Bad power supply?

jaydee · 08-18-2004, 09:45 PM

I noticed my power supply doing some pretty wild and continuous swings. It should default to 1.6V (TBred A 1700+) but it is going from 1.54 to 1.62 with constant jumping around.

+12 rail jumps around from 11.80 to 11.86.
+5 rail jumps from 5 to 5.03.

System is running default unoverclocked. Sound bad?

greenman100 · 08-18-2004, 10:12 PM

how were those readings obtained?

nikhsub1 · 08-18-2004, 10:16 PM

Quote:

Originally Posted by greenman100

how were those readings obtained?

Yeah, best to check with a multimeter. I have had plent of mobo's that would suggest a bad PSU when in fact, the PSU was fine but the mobo was borked. It could be bad VR on the board.

jaydee · 08-18-2004, 10:41 PM

Onboard from Abit hardware monitor. I will try a multi meter when I can. What raised a suspicion was the last CPU I had in there was constantly crashing for no obvious reason. Changed CPU's and the problem was solved. Just hoping the voltage jumping didn't damage the CPU over time.

nikhsub1 · 08-18-2004, 11:13 PM

Quote:

Originally Posted by jaydee116

Onboard from Abit hardware monitor. I will try a multi meter when I can. What raised a suspicion was the last CPU I had in there was constantly crashing for no obvious reason. Changed CPU's and the problem was solved. Just hoping the voltage jumping didn't damage the CPU over time.

I'd bet the farm (almost) that it is the board, Abit is known for sketchy VR at times. My first IC7 was so bad it wasn't even funny, had to return it.

Etacovda · 08-18-2004, 11:34 PM

If its an NF7, try throwing an 80mm above the mosfets - stablized my voltages very nicely. Of course, it could just be crappy monitoring.

DrMemory · 08-18-2004, 11:58 PM

When I first started overclocking (and water cooling), I was using a home made gold finger device on a 700 MHz Athlon to set the core voltage and CPU frequency. To monitor the core voltage I hooked up a multimeter to the core voltage supply and saw exactly the same kind of voltage swings you are talking about. Since the CPU continued to function correctly for over a year while being overclocked to 900 MHz (until my motherboard caps started making pop noises) I think what you are seeing is normal. Because of VERY large current swings in the CPU cores themselves, some amount of voltage "instability" is inevitable. The thing I still have trouble understanding is why this can be seen on slow monitoring equipment.

billieboy · 08-19-2004, 05:40 PM

Fluctuating is a result of PSU integral design and power consumption. Translation…power supply isn’t the best. But I guess that’s debateable

bigben2k · 08-19-2004, 05:55 PM

Quote:

Originally Posted by Etacovda

If its an NF7, try throwing an 80mm above the mosfets - stablized my voltages very nicely. Of course, it could just be crappy monitoring.

That'd be my advice.

Otherwise clamp a multimeter to the PSU output rails and see what you read under various loads.

jaydee · 08-19-2004, 05:59 PM

It is an Abit KD7. Has some airflow around the mosfets already. Mosfets are tiny on this board.

Brians256 · 08-19-2004, 06:07 PM

Voltage to CPU should remain stable. Modern power supply design uses active monitoring of output voltage to accomodate large and swift swings in current demands. Using 100KHz pulses, there a PSU should be able to react far quicker than the human eye can read a number. Plus, the DC-DC converters on the mobo smooths the demands even further so that the PSU should see much slower swings in demand.

Take home message: if you see large swings, you have a bad PSU or bad wiring/connections.

These power supplies are made for about $6-$14 deep in China, so they aren't exactly state of the art paragons of quality.

However, the swings you state really aren't big enough to hurt anything. If they swing 5% to 10%... then I'd be worried. All really voltage sensitive devices already have dedicated DC-DC converters that supply a more exact voltage to the component (e.g. RAM and CPU).

edit: spelling error

RoboTech · 08-19-2004, 09:23 PM

Another thing you might check is your ATX connector. If the board has been in service for awhile, it's possible that some of the contacts may have started to oxidize, creating higher resistance and possibly fluctuating voltages. Worst case is when some of the pins actually overheat and burn/melt the connector to the mobo.

Pull the main ATX mobo power connector and check for any signs of over heating or burning. If all looks OK, try re-seating it a few times to clean up the connections. Might be worth a look...

jaydee · 08-19-2004, 10:50 PM

Thanks for the input guys. I will not only check the connector but swap in a couple other power suplies I got stored away.

