• Please review our updated Terms and Rules here

When Do You Know a Motherboard is Bad?

This explanation/definition is floating around in several places :

The term comes from the power lines on motherboards. Power must be available throughout a motherboard; hence, voltage lines tend to run in long strips like railroad tracks.
Click to expand...
 
I suspect the term actually predates motherboards as we know them and is a bit idiomatic. Probably avoids the use of "bus" and confusion. However, in that case, what does one use bus bars for? :huh: If the durned thing hits a bump, you end up spilling your drink...

If one goes to an IBM 5150, there is no ground "track"--it's buried in a middle layer as a plane. So, you're running what, a monorail? Where's North Haverbrook? :)
 
I guess motherboards involve railroad tracks, engineers, and buses? :p

Well power supply rail goes back at least as far as 1972, according to Google books:

...All devices in the T.T.L. range work from a single +5V power supply rail...
Click to expand...

Nothing before 1970, except references to railroad rails. Maybe it did start with PC Boards..?
 
Dunno--I'll have to look at some of my old books. I expect the terminology to be coincident with solid-state electronics, but that's just a guess.
 
LOL, the OP title sounds like a joke waiting for a punch line.

When Do You Know a Mother Board Is Bad? ....

When the daughter board re-volts?

:p
 
Just to kinda wrap this thing up, I replaced the Gigabyte motherboard with a slightly newer model but was able to use the same processor. Had to also replace the DDR2 memory with DDR3. The board booted without issues using the same power supply.

The two eMachine motherboards are probably also shot. One of them goes on as soon as it's plugged in and you can't turn it off unless you pull the plug. The other starts and all the fans spin, but nothing else. If I feel like it, I'll get a couple of micro ATX boards for them as well.

Thanks...Joe
 
If you ever decide to repair them, my first thought is to replace the electrolytic capacitors - especially the larger ones near the processor.
But generally today it's easier to just do what you did, and replace it with one that still works.
 
It's not as easy to toss a vintage board. I just ordered a couple set of capacitors for an old PPro motherboard to fix it since they are hard to find and pricey.
 
Hi Guys,
After getting one of the PCs working again by buying a new motherboard (processor and power supply appear to be good), I started working on the other three. As an aside...the deal he got at Micro Center was amazing. To buy a new board and 8GB of memory only cost him (he says) about $45 because the board was free (usually $45) if you also bought a processor. Since we never used the processor, he returned it and was able to keep the board discount.

Now, about those other machines...

First up was a machine using an AMD Duron (I assume that is a K7 based on the motherboard). When I tried to fire this one up, I got an electrical crack followed by a little puff of smoke. When I disassembled the machine, I found a capacitor and I think I see where it fell off the board. I am going to try replacing it and see what happens.

Question...if the capacitor is marked "100uf 16v" can I safely use a capacitor marked "100uf 35v"? I only ask because the 35v version was the only one available at the local Radio Shack. If I can't use it, I should be able to locate an exact replacement but may need to do so online.

On to the other two machines. These are eMachines. One uses an Athlon 64 and the other an Intel Celeron.

Athlon 64: When switched on, this machine powered up all the fans but nothing else occured. No video, no beeps, nothing. The fans continue to run until I switch it off.

Celeron: All the fans go on as soon as the power supply is plugged in and you cannot switch it off. There is no video and no beeps.

After disassembling both machines and placing the motherboards on the table, I tried to boot each one with only memory installed (both have onboard video). Both acted the same way as they did fully assembled.

I then tried to boot the Celeron board using the power supply from the Athlon machine. Nothing happened - no fans, nothing. I still need to try the Celeron PS on the Athlon board and will update this post once I do. I also want to try to boot them using a known good power supply.

Question...if either or both eMachine power supplies are bad, can they hurt a good motherboard? I'm thinking the only way I can verify whether the eMachine power supplies are good (assuming my good supply doesn't power the boards up) is to hook them up to a known good board. And just the opposite...can a bad board hurt a good power supply?

Thanks...Joe
 
I think I mentioned this at least once before -- you are never going to get anywhere if you keep testing devices of unknown viability with devices of unknown viability. You need to use a meaningful test procedure, the cornerstone of which is a postulate which only allows for *one* variable per test/experiment. You *must* use test equipment, i.e., components, of known good status to test any questionable component. If you don't do this you are just spinning your wheels and wasting everybody involved's time. You cannot determine whether a component is good or not unless you are 100% certain that every component you are using to test it with is 100% good! Going about testing any other way is just an exercise in futility. Does this scenario sound familiar? :) If it doesn't then you're either not paying attention or just not up to the task of testing these components. You really should reevaluate your test method(s) and make sure every individual component you use each time you attempt to test something is in 100% working condition itself before you actually use it because if this is not the case none of your results will be useful. IOW, using unknowns to test other unknowns is not an accepted method and therefore will not give you the answers you are seeking.
 
