Building new system now: Antec P182 case: Gigabyte DS3P mobo: Xeon E3110, etc. So far I've got the cpu, hsf, psu, motherboard, 4gb ram, and all fans installed. Have not begun the wiring, that's next. I wanted to know if there's any way to use a multimeter to check for possible shorts, etc, before actually using A/C power. If not, what do I absolutely need wired to check system so far, especially psu, mobo, cpu & hsf, and all fans (blowing the right direction). A bit concerned about mobo short, though I've followed all guides closely. Oh, I am using the Xigmateck S1283 with the free bolt-thru w/backplate kit, which is chief concern.
jsc
Champion
- Jul 14, 2004
- 16,893
- 6
- 52,960
Hmmm. I never thought of doing that. I don't worry about accidental grounds. I'm just really careful and I breadboard before I install the parts in a case. That way, I KNOW I am installing good parts.
But talk about coincidences: I just happen to have a bread boarded DS3P on my worktable plugged into my KVM switch.
So, after unplugging the PSU, using my handy-dandy Rat Shack dmm with range switch set to OHMS with the black lead on a black (ground) PSU pin:
orange wire (3.3 volts) to ground - about 350 ohms
red wire (5 volts) to ground - open
purple wire (5 volt SB) - about 300 ohms
green wire ( -12 volts) - open
yellow wire (12 volts) to ground - about 350 ohms for an instant, the jumping to open. Using KOHM range - about a thousand ohms.
In your case, with the system in a case, unplug all the PSU cables. Ground black (negative) probe to case. Red probe to all the motherboard PSU pins. The gound pins (where PSU black wires go) should read a short to ground. Everything else should read a few hundred ohms or higher.
If you do not read a short on the pins where black wires go, chose a ground pin on the main motherboard power connector, put the black lead of dmm on it and measure everything else.
Remember, we are measuring solid state electronics here. Different dmm's use different voltages and currents to measure resistance. And unlike resistors, semiconductors will respond differently. If you measured anywhere from, oh, a hundred to a thousand ohms, you probably do not have a detectable power suppy short.
I do not know what else can be done with home test equipment. Keep in mind that motherboard PCB's are multilayer and if anything else is internally shorted, you probably cannot detect it.
Fans - the good fans have an arrow indicating direction of airflow somewhere on the outside of the shroud. If it doesn't, the side of the hub with the label generally indicates the output side.
"If not, what do I absolutely need wired to check system so far, " Everything except the hard drives. With CPU and HSF, RAM, and video card installed, the system should successfully POST - the single short beep.
But talk about coincidences: I just happen to have a bread boarded DS3P on my worktable plugged into my KVM switch.
So, after unplugging the PSU, using my handy-dandy Rat Shack dmm with range switch set to OHMS with the black lead on a black (ground) PSU pin:
orange wire (3.3 volts) to ground - about 350 ohms
red wire (5 volts) to ground - open
purple wire (5 volt SB) - about 300 ohms
green wire ( -12 volts) - open
yellow wire (12 volts) to ground - about 350 ohms for an instant, the jumping to open. Using KOHM range - about a thousand ohms.
In your case, with the system in a case, unplug all the PSU cables. Ground black (negative) probe to case. Red probe to all the motherboard PSU pins. The gound pins (where PSU black wires go) should read a short to ground. Everything else should read a few hundred ohms or higher.
If you do not read a short on the pins where black wires go, chose a ground pin on the main motherboard power connector, put the black lead of dmm on it and measure everything else.
Remember, we are measuring solid state electronics here. Different dmm's use different voltages and currents to measure resistance. And unlike resistors, semiconductors will respond differently. If you measured anywhere from, oh, a hundred to a thousand ohms, you probably do not have a detectable power suppy short.
I do not know what else can be done with home test equipment. Keep in mind that motherboard PCB's are multilayer and if anything else is internally shorted, you probably cannot detect it.
Fans - the good fans have an arrow indicating direction of airflow somewhere on the outside of the shroud. If it doesn't, the side of the hub with the label generally indicates the output side.
