Computer not posting

**starfury1** · 03-18-2007, 05:13 AM

Re: Computer not posting

Hi AKBessy

Did you replace all caps or only those two....?

With electrolytic's and age/heat (they do have a finite life anyway)
it best to replace them all ....esp if they are known dud brand caps

Also as I have pointed out in your other post, psu may well be stuffed too.
So you need to check that as well
BE CAREFUL HERE THEY ARE DEADLY, GIVE THEM THE RESPECT THEY DESERVE

UNPLUG IT AND LET IT STAND FOR AWHILE before you open it and have alook at the caps

If its been running since around 2001
and has had a lot of use then yep good chance.

Caps can fail and still look ok so you cant just go by what they look like.

you also must replace with good quality LOW ESR type for best life expectancy
Using hobby store General Purpose type caps will not do.

If you need a recap kit and are USA based check with TC
(to my mind this would be quicker then sourcing all over the place, if he does a kit for them?)

Can't help you out on the door thing with the CD maybe a symptom of your problem or maybe another fault...have to leave that to those who know these puters better then I.

Some models of PSU can destroy HDDs and peripherals

You first need to find out about the psu and MB
workout which is, or are they both part of the problem.

I guess the big question is
"how much time and money do you want to throw at keeping this old box flying"

Yes there would be a chance of other damage so you could be throwing good money after bad

Anything of this age you should check both PSU and MB as a matter of course in my opinion anyway

It does help if you have another some what expendable box to test components on like the CDROM psu etc,
although I would do other checks before risking a box on a possible dud PSU

Cheers

**tazwegion** · 03-18-2007, 06:30 AM

Re: Computer not posting

Test the PSU @ the ATX connector (with a digital multimeter) whilst powering up the system

**AKBessy** · 03-18-2007, 10:51 AM

Re: Computer not posting

I ordered a custom cap kit from badcaps and replaced all of the caps 1000uf and up.

taz, I took the dummy factor out of testing and put in a known good psu a few minutes ago- still the same symptoms.

(I am working on this old computer for the experience. Throwing away a little money is okay if I am learning something.

)

**AKBessy** · 03-18-2007, 11:07 AM

For starfury from another board

starfury, I hope you got my message on the other board. Anyway, here's what you said.

"Hi

Not sure if its the MB pcb only or a full system in came in you are talking about
Sounds like the box as a whole unit

Sounds like you checked the MB and replaced the caps there
if this was done correctly and solder joints etc are ok
Then there a very good chance it should have fired up

Since it hasn't

OK I'll ask the dumb question.

Have you checked the PSU for bad caps?

have you tried a known working PSU

is other damage possible....yes

if the MB had badcaps and the box has an el-cheapo psu its a fair chance its cactus too.

here's a post by TC on that MB

didn't check version thought
Is it the same as yours ?"

+++++

The motherboard came as part of a whole computer. My dad's computer quit working and he said I could fool around with it to see if I can fix it. He got a new computer in the meanwhile, so I am not under pressure to "get 'er done" right away ;-). Anyway, the board is the MS-6312 v.1 .

My soldering is questionable, although I'd like to think that it wasn't ;-). As I mentioned in the above posting. I swapped out the psu and put a known good one in with the same results.

I am not an electronics person. I trained and got certified in refrigeration and heating, and any electronics experience has been in swapping out boards - not troubleshooting. Although I got certified, I still suck at electrical troubleshooting! Electrical "water flowing" is a concept that I just haven't been able to grasp when I have a multimeter in my hand. I am working on it though.

**AKBessy** · 03-20-2007, 02:17 AM

Re: Computer not posting

I removed the FETs near the DIMM slots tonight. I did an Ohm test and a test using the Diode test function on my Fluke 29 Series II Multimeter to get some numbers for the FETs. Are either of those tests applicable for determining a good FET? If so, could someone tell me how to interpret my numbers if I post them?

Another question: Can someone tell me what the different legs are called for the FET shown below? I have the legs labeled S=, G=, and D=. I understand that there is a "source", "gate" and "drain". I thought the source was "s", "g" was the gate and "d" is the drain. Am I correct?

Thank you in advance.

