Effect of extension cables

[Dish]

The discussion about whether extension cables is something to try and avoid as far as possible came up in my other thread.

Here

I'm leaning towards a Seasonic S12 for my next build and since i'm going for an ASUS A8N Premium, (might wait to see where the price will start on the A8n32), i'm probably gonna need to use an extension for the ATX12V cable.

My main questions are:

1) Should having to use an extension affect my choice of PSU? (Cable management/clutter issues taken aside.)

2) Just out of curiousity how much does different extensions actually affect things? I'm guessing extending cables to a fan isn't of the same importance to the system.

P.S. I hope this shouldn't be in the off topic section. If so i'm sorry. Just figured since the main question is related to Power Supply this is the right place.

[Devonavar]

1. Will you be overclocking? Do, you have a power hungry CPU (or GPU), and are the programs you typically use highly CPU (or GPU) dependent? Extension cables will cause voltage drops, will can affect how far you can overclock while remaining stable. Similarly, if you are stressing the CPU at full load whenever you run it, lower voltages mean a higher current needs to be drawn to maintain the same power output, which leads to more heat in the cables and generally a slightly less efficient system. If your power supply is at the edge of its capabilities already, the additional current required could overload one of the lines. With a reputable (non-generic) power supply however, it's almost impossible to do this.

2. See (1) above. A voltage drop for a fan is insignificant. The more power a component requires, the more a voltage drop will push up the current it draws, so the CPU and GPU are most affected.

In real use, extensions are almost always perfectly safe and usable. Unless you have an extremely unusual setup (dual CPU/GPU) or plan to overclock, there's no reason not to use them.

[Ackelind]

I use a Seasonic S12 430W together with ASUS A8N SLI Premium in a P180 setup, and I had to use extenders for the P4 connector. My CPU is a venice 3000+, running at [email protected] with no problems at all. My computer has never rebooted or even crashed once. Current uptime is about two weeks since I had to shut it down when I did some cablegami.

When i had just built it, I tried many different fan setups, and everytime I changed something, I ran 3dmark03, prime95 etc at least over night to check temps/stability.

How's that for stability with extenders? I haven't had any problems at all atleast.

[Dish]

What i'm going for is:

P180
ASUS A8N SLI Premium (maybe the 32 one)
AMD 64 X2 4400+
Scythe Ninja
2048 RAM
2 x WD Raptor 74Gb
DVD RW/CD RW Combo
2 x ASUS 7800GTX TOP
Soundblaster Audigy 2
Up to 5 120mm Fans depending on temp/noise
SeaSonic S12-600W

Might add another HD if i ever feel i run out of space too fast. Doubt it though.

More than that i don't think i'll ever put in there. And the example above is what i want the system to handle probably not what I'm gonna get after calculating the price.

Gonna be running mostly 3D-studio, CAD, Photoshop, SketchUp, Illustrator, InDesign.
Even if I'm trying to deny it it's gonna be alot of gaming too.

Probably won't be overclocking too much but in a few years i might have changed my mind on that.

I understand that the PSU should handle an extension with these components but I just want to get a general idea of what and how much effect adding an extension will have. Since I don't have any real knowledge of electrics I can't evaluate the problem myself.

[chirs]

1) Should having to use an extension affect my choice of PSU? (Cable management/clutter issues taken aside.)

2) Just out of curiousity how much does different extensions actually affect things? I'm guessing extending cables to a fan isn't of the same importance to the system.
.

1) The issue with extenders is not the extra wires so much as the additional contact point. If you plan on using it and rarely taking it apart, then you might want to consider some dielectric grease in the junction to minimize corrosion. Even without the grease, unless you live in corrosive areas I doubt it will be an issue in the short (1-2 year) term.

2) The issue with the additional voltage drop will cause a larger power loss when there is more current flowing through the cable. An extension to a fan shouldn't be a problem, and in fact I've used such for the past two years with no problems.

[Dish]

SO is the extra contact point reason enough to go with a less stable PSU or is the effect negligible.