DrMemory · 08-19-2004, 11:11 PM

I had none of the "possible" problems listed here. It was a 300 W Antec PC power supply on an Abit motherboard. The input voltage at the CPU power supply was NOT fluctuating. The output voltage was. For a CPU dissapating 70 watts with a 1.65 V core we are talking about over 40 amps. Modern CPUs can have large (more than 10 amps) current fluctuations that occur on the order of several CPU clock cycles. Even at "only" 900 MHz that is much faster than any switching power supply can compensate for. And at those currents, even a very small resistance can cause 50-100 mv drops. On Pentium IIs and I belive Pentium III's too, Intel had output pins that gave the CPU power supply a "heads up" when it needed to prepare for such a change. Large capacitors (usually 3 or more 1500 uF caps) on the power supply output can supply the current (only for a short time), but the larger the capacitance, the longer it takes (this may be why these fluctuations are "visible"). Hugh numbers of small bypass caps right on the CPU power pins help, but they simply can't supply enough current. I'm just telling you what I actually saw by actually putting a digital multimeter directly on the output of a CPU power supply. Try it for yourself.

jaydee · 08-19-2004, 11:16 PM

Will put the DMM to work aswell.

AveMORphine · 08-19-2004, 11:45 PM

Check the condensators beside the Mosfets (look at the two at the far end in this picture and you`ll see why the Power connector is "fried" as well. That is the last thing that goes bad). Abit KG7 is well known for having poor condensators. So I would check if they have holes on top.

jaydee · 08-20-2004, 09:11 AM

Quote:

Originally Posted by AveMORphine

Check the condensators beside the Mosfets (look at the two at the far end in this picture and you`ll see why the Power connector is "fried" as well. That is the last thing that goes bad). Abit KG7 is well known for having poor condensators. So I would check if they have holes on top.

Those are capacitors that are blown. All mine are fine. You can replace those to.

greenman100 · 08-20-2004, 10:29 AM

condensators?

sa-weeeeet

Tempus · 08-20-2004, 01:03 PM

isn't that what luke skywaker was working on when the imperials killed is family?

DrMemory · 08-20-2004, 07:18 PM

Originally, when experiments with electricity began (1700's?) capacitors were called condensors. I don't know when electronic circuit designers started calling them capacitors but it was a relatively short time ago.

Abit quit using conterfeit Chinese (well Taiwanese) caps several years ago. They were one of the few (only?) motherboard manufacturers that owned up to the problem and replaced the motherboards. Unfortunately, this was well after I threw my Abit "popping caps" motherboard away.

08-18-2004, 09:45 PM	#1
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	Bad power supply? I noticed my power supply doing some pretty wild and continuous swings. It should default to 1.6V (TBred A 1700+) but it is going from 1.54 to 1.62 with constant jumping around. +12 rail jumps around from 11.80 to 11.86. +5 rail jumps from 5 to 5.03. System is running default unoverclocked. Sound bad?

08-18-2004, 10:12 PM	#2
greenman100 Cooling Savant Join Date: Jun 2004 Location: Florida Posts: 414	how were those readings obtained? __________________ "Employ your time in improving yourself by other men's writings, so that you shall gain easily what others have labored hard for." --Socrates "greenman100 = obnoxious ass hole"-gazorp

08-20-2004, 10:29 AM	#18
greenman100 Cooling Savant Join Date: Jun 2004 Location: Florida Posts: 414	condensators? sa-weeeeet __________________ "Employ your time in improving yourself by other men's writings, so that you shall gain easily what others have labored hard for." --Socrates "greenman100 = obnoxious ass hole"-gazorp

08-20-2004, 01:03 PM	#19
Tempus Cooling Savant Join Date: Jul 2003 Location: Michigan, USA Posts: 456	isn't that what luke skywaker was working on when the imperials killed is family? __________________ Thou art God.

08-18-2004, 10:41 PM	#4
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	Onboard from Abit hardware monitor. I will try a multi meter when I can. What raised a suspicion was the last CPU I had in there was constantly crashing for no obvious reason. Changed CPU's and the problem was solved. Just hoping the voltage jumping didn't damage the CPU over time.

08-18-2004, 11:34 PM	#6
Etacovda Cooling Savant Join Date: Feb 2004 Location: Dunedin NZ Posts: 735	If its an NF7, try throwing an 80mm above the mosfets - stablized my voltages very nicely. Of course, it could just be crappy monitoring.