Attached Files

**tazwegion** · 03-20-2007, 03:43 AM

Re: Computer not posting

Originally posted by AKBessy

I ordered a custom cap kit from badcaps and replaced all of the caps 1000uf and up.

taz, I took the dummy factor out of testing and put in a known good psu a few minutes ago- still the same symptoms.

Hmmmm... I'm not totally familiar with Gateway PC's, but from what I've read Gateways' along with early Dell PC's and possibly some others had propriety wired ATX connectors & motherboard headers OR oddly shaped PSUs, in short industry standard ATX PSUs may not either work or fit propriety case(s)

Originally posted by AKBessy

(I am working on this old computer for the experience. Throwing away a little money is okay if I am learning something.

)

Great sentiment... more power to ya'

Back on topic...

Did you test the motherboard/system for any activity prior to recapping? and if so, does it exhibit any similar characteristics now?

Personally I'd gut the system & set it up with only the bare essentials (on a non-static surface)

Test for power, video signal & RAM/HDD recognition (progress from there)

It may be possible that something other than the capacitors is the issue, a short on the underside of the mobo (usually caused by errant standoffs) can bring it all to a screeching halt

Here's a guide to PC troubleshooting (there's plenty around) @ PC world

**AKBessy** · 03-20-2007, 10:57 AM

Re: Computer not posting

I didn't test for anything when I first got the system - that's too bad too in my hindsight vision.

In my picture above, I labeled the S and G wrong - I didn't take into consideration that the FET was upside down! LOL. Anyway. I forgot about my books on electricity from school. So, I cracked them open this morning and was reading about transistors. I am understanding that an FET is basically comprised of two circuits within it: the gate being one circuit and source-to-drain being the other circuit. The theory of how it works reminds me of a step transformer. The gate circuit controls (induces) the flow in the source-to-drain circuit.

Ok, so looking at this test I have been trying to conduct using the diode testing method, I still have a question.

Here is the test instructions: (see http://www.4qd.co.uk/serv/mostest.html for the full article)

"So: connect the meter negative to the MOSFET's source. This is arrowed in the drawing above which shows the most popular TO220 MOSFETs.

Hold the MOSFET by the case or the tab if you wish, it doesn't matter if you touch the metal body but be careful not to touch the leads until you need to.

First touch the meter positive on to the gate.

Now move the positive meter probe to the drain. You should get a low reading. The MOSFET's gate capacitance has been charged up by the meter and the device is turned on.

With the meter positive still connected to the drain, touch a finger between source and gate (and drain if you wish, it matters not). The gate will be discharged through your finger and the meter reading should go high, indicating a non-conducting device.

Such a simple test cannot be 100% - but it is useful and is usually adequate."

In this test, after touching the positive lead to the drain, it says to leave the positive lead on the drain and touch the source-to-gate with a finger. Do I leave the negative lead on the source leg while doing this? I am thinking that if I am trying to drain with my finger, that I should be taking the negative lead off - the instructions don't specify.

Ohms test results on the two FETs - both are the same, CEB703AL:

FET #1: S-G= 8.8, G-D=9.2, S-D=0.9
FET #2: S-G= 30.5, G-D=34.1 S-D=4.5

Both FETs show continuity using the beeper on all three tests between legs. But, if the gate is physically separate from the source-drain, it shouldn't be showing continuity with the other circuit should it - shouldn't it read OL when I am doing an Ohm test?

**tazwegion** · 03-20-2007, 10:00 PM

Re: Computer not posting

Ah it's all greek to me! LMAO

It did prompt me to do a little research however, came across THIS little interesting (interactive) N-type MOSFET circuit diagram/explaination

Good luck with it all, I'm sure one of our resident electronic engineers can steer you straight (if need be)

BTW I don't know what happened to cause that double post of mine above, perhaps my out-of-date Opera version?

**AKBessy** · 03-21-2007, 01:16 AM

Re: Computer not posting

taz, do I have to take the quiz too? LOL

**tazwegion** · 03-21-2007, 02:01 AM

Re: Computer not posting

If that's what you'd like to do... I just figured it may be of some (limited) assistance

* cues quiz timer *

**AKBessy** · 03-21-2007, 11:15 AM

Re: Computer not posting

Thanks taz. I like the interactive thingy and what it is displaying. I keep making the mistake of looking at it when I am tired, so I get easily confused and can't keep things straight. I need to look at it during the day. Understanding what is happening (the process) is the whole key to understanding the numbers I am looking at when I put a multimeter on the components!