Grandcheapskate said:
Question...if the capacitor is marked "100uf 16v" can I safely use a capacitor marked "100uf 35v"? I only ask because the 35v version was the only one available at the local Radio Shack.
Click to expand...

The "16v" is the manufacturer's working voltage/voltage rating of the capacitor. The "35v" should work providing it is the right type (electrically) and size isn't a factor.
 
I routinely do that--as long as the ESR is comparable--and I prefer to replace with 85C 'lytics with 105C ones--the inside of a PC can get pretty toasty. The same goes double for LCD monitors.
 
Stone said:
I think I mentioned this at least once before -- you are never going to get anywhere if you keep testing devices of unknown viability with devices of unknown viability. You need to use a meaningful test procedure, the cornerstone of which is a postulate which only allows for *one* variable per test/experiment. You *must* use test equipment, i.e., components, of known good status to test any questionable component. If you don't do this you are just spinning your wheels and wasting everybody involved's time. You cannot determine whether a component is good or not unless you are 100% certain that every component you are using to test it with is 100% good! Going about testing any other way is just an exercise in futility. Does this scenario sound familiar? :) If it doesn't then you're either not paying attention or just not up to the task of testing these components. You really should reevaluate your test method(s) and make sure every individual component you use each time you attempt to test something is in 100% working condition itself before you actually use it because if this is not the case none of your results will be useful. IOW, using unknowns to test other unknowns is not an accepted method and therefore will not give you the answers you are seeking.
Click to expand...

I agree with you completely. That is why I am asking if a bad motherboard can hurt a good power supply and whether a bad power supply can hurt a good motherboard (note the last paragraph in my prior post). In order to test these components I need to use a good power supply and good motherboard/memory combination. I just don't want to destroy something good to test something bad.

Unfortunately I cannot test the CPUs seperately as I only have the one board for each type.

Joe
 
Grandcheapskate said:
Unfortunately I cannot test the CPUs seperately as I only have the one board for each type.
Click to expand...
Then you can't test anything. (If it fails to start how will you know what failed, the board or the CPU?) :)

I will add that if the above occurs, in all probability, it's the board as CPUs tend to be a bit on the bulletproof side in most circumstances. But it's not a certainty.
 
I took the P2 machine which I just got fully functional ( the one with the EMM386 issue) and tried the two power supplies from the eMachines. Neither one powered up the board. When I put the original power supply back...nothing. The machine no longer powers up.

After days of working on this machine to get it working, it appears one of the eMachine power supplies must have blown the motherboard. I cannot express how deflated I am right now after spending so much time getting this machine running. This will be the last time I subject a working motherboard to a power supply test.

Joe
 
Grandcheapskate said:
After days of working on this machine to get it working, it appears one of the eMachine power supplies must have blown the motherboard.
Click to expand...
Somehow I really doubt that this is what actually happened.

You can probably test the PSU(s) with an old HD that spins but doesn't function otherwise. Hook it up and if it spins check the PSU's voltages.
 
Stone said:
Somehow I really doubt that this is what actually happened.

You can probably test the PSU(s) with an old HD that spins but doesn't function otherwise. Hook it up and if it spins check the PSU's voltages.
Click to expand...

eMachines used non-standard pinouts on some of their power supplies with correspondingly rearranged connectors on the motherboard. Bad things happen when the wrong voltages are plugged into the wrong places.

Edit: Even the "good" power supplies used by eMachines were bottom of the barrel models with no protection against excessive current or incorrect voltages and sometimes did take out the motherboard on failure. Reusing the power supply from an eMachines system tends to be unwise. A web search can find many complaints about the problems of eMachines.
 
Last edited:
krebizfan said:
eMachines used non-standard pinouts on some of their power supplies with correspondingly rearranged connectors on the motherboard. Bad things happen when the wrong voltages are plugged into the wrong places.

Edit: Even the "good" power supplies used by eMachines were bottom of the barrel models with no protection against excessive current or incorrect voltages and sometimes did take out the motherboard on failure. Reusing the power supply from an eMachines system tends to be unwise. A web search can find many complaints about the problems of eMachines.
Click to expand...

I had heard about this before (and also about Gateways) and should have simply trashed the PSUs rather than try them with a good board. I will therefore trash the eMachine motherboards without testing them with a good power supply because if the pinouts are different, bad things will happen again.

I had one eMachine power supply go bad a couple years ago and I replaced it without a problem. Maybe I just got lucky and the motherboard used the standard pinout.

Stone...if I wanted to try your suggestion at a later date with other power supplies, how do I turn on an ATX power supply if it's not connected to a motherboard since it doesn't have a switch like AT power supplies? Do I have to jump a couple of the wires on the power connector?

Thanks...Joe
 
You must log in or register to reply here.
Back
Top