08-18-2004, 11:58 PM	#7
DrMemory Cooling Savant Join Date: Jun 2004 Location: Dallas, TX Posts: 101	When I first started overclocking (and water cooling), I was using a home made gold finger device on a 700 MHz Athlon to set the core voltage and CPU frequency. To monitor the core voltage I hooked up a multimeter to the core voltage supply and saw exactly the same kind of voltage swings you are talking about. Since the CPU continued to function correctly for over a year while being overclocked to 900 MHz (until my motherboard caps started making pop noises) I think what you are seeing is normal. Because of VERY large current swings in the CPU cores themselves, some amount of voltage "instability" is inevitable. The thing I still have trouble understanding is why this can be seen on slow monitoring equipment.

08-19-2004, 05:40 PM	#8
billieboy Foo's Been Banned Join Date: Aug 2004 Location: england Posts: 3	Fluctuating is a result of PSU integral design and power consumption. Translation…power supply isn’t the best. But I guess that’s debateable

08-19-2004, 05:59 PM	#10
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	It is an Abit KD7. Has some airflow around the mosfets already. Mosfets are tiny on this board.

08-19-2004, 06:07 PM	#11
Brians256 Pro/Staff Join Date: Oct 2001 Location: Klamath Falls, OR Posts: 1,439	Voltage to CPU should remain stable. Modern power supply design uses active monitoring of output voltage to accomodate large and swift swings in current demands. Using 100KHz pulses, there a PSU should be able to react far quicker than the human eye can read a number. Plus, the DC-DC converters on the mobo smooths the demands even further so that the PSU should see much slower swings in demand. Take home message: if you see large swings, you have a bad PSU or bad wiring/connections. These power supplies are made for about $6-$14 deep in China, so they aren't exactly state of the art paragons of quality. However, the swings you state really aren't big enough to hurt anything. If they swing 5% to 10%... then I'd be worried. All really voltage sensitive devices already have dedicated DC-DC converters that supply a more exact voltage to the component (e.g. RAM and CPU). edit: spelling error

08-19-2004, 09:23 PM	#12
RoboTech Cooling Savant Join Date: Sep 2002 Location: Cincinnati, OH Posts: 229	Another thing you might check is your ATX connector. If the board has been in service for awhile, it's possible that some of the contacts may have started to oxidize, creating higher resistance and possibly fluctuating voltages. Worst case is when some of the pins actually overheat and burn/melt the connector to the mobo. Pull the main ATX mobo power connector and check for any signs of over heating or burning. If all looks OK, try re-seating it a few times to clean up the connections. Might be worth a look...

08-19-2004, 10:50 PM	#13
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	Thanks for the input guys. I will not only check the connector but swap in a couple other power suplies I got stored away.

08-19-2004, 11:11 PM	#14
DrMemory Cooling Savant Join Date: Jun 2004 Location: Dallas, TX Posts: 101	I had none of the "possible" problems listed here. It was a 300 W Antec PC power supply on an Abit motherboard. The input voltage at the CPU power supply was NOT fluctuating. The output voltage was. For a CPU dissapating 70 watts with a 1.65 V core we are talking about over 40 amps. Modern CPUs can have large (more than 10 amps) current fluctuations that occur on the order of several CPU clock cycles. Even at "only" 900 MHz that is much faster than any switching power supply can compensate for. And at those currents, even a very small resistance can cause 50-100 mv drops. On Pentium IIs and I belive Pentium III's too, Intel had output pins that gave the CPU power supply a "heads up" when it needed to prepare for such a change. Large capacitors (usually 3 or more 1500 uF caps) on the power supply output can supply the current (only for a short time), but the larger the capacitance, the longer it takes (this may be why these fluctuations are "visible"). Hugh numbers of small bypass caps right on the CPU power pins help, but they simply can't supply enough current. I'm just telling you what I actually saw by actually putting a digital multimeter directly on the output of a CPU power supply. Try it for yourself.

08-19-2004, 11:16 PM	#15
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	Will put the DMM to work aswell.

08-20-2004, 07:18 PM	#20
DrMemory Cooling Savant Join Date: Jun 2004 Location: Dallas, TX Posts: 101	Originally, when experiments with electricity began (1700's?) capacitors were called condensors. I don't know when electronic circuit designers started calling them capacitors but it was a relatively short time ago. Abit quit using conterfeit Chinese (well Taiwanese) caps several years ago. They were one of the few (only?) motherboard manufacturers that owned up to the problem and replaced the motherboards. Unfortunately, this was well after I threw my Abit "popping caps" motherboard away.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)