19-09-2019, 12:14 PM
The fuse has to handle the inrush current. Also, this power supply needs to work down to 100V, which results in a greater current pull.
Being from a HP machine, this is a good quality power supply, with active PFC and decent filtering. It is a 12V-only type, as I was discussing in the other thread - so it will be much better than a cheap generic multi-rail type. The safety approvals will be genuine.
This isn't from the 6200 mini-towers that Dave sold a few years back - this is from a small form factor (SFF) machine. No problem; it'll work well enough if the caps are still in good health.
Regarding joining together the two 12V rails, I think that should be OK. As you've seen, the converter makes a single 12V. For a while, the ATX spec included a 20A limit for each 12V output, so separate current limiting was required (hence the shunt resistors you've found). That requirement has changed from "should" to "recommended", for what that's worth.
It's likely that the resistance of the 12V cables will be more than the shunt resistors, so I'm fairly sure the current will share reasonably equally. It does need checking, however, as if either rail is overloaded, the whole supply is throttled back. The easiest way is with a DC current clamp, such as this one: http://www.markhennessy.co.uk/budget_mul...ut210e.htm
PS_ON is pin 3 of P2. Hopefully it's green wire. Personally, I'd use a switch on the front panel.
This power supply has a fan input control (pin 1 of P2) so the motherboard can control the speed of the fan. In the absence of the this signal, I assume the fan will run at a reasonable speed - it's a 4 wire fan, and these usually run at full speed in the absence of a PWM control signal. Pin 2 is the tacho signal from the fan, and can be ignored.
I don't know what the minimum load requirements for this supply might be, but that should be pretty easy to check. Don't forget the standby rail might have a modest minimum load requirement as well. The -12V output probably doesn't.
Most likely those presets adjust the output voltage of the main and standby converters. It's unusual to have an adjustable output for the standby supply, so I'm guessing slightly! Having identified which adjusts the main converter, we might need to do some reverse-engineering and resistor changing to get the adjustment range you want. Should be possible, but good close-up photos will be needed...
Finally, I don't think you mentioned the earth thing before.
Yes, most switched-mode power supplies do indeed have their outputs referenced to ground. This is mostly for EMI reasons. And with a cheap and nasty power supply, it might also help with safety as well. Floating outputs are obviously possible, but less common.
It might be possible to lift the ground in your supply, but honestly, that's risky territory. Definitely not something I like to do, and that's with the unit in front of me - I certainly wouldn't want to advise on it from a distance.
Being from a HP machine, this is a good quality power supply, with active PFC and decent filtering. It is a 12V-only type, as I was discussing in the other thread - so it will be much better than a cheap generic multi-rail type. The safety approvals will be genuine.
This isn't from the 6200 mini-towers that Dave sold a few years back - this is from a small form factor (SFF) machine. No problem; it'll work well enough if the caps are still in good health.
Regarding joining together the two 12V rails, I think that should be OK. As you've seen, the converter makes a single 12V. For a while, the ATX spec included a 20A limit for each 12V output, so separate current limiting was required (hence the shunt resistors you've found). That requirement has changed from "should" to "recommended", for what that's worth.
It's likely that the resistance of the 12V cables will be more than the shunt resistors, so I'm fairly sure the current will share reasonably equally. It does need checking, however, as if either rail is overloaded, the whole supply is throttled back. The easiest way is with a DC current clamp, such as this one: http://www.markhennessy.co.uk/budget_mul...ut210e.htm
PS_ON is pin 3 of P2. Hopefully it's green wire. Personally, I'd use a switch on the front panel.
This power supply has a fan input control (pin 1 of P2) so the motherboard can control the speed of the fan. In the absence of the this signal, I assume the fan will run at a reasonable speed - it's a 4 wire fan, and these usually run at full speed in the absence of a PWM control signal. Pin 2 is the tacho signal from the fan, and can be ignored.
I don't know what the minimum load requirements for this supply might be, but that should be pretty easy to check. Don't forget the standby rail might have a modest minimum load requirement as well. The -12V output probably doesn't.
Most likely those presets adjust the output voltage of the main and standby converters. It's unusual to have an adjustable output for the standby supply, so I'm guessing slightly! Having identified which adjusts the main converter, we might need to do some reverse-engineering and resistor changing to get the adjustment range you want. Should be possible, but good close-up photos will be needed...
Finally, I don't think you mentioned the earth thing before.
Yes, most switched-mode power supplies do indeed have their outputs referenced to ground. This is mostly for EMI reasons. And with a cheap and nasty power supply, it might also help with safety as well. Floating outputs are obviously possible, but less common.
It might be possible to lift the ground in your supply, but honestly, that's risky territory. Definitely not something I like to do, and that's with the unit in front of me - I certainly wouldn't want to advise on it from a distance.