"If not, what do I absolutely need wired to check system so far, " Everything except the hard drives. With CPU and HSF, RAM, and video card installed, the system should successfully POST - the single short beep.
Great response jsc! Yeah, I may as well run some multimeter tests to see if I can learn something. Then use A/C (maybe in GFI?) to check, after adding video card. That'll also tell me if I'll hear any short beebs at all, cause I can't find those tiny noisemakers on the chassis or the motherboard. Nothing found in both manuals.
jsc
Champion
- Jul 14, 2004
- 16,893
- 6
- 52,960
Unlike my 680i motherboard, the DS3P does not have a builtin piezoelectric beeper. You need to get a small speaker with a 4 pin header. If you check the manual on Page 29 Section 15 Front Panel Header, it shows you where to connect the speaker. I'm using one salvaged from an old case.
I don't think the Antec cases come with one. Apparently your P182 doesn't, and I know my 900 doesn't either.
You need one. They are invaluable for troubleshooting. And it will provide an audible alert for CPU overtemp (BIOS settings).
I don't think the Antec cases come with one. Apparently your P182 doesn't, and I know my 900 doesn't either.
You need one. They are invaluable for troubleshooting. And it will provide an audible alert for CPU overtemp (BIOS settings).
jsc
Champion
- Jul 14, 2004
- 16,893
- 6
- 52,960
True, Zorg.
"Poking" isn't the problem. Knowing when, where, why, and with what to poke is. For example, using a regular meter-type multimeter runs the risk of damaging solid state electronics.
On one hand, if there's no NEED to poke, you shouldn't. A related corollary: If it ain't broke, don't try to fix it.
On the other hand, I have more than 45 years experience poking about various types of electronics.
On the third hand, I was so focused on the best way for rickkip to do what he wanted that I lost track of the fact that, in the absence of trouble, he shouldn't look for it.
I was looking at the trees, not the forest.
"Poking" isn't the problem. Knowing when, where, why, and with what to poke is. For example, using a regular meter-type multimeter runs the risk of damaging solid state electronics.
On one hand, if there's no NEED to poke, you shouldn't. A related corollary: If it ain't broke, don't try to fix it.
On the other hand, I have more than 45 years experience poking about various types of electronics.
On the third hand, I was so focused on the best way for rickkip to do what he wanted that I lost track of the fact that, in the absence of trouble, he shouldn't look for it.
I was looking at the trees, not the forest.
Zorg
Splendid
- May 31, 2004
- 6,732
- 0
- 25,790
I understand what you are saying. I know "poking" was a little harsh, I used the word intentionally. I'm sure your credentials are impeccable. The question, as you realized, was regarding his credentials. I have often seen the trees and not the forest when explaining things. The problem is that you can't see if his eyes glaze over or not. I am a firm believer in "If it ain't broke, don't fix it". I also like KISS, unfortunately I don't follow it often enough.
In either case, OP did you fire it up? Does it run?
In either case, OP did you fire it up? Does it run?
RetiredChief
Judicious
- Feb 22, 2007
- 8,528
- 0
- 32,960
Agree with Zorg, Comments for info only.
Altho I agree, poking around for some can be a hazard. On the other hand we have seen several install MB without standoffs which the DVM would have spotted. It could prevent reflected problems, ie problem one subsystem "destroying" other subsystems.
(1) Output of DVM, in ohms position, will not damage home computer components. Voltage output is normally less than 1 volt and is current limited. If DVM could damage the subsytem, just think how may would die from ESD.
(2) Problem of understanding some readings, ie negative readings which could be caused by a small capacitve charge (Most power connections are connected to Caps. Readings that start off low and increase - again Caps that are charging. Then again, unless you know what the reading should be, all that you are check is for "Shorts".
(3) Equally valid, if going this far, would be a ohms check from supply -> supply pins on MB, ie +3.3V to +5V to +12V Before and after installing Memory and GPU.
NOTE: Do I do this - NOPE. Was a required check on SAGE III satelite instrument any time the cables were demated and remated. But that was a $10 million instrument.
Altho I agree, poking around for some can be a hazard. On the other hand we have seen several install MB without standoffs which the DVM would have spotted. It could prevent reflected problems, ie problem one subsystem "destroying" other subsystems.
(1) Output of DVM, in ohms position, will not damage home computer components. Voltage output is normally less than 1 volt and is current limited. If DVM could damage the subsytem, just think how may would die from ESD.
(2) Problem of understanding some readings, ie negative readings which could be caused by a small capacitve charge (Most power connections are connected to Caps. Readings that start off low and increase - again Caps that are charging. Then again, unless you know what the reading should be, all that you are check is for "Shorts".
(3) Equally valid, if going this far, would be a ohms check from supply -> supply pins on MB, ie +3.3V to +5V to +12V Before and after installing Memory and GPU.
NOTE: Do I do this - NOPE. Was a required check on SAGE III satelite instrument any time the cables were demated and remated. But that was a $10 million instrument.
RetiredChief
Judicious
- Feb 22, 2007
- 8,528
- 0
- 32,960
Zorg
Myself, I don't. Based on "old design thru late 80's" . However I think the newer models say after late 90's the manuf. wised up and placed a min load resistance on the outputs, ie draw maybe 50 -> 200 millis BUT can not swear to it, nor that the internal ouput resistance satisfies that min load current.. Haven't heard "dummy Load" for a while (also used as Radar antenna load), another term is bleeder resistor as it allows filter caps to discharge.
Some labels (or in specs) will state minimum current load.
Spec on antec 850 W (Min load specs)
+5 - 200mA, +3.3 - 100 mA, +12 V1 - 100 mA, +12 V2 - 500 mA
+12 V3 & V4, +5Vsb are Zero
Would have to power down and disconnect to measure internal resisance.
ohms ck 20 Pin connector coolerMaster iGreen 600W Not Matted
Internal Load resistance
+3.3 29 ohms
+5 V 90 ohms
+12V1 255 ohms
Added.
If I wanted to Check a PSU out I would place it under a nominal load, at least 50 % to 75% load. Would like to make a load Bank. Probably could use some high wattage light bulbs or toaster wire (Wife would not like but could go by used one for $5.
Myself, I don't. Based on "old design thru late 80's" . However I think the newer models say after late 90's the manuf. wised up and placed a min load resistance on the outputs, ie draw maybe 50 -> 200 millis BUT can not swear to it, nor that the internal ouput resistance satisfies that min load current.. Haven't heard "dummy Load" for a while (also used as Radar antenna load), another term is bleeder resistor as it allows filter caps to discharge.
Some labels (or in specs) will state minimum current load.
Spec on antec 850 W (Min load specs)
+5 - 200mA, +3.3 - 100 mA, +12 V1 - 100 mA, +12 V2 - 500 mA
+12 V3 & V4, +5Vsb are Zero
Would have to power down and disconnect to measure internal resisance.
ohms ck 20 Pin connector coolerMaster iGreen 600W Not Matted
Internal Load resistance
+3.3 29 ohms
+5 V 90 ohms
+12V1 255 ohms
Added.
If I wanted to Check a PSU out I would place it under a nominal load, at least 50 % to 75% load. Would like to make a load Bank. Probably could use some high wattage light bulbs or toaster wire (Wife would not like but could go by used one for $5.
OP Here: New day! Last night I triple checked everything and had everything done including a fax modem and external esata ports, came with mobo. Left both sides off, hooked up monitor and ps2 keyboard/mouse and turned on the a/c. Zap! Not a sound of that, just the whirring of nice fans. All turning in the correct direction. All seems to be great so far. I've got XP on a 30gb partition (out of my 2 Samsung Spinpoint 320's) and have formatted the rest. Now I'm backing stuff off the old computer (wabs, favorites, Roboform data, banking data, jokes, etc., all going to thumb drive. Got to switch over to get internet setup on the network, and install many more programs. So far, great success. I'd like to thank all here and throughout the Tom's and Anandtech reviews and especially forums for all the help. I've been studying, printing, and saving to favorites many articles that I culled through. Guess I gotta say, if a 62 year old geezer with 100% disability can do this, most anyone can if they apply themselves. Hmmm, been thinking of a big screen and HTPC. Naw, gotta give the wife a break, cause she's been awfully patient and understanding with my bouts of insanity! Thanks again.
