Sign in with
Sign up | Sign in
Your question

GA-EP45-ED3P no post, no beep, lights up for ~10 seconds, power cycles

Last response: in Motherboards
Share
February 26, 2010 11:57:31 PM

GA-EP45-ED3P no post, no beep, lights up for ~10 seconds, power cycles

Hello all, just want to make sure I'm not missing anything. I just purchased this mb from newegg. I've read the sticky and verified all steps . I used anti-static wrist band. I've pulled the mb out and just tried to boot with just CPU and cooler (no ram, no memory, just pwr switch and speaker, red is positive wire on speaker right?). Verified the 4x2 ATX CPU power plug is in, and ATX power plug is in. I get 8 lights, two red, two yellow, two green, cpu fan flickers. About 10 seconds later it power cycles.

Specs:
GA-EP45-UD3P rev 1.6
Prescott 3.2 ghz (should be supported)
Power supply is a Xscale 470 watt (review)

I know the power supply is a tad weak, but should have no problems booting just a cpu and cooler. The dell the CPU came out of (which I put back in to verify it wasn't fried, and it still works) had a 350W supply. Memory doesn't really matter right now cause I can't even get it to beep at me.

So, do I need to RMA it back to newegg? Any help would be appreciated :sol: 
a b à CPUs
a c 177 V Motherboard
February 27, 2010 1:54:26 AM

No polarity to speaker - doesn't apply... First thing I'd verify is PSU's PWR_OK. Board can't do much of anything w/o RAM, but should produce RAM error beeps... Sequence is this: MOBO power on pulls PSU pin 16 (/PS_ON - the "/" means it's 'low true') to ground, tells the PSU to 'go', PSU starts a timer, powers up rails, when timer times out, multi-channel comparator 'looks at' the rails, of all OK, pulls pin 8 (PWR_OK) high (and I think it's an old TTL level high - should be somewhere above 3.3, close to or at 5) to tell board it's ready to go - if no PWR_OK, board cycles power to try again...

Search "power supply paper clip trick", fool the PSU into turning on, check PWR_OK - if low, check rails; all rails can be fine, but if PSU doesn't bring up 8, you're screwed anyways...
m
0
l
March 2, 2010 1:46:48 AM

bilbat said:
Search "power supply paper clip trick", fool the PSU into turning on, check PWR_OK - if low, check rails; all rails can be fine, but if PSU doesn't bring up 8, you're screwed anyways...


Thank you bilbat for your advice. Just to clarify, your suggesting that I test the power supply to see if it actually maintains power by itself? If that's the case, then I know the power supply works fine becuase I've been using it for over a year in another PC. I've also tested the motherboard against another power supply and the motherboard is acting the same. In the meantime, i've reloaded the Prescott into it's old pc and plugged the new memory into the old pc and ran memtest86+ for two days with no errors. So do you think at this point it would be prudent to rma the board?
m
0
l
Related resources
a c 172 à CPUs
a c 156 V Motherboard
March 6, 2010 2:08:19 PM

bilbat said:
No polarity to speaker - doesn't apply... First thing I'd verify is PSU's PWR_OK. Board can't do much of anything w/o RAM, but should produce RAM error beeps... Sequence is this: MOBO power on pulls PSU pin 16 (/PS_ON - the "/" means it's 'low true') to ground, tells the PSU to 'go', PSU starts a timer, powers up rails, when timer times out, multi-channel comparator 'looks at' the rails, of all OK, pulls pin 8 (PWR_OK) high (and I think it's an old TTL level high - should be somewhere above 3.3, close to or at 5) to tell board it's ready to go - if no PWR_OK, board cycles power to try again...

Search "power supply paper clip trick", fool the PSU into turning on, check PWR_OK - if low, check rails; all rails can be fine, but if PSU doesn't bring up 8, you're screwed anyways...


" MOBO power on pulls PSU pin 16 ..." - that's a TTL logic HIGH (5 volts) being grounded by the case power switch. On an Antec 650 watt TP3, about 3 ma. (.003 amps) flow to ground through the switch.

"high (and I think it's an old TTL level high - should be somewhere above 3.3, close to or at 5" - I checked this on three different PSU's (a cheap Raidmax, the Antec, and a Corsair 750TX) and it goes to 4.8 - 5.0 volts so there must be a pullup resistor involved.

When the gray wire (pin 8) goes HIGH, it removes a hardware reset from the CPU.
m
0
l
!