I'm guessing it's not a disaster since many power supplys come with adapters if they lack any needed connectors I just want to figure out if i'd be better off going for another PSU even if it meant getting something i put lower on the list of power supplys I want.

[hofffam]

I think you guys are not really applying the simple principles of electricity to this issue. Wire has resistance - it is related to the material it is made of, the diameter, and the length of the wire. The effect of length is VERY small at the distances you are talking about here. I don't have a Belden catalog handy but the voltage drop over 100 feet of 20 gauge wire is almost nothing. These effects are not related to the voltage or the current across the wire. A connector adds resistance to the circuit. A resistance will cause a voltage drop. If it is a good connector, the voltage drop will be very small. Again, this resistance has nothing to do with the voltage or the current running in the wire.

A wire with a lot of resistance will heat up. The amount of heat will go up with current. The more current - the more heat. So a bigger cpu will draw more current.

All things considered - a 12 inch power supply extension might have a measurable (with a digital voltmeter) effect, but it is so small to be completely negligible.

[Dish]

Thanks, finally a straight answer. Appreciate it.

SeaSonic it is then. ^^

[windsok]

Noone in Australia seems to stock ATX12V 4Pin extender cables

They do however have Molex -> ATX12V 4pin cables. Would using one of these be fine? is there any difference in whats coming down the ATX12V 4pin cable, and the Molex cable?

http://www.auspcmarket.com.au/show_prod ... y_id]=1128

Thanks.

[teknerd]

They do however have Molex -> ATX12V 4pin cables. Would using one of these be fine? is there any difference in whats coming down the ATX12V 4pin cable, and the Molex cable?

the molex -> 12V converter isnt a good idea on ATX v2.0+ power supplies. It is fine on older power supplies though.
The atx 2.0+ spec has 2 12V rails, one for the main system power (the 12V connector) and one for basically everything else. Therefore if you convert a molex to the 12V connector you are basically putting everything on the 12V1 rail, thereby halving the 12V output capacity of your power supply.
atx 1.3- power supplies, however, have only 1 12V rail so you dont have to worry about it.
If anyone is convused at all by this, tell me and ill try to re-explain it more cleaerly.

[chirs]

A resistance will cause a voltage drop. If it is a good connector, the voltage drop will be very small. Again, this resistance has nothing to do with the voltage or the current running in the wire.

You're not totally accurate here. Ohm's law says that the voltage drop across the resistance will be related to the current you push through it. (V=IR)

However, you are correct that for the vast majority of cases it won't be an issue.

[MikeC]

This issue comes up from time to time when testing PSUs in the lab. The leads on some PSUs are not long enough to reach, so we end up having to extend the cables. Extra voltage drop through the extension cable (either 6" or 12" long) does occur, mostly at higher power, but it's in the order of ~0.05V max. It might have hit 0.1V once or twice. In this case, if the female connectors are crimped a bit tighter, the drop gets smaller. As already pointed out, that's where the potential drop is -- in the connectors, not in the wire.

[hofffam]

chirs - what I meant to say was that resistance doesn't change with voltage or current. You are correct - the voltage across a resistor will change as Ohm's law says. So if a connector adds .05 ohms of resistance it will cause a greater voltage drop at high current. Thanks for pointing that out.

[PositiveSpin]

Noone in Australia seems to stock ATX12V 4Pin extender cables

They do however have Molex -> ATX12V 4pin cables. Would using one of these be fine? is there any difference in whats coming down the ATX12V 4pin cable, and the Molex cable?

http://www.auspcmarket.com.au/show_prod ... y_id]=1128

Thanks.

The ATX12V connector puts the load over 2 extra wires (2xblack=0V, 2xyellow=12V). Using a molex adapter puts all the load onto one wire (only one yellow on a molex). So you'll put twice the current onto the yellow wire. Better not put anything else on that molex lead!

And, as pointed out earlier, you'll be overloading one side of the power supply.

I think you'd be better off getting a supply with longer leads - there are plenty of choices that will do.