Zorg
Splendid
- May 31, 2004
- 6,732
- 0
- 25,790
Good job, see the meter wasn't needed.
100% disability? My partner was Airborne in Vietnam and got hit by a piece of a rocket. It took out half his face, one lung a couple of fingers and more and he only got 60%.
They did a good job fixing his face, you can't tell.
100% disability? My partner was Airborne in Vietnam and got hit by a piece of a rocket. It took out half his face, one lung a couple of fingers and more and he only got 60%.
They did a good job fixing his face, you can't tell.
radnor
Distinguished
- Apr 9, 2008
- 1,021
- 0
- 19,290
There is no danger in Turn on a PSU with no loads. I had to do it every PSU in another tech job i had. We just couldnt risk it. But be careful. Ill give you a "Ghostbuster" quote. Dunno why but i feel its.....apropriate in this thread.
Dr. Egon Spengler: There's something very important I forgot to tell you.
Dr. Peter Venkman: What?
Dr. Egon Spengler: Don't cross the streams.
Dr. Peter Venkman: Why?
Dr. Egon Spengler: It would be bad.
Dr. Egon Spengler: There's something very important I forgot to tell you.
Dr. Peter Venkman: What?
Dr. Egon Spengler: Don't cross the streams.
Dr. Peter Venkman: Why?
Dr. Egon Spengler: It would be bad.
jsc
Champion
- Jul 14, 2004
- 16,893
- 6
- 52,960
I thought about that. I am not sure that light bulbs (have to use automotive) or toaster wire would work. Cold resistance is much lower than operating resistance. A computer PSU might overload and shutdown. I have thought about getting some power resistors and bolting them onto an old P4 heatsink to use with my inexpensive PSU tester. Hermetically sealed, 50 watt, low ohm power resistors are available from Mouser Electronics for about $5 apiece.
Hey, Chief, we could suggest that rickkip use a megger.
RetiredChief
Judicious
- Feb 22, 2007
- 8,528
- 0
- 32,960
jsc
100% correct on cold resistance. 120 V / 55 watt Bulb has a off resistance of approx 20 ohms, on resistance at 120 V R = 260 ohms.
Need to ck a 500, or 1000W floodlight.
using the 12 V rail with a 20A limit.
R load would need to be 600 milliohms. The real question would be as the filiment heated up what would the real load be. w/the 1000 W bulb it would be 14 ohms (too high) IF 120 was applied, but with only 12 Volts, don't know upper R ( would be less than 14 ohms.
If cold resistance was less than 600 milliohms, Then the question of overcurrent is a question of race. Most PS's can withstand a higher inrush current, normally in mSec and not normally found on consumer grade PS's - Then it becomes a question of, will the filiment heat up fast enough to drop I to below threshold before triggering overload circuit. ( Same comments about toaster wire - will have to check cold R.
100% correct on cold resistance. 120 V / 55 watt Bulb has a off resistance of approx 20 ohms, on resistance at 120 V R = 260 ohms.
Need to ck a 500, or 1000W floodlight.
using the 12 V rail with a 20A limit.
R load would need to be 600 milliohms. The real question would be as the filiment heated up what would the real load be. w/the 1000 W bulb it would be 14 ohms (too high) IF 120 was applied, but with only 12 Volts, don't know upper R ( would be less than 14 ohms.
If cold resistance was less than 600 milliohms, Then the question of overcurrent is a question of race. Most PS's can withstand a higher inrush current, normally in mSec and not normally found on consumer grade PS's - Then it becomes a question of, will the filiment heat up fast enough to drop I to below threshold before triggering overload circuit. ( Same comments about toaster wire - will have to check cold R.
TRENDING THREADS
-
-
[SOLVED] Is it safe to use a recapped PSU for a long time ?- Started by zwtch_17
- Replies: 7
-
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 3K
Facebook
Twitter
Instagram