win xp hangs at BIOS boot menu screen