**starfury1** · 03-22-2007, 10:14 AM

Re: Computer not posting

Hi

Bit late with this reply seems you got good info on the fets and also think someone posted in the other thread on how to test in circuit.

Fets should never read zero ohms between pins.

Tazz is right about the psu connector on some older models.... but then I think they seen the light
(this was in the back of my mind and why I said standard somewhere)

here the nonstandard from a search link1 link2
another how to test link3

hers another on making an adapter...here
and other info

BTW I haven't really check any of the Above so you are on your own there although others may have a comment or 2 on it

on testing semiconductors of all sorts link1

that whole site is worth a Browse

OK on understanding electricity and yeah its not hard
just hard getting the idea working in your head...oh and BTW being sleepy and trying this is really not recommended

if you can fathom whats happening in this post here on neutral earth you will be on your way.

there is more to it of course;

AC frequency passive components and active components how they work and behave with DC AC frequency phase shifts between current and voltage.

important part is to first realize that current and voltage are 2 different animals and yes the water model doesn't quite fit as a way to explain it in some respects

I have been slack I meant to add a little more to that post.

Taz's suggestion of removing it from box and laying it up (bare min)
is a good idea as I think you may also have some other issue.

using it as training wheels is alway a good way to learn, practical

I would still check that supply but I know others here could probably tell you what you will find....they work with them on a day by day basis I dont
sadly

well hope this helps you out and did I say Welcome? if not WELCOME AKBessy

You have now enter the twilight zone......add your own spooky music here

Cheers

**AKBessy** · 03-23-2007, 11:37 AM

Re: Computer not posting

Thanks for the information starfury. I have had some training and experience with electrical stuff - but, it's one of those things where I definitely have to give my undivided, rested attention to learning and understanding. I haven't had the opportunity to work with electronics much at all. I enjoy this board because I am getting so much good information to read and people to talk to. It's as good as going to school - well, almost - hehe, I don't have a teacher to come over and watch me while I run through a test and correct me as needed.

Yes, I am using my first pc board as training wheels and tearing it apart and doing testing on it - what a great way to go!

Thank you for the welcoming. I am glad to meet all of you and enjoy being in this cyber twilight zone with all of you!

**tazwegion** · 03-24-2007, 05:08 PM

Re: Computer not posting

Originally posted by AKBessy

I am glad to meet all of you and enjoy being in this cyber twilight zone with all of you!

There's no need to make it sound so seedy! ROTFLMAO

* puts on virtual cyber sunglasses *

**AKBessy** · 03-25-2007, 12:09 AM

Re: Computer not posting

Oh gosh, you're right taz...I just cleaned my glasses and see what the heck I wrote. I am so sorry folks - I didn't mean to get sappy on you! I don't even drink and I was still spilling tears of joy in a beer. LOL

**kc8adu** · 03-25-2007, 09:36 AM

Re: Computer not posting

i just recapped 4 of those mobo's and none would post till i pulled the battery and cleared cmos.
the only ones that havnt survived were ones with burned mosfets.this is with a sample size of about 300.
about 50% of the ones with burned mosfets were fixable.
high on the problem list are busted rear usb .the pins short to the grounded shell when they break.

**AKBessy** · 03-25-2007, 07:18 PM

Re: Computer not posting

Hi kc8adu,

Well, I did attempt to clear the CMOS using the switch and pulling the battery. The computer still wouldn't boot.

I removed the all of the caps 1000uf and above. Used soldering gun and wick to get about 1/2 the holes cleaned out. Then, I did the "no-no" and used a tiny drill bit and cleaned out the remaining holes because I was worried that I was overheating the board. I kept the bit centered and drilled out only enough to be able to get the new cap leads through the holes - so I felt pretty good about my drill job. I blew off the board. I soldered on the new caps, but had a bit of a time with the first two as the solder wasn't adhering. I felt that I should have used flux on the board to clean it up the way the solder wasn't adhering, but I thought that the 60-40 had rosin?

