Molex wires

[ghidu]

I sleeved my PSU cables for some time and didn't pay any atention to the ground wires.
Now I read that they're not the same, but one is 12V return and the other 5V return.
Is this BAD? How can I repair this? Is there a way to find them now?

 

[Rike]

The black ground wire is matched to the nearest colored wire. You need to jump two pins (someone else will say which ones since I'm too lazy to look it up) to turn the PSU on with out hooking it to any components. Then test each connection with a multi-meter. Yellow to black should be 12V and red to black 5V. If you get something other than that, you?ve mixed up the wires. Test everything or you could cook some parts. Good luck.

 

[FlyingPenguin]

On most PSUs then two black wires are the same - they're both connected to chassis ground. On some fancy premium PSUs they MIGHT be different, but I doubt it. Knowning the way their designed they really both have to be tied to chassis ground.

If in doubt, use a multimeter in the OHMS setting to measure resistance between the two black wires. If there's a dead short (no resistance) they're tied together internally and it doesn't matter which one is used for what.

 

[Rike]

Originally posted by: FlyingPenguin
If in doubt, use a multimeter in the OHMS setting to measure resistance between the two black wires. If there's a dead short (no resistance) they're tied together internally and it doesn't matter which one is used for what.

That makes sense. But if that is true, why bother to have separate pins for the ground wires to begin with?

 

[ghidu]

OK, can I use the multimeter when the computer is on? I've been using the PC like that months and the only problem I noticed is that my HD is making cracking noises.
So with multimeter on ohms, there have to be now resistance between the ground wires. On DCV, 12V between 12V line and its ground wire and 5V between 5V line and return 5V. Did I got it right?
Thanks for the help, it's really appreciated

 

[Rike]

Oh, you've already been running.

Well if it works, I guess you've OK so far. The fact that you have resistance between ground lines seems to indicate there is a meaningful difference between them. Just to be sure, you should check all your Molex connectors to see if you switched any of them. If everything is OK, you should get 12 V and 5V on the correct wires and nothing else.

I'd guess that your strange HD noises are something unrelated.

 

[ghidu]

Originally posted by: Rike
Oh, you've already been running.

Well if it works, I guess you've OK so far. The fact that you have resistance between ground lines seems to indicate there is a meaningful difference between them. Just to be sure, you should check all your Molex connectors to see if you switched any of them. If everything is OK, you should get 12 V and 5V on the correct wires and nothing else.

I'd guess that your strange HD noises are something unrelated.

My question still stands. Measuring with the pc on isn't going to show wrong numbers? I mean on one molex I have 0.5Ohms between the 2 ground lines and on another 9ohms and another 8 ohms. I read that starting a PSU without a MB could damage it
And shouldn't be the other way: no ohms means that their not connected (not on the same line?). I know very little electronics, that's why I'm not sure what to do.

 

[shiranai]

The numbers better not change when you turn on your machine. Ground should always be ground, 12V should be 12V, and 5V should be 5V, regardless of what is connected to them (within a certain margin, of course). If this is not the case, then something is wrong.

Zero input resistance between two terminals indicates a short. An open (where two terminals are not connected) has 'infinite' resistance. Therefore no, it isn't the other way, the resistance between your two ground lines should be very close to zero.

 

[ghidu]

I just did a test. With the PSU connected to my PC I cleared everything on 1 line of molex. The resistance between the 2 ground wires is 0.6ohm on every molex. The voltage between 12V and any ground (close to it or close to 5V) is 11.94V; although on one molex there was 11.94 and 11.93. On the other line of molex the voltage was 11.95V betwen 12V and ground. Between 5V and ground - 5.and something- same number anyway.
So, the resistance between my ground lines is 0.6, pretty close to 0. Does that mean they're all the same? Is there any other tests to run to make sure?
Normally I'm very careful when it comes to my computers and modding them, but previously I read that they're both the same and this morning I find that's not quite so.

 

[furballi]

You may be seeing the contact resistance of the test probes. Short the two probe leads and verify that there is less than 0.2 ohm. The resistance between the two ground wires should be within 0.1 ohm of the test probes.

For example, my test proble resistance is 0.2 ohm. The resistance between the ground wires is 0.2 ohm per this meter. If I use the delta mode on the meter to zero the effect of the test probes, then I see 0.0 ohm resistance between the ground leads.

 

[JimPhelpsMI]

Hi, You got it right. The black wires come from the exact same place in the power supply. The system doesn't care which way you connect them. It's done to reduce the voltage drop in the wires. Same as using larger wire. Jim

 

[RaiderJ]

Originally posted by: JimPhelpsMI
Hi, You got it right. The black wires come from the exact same place in the power supply. The system doesn't care which way you connect them. It's done to reduce the voltage drop in the wires. Same as using larger wire. Jim

Yup. Ground is ground.

 

[ghidu]

Originally posted by: RaiderJ

Originally posted by: JimPhelpsMI
Hi, You got it right. The black wires come from the exact same place in the power supply. The system doesn't care which way you connect them. It's done to reduce the voltage drop in the wires. Same as using larger wire. Jim

Yup. Ground is ground.

That's excellent Jim and Raider, I put alot of time and money into this PC and I was afraid I f*cked up.

Originally posted by: furballi
You may be seeing the contact resistance of the test probes. Short the two probe leads and verify that there is less than 0.2 ohm. The resistance between the two ground wires should be within 0.1 ohm of the test probes.

For example, my test proble resistance is 0.2 ohm. The resistance between the ground wires is 0.2 ohm per this meter. If I use the delta mode on the meter to zero the effect of the test probes, then I see 0.0 ohm resistance between the ground leads.

I get different numbers when I short the to 2 leads from my multimeter, but it stays longer on 0.6. What's delta mode?
Thank you every1 for clearing this up for me.

 

[PsYcHoCoW]

The wires you measure resistance with, ARE resistive ; touch them together and see what reading you get (most likely around half an ohm). Delta mode has you touch both wires together when entering it, and substracts that value from the resistance read until you take delta mode off.

Grounds are linked together at the PSU, possibly to avoid ground noises from spreading from 12v to 5v grounds. Mixing them shouldn't cause problems though.

 

[Zepper]

In the old days when real computer power supplies were linear (huge transformers and caps), the grounds for the various voltages could be floating (which means they could be at some voltage relative to the others and or true earth ground). That was when the drive connectors were set in concrete. Anymore all they would need is three wires as ground is common among all voltages and should be a true earth ground. And the pattern is (looking at the working end of the connector with the beveled corners up) +12V, +5V return, +12V return, +5V. It has been a long, long time since I've seen floating grounds anywhere (on purpose that is )...

.bh.

 

[Rike]

Thanks for the history lesson, Zepper. What everyone was saying about "the ground is the ground" made complete sense, but then why have separate ground wires to begin with? Thanks for filling in the gaps.

I guess it's just another example of how today we are bound to how things were done in the past, even when the reason for it has long become irrelevant.

 

[ghidu]

Originally posted by: Rike
Thanks for the history lesson, Zepper. What everyone was saying about "the ground is the ground" made complete sense, but then why have separate ground wires to begin with? Thanks for filling in the gaps.

I was asking myself the same thing and that's why I started to have doubts.
Thanks Zepper for the explanation. I have a testing PSU and I am using any black wire as ground and didn't think for a moment that actually they might be different - when in fact they were, but not anymore
PsYcHoCoW, thanks for the lesson in electronics, but I don't think my multimeter is that good.
See ya all:beer: