the test run, because there is no such documented try on using an LH-PSU in Crimson).

Without cards installed, the PSU only sends 8 Volts instead of 12. I think that this extremely low voltage failed the temperature sensor, which needs power itself. This, and/or that was also the case:

Code:

Thermal circuit braker - Trips when the system air temperature exceeds 140 °F (60 °C) [...] The breaker will also trip if the system senses a DC voltage that is 40% out of range. --Field Service Handbook p2 - 11

I asked myself what this PCB on the back does, now I know that it is the temperature control unit. :-)  (I checked all fans recently, I won't let this machine ever run unattended)

There's a bunch of voltage status LEDs and test points on the PS3, but if they trip the breaker then at least the terminology in the Field Service Handbook is off. It might light the fault LED.

Yes, I used all these test points to check the voltage. And the fault light on the corresponding LED ("-12V UVP) and the "fault LED yellow" lit, when I undervolted the system.

I don't have the pinout for the flatcable so I can't test PS3 functionality outside the system, and I'm not going to subject my Crimson to voltages far enough out of spec hoping to trigger it and see what happens...

No please don't torture your Crimson! 


Because I already tortured mine unintentionally, by doing a massive undervolting to the backplane. 

[Geoman]

Possibly. I would probably pay for shipping from the EU if another one came up.

I was contacted by the seller, that he found another one. It came with the same original packaging and looks and smells prestine new old stock. BUT: there is only fan span, no voltages at all, with the same test-setup as depicted above. So for this one and the Fuji-PSU I would have to find a repair service in Germany I guess.

Perhaps I'm going to take the Mighty  Mite apart to make a photo documentary.