Anyway, so after all that and the computer not booting: I posted about what I did. I learned that drilling was a no-no. I then ventured to pull off the two mosfets by the DIMM slots. I did tests on them (not sure I did the tests properly); I posted the results because I wasn't sure what the tests were telling me. One fet had higher readings than the other. Anyway, before I go into that anymore, I need to redo the Ohm test and a diode test (if those are the correct tests?). The fets were hard to pull off, wasn't happy about that.

I don't see that anything is broken on the board. The USB, printer, monitor, speakers, and phone jack are all still intact.

**starfury1** · 03-27-2007, 11:01 AM

Re: Computer not posting

Normally you find 60/40 solder does have flux in it
It does help to have a fux pen or liquid flux
but if soldered correctly it should flow up the leads and look like a volcano with the lead sticking out the top.

You are right to worry about excessive heat too, the last thing you want to do is damage a PCB
(why the junk pcb's and practice practice practice)

again a clean iron and just a bit of solder to get heat transfer then add more solder while holding the iron on the pad & lead.

If done correctly it should only take 2~3 (max) seconds to make a good joint.

on FET removel

others will probably have the own methods I am sure and maybe they could post

best is to work one leg then the other free of the pcb, then hit the Tab, hold the body with a pair of tweezers or small pliers, (dont apply to much pressure they are brittle) add a bit of solder to the tab/pcb and heat it up
Once the solder goes liquid lift it off
( lift the legs so they are just free you dont want a 90 degree bends in them)

get some desolder wick and if you can find it liquid flux

It sounds strange but alot of the time adding some solder make it easier to remove things

flux (too much of it) tends to cause the solder to ball

there are things called desolder tweezers (tongs)
That just make it so much simpler
(go Here to see what I am talking about )

But with any Desolder gear $$$$$$

here's a page I found on soldering & desoldering SMC's

As far as the test goes of fets,
I think they are either going to be working or not.
(as in they will be short circuit if stuffed)

but I'll leave that open to those that do it each day to comment on.

Ok next question time

What type of Iron are you using and what type of multimeter?

If you don't mind telling use

Also if you dont mind
post a pic of your handy work and point out which ones you were having a hard time with.

Cheers

**AKBessy** · 03-27-2007, 12:05 PM

Re: Computer not posting

Thanks for the feedback starfury1. Those links look useful to me ;-).

I am getting another board, so this board that I have been working on is my official practice board. Moreover, one of the local computer store/repair shops is happy to let me have boards to practice on, so someday I should get the hang of this stuff and have a pretty look, professionally soldered board ;-). In the meanwhile.....

My multimeter is a Fluke 29 Series II that I bought back in 1996. The soldering iron I have been using is a RadioShack Cat. No. 64-2060B 45 watt iron with a slanted tip that hubby made for me (original tip was actually a sucker thingy). The solder I used is RadioShack brand again 64-002 E .062 dia. 60/40 Standard Rosin-Core Solder.

One of these days, I am going to order a hotter iron because I can see where I have problems with less heat. For soldering things back on, the 45 watt is okay though huh?

Here are pictures of the areas that I had problems with. No, they don't look pretty, so be gentle with me LOL.

The first one is of the area that the two FETs were in. I had problems getting enough heat into the areas; the FET in the upper right was especially hard with the power strip and DIMM slot brackets - I kinda melted some plastic on both of those.

The other one is of the first (ever) capacitors I tried to remove and put back on. I had a hard time getting the original capacitors off - the solder would not melt - it was weird. Then I figured out to dump some fresh solder on and that helped greatly. Trying to solder new capacitors on in this area was hard - the solder was acting like it didn't have any flux in it. I don't know if this is because I didn't wipe the board after removing the originals? Anyway, the soldering looks pretty funky. Later desoldering and soldering went ok as I started getting the hang of it.

I did practice on some other boards before tackling my project board, but the solder must have been different and "behaving" like I thought it should. When I got on the board I wanted to work on, all of a sudden I couldn't desolder and solder satisfactorily - I wasn't a happy camper and thought, "Well, here's some learning experiences happening!" ;-)

Attached Files

Computer not posting

Computer not posting

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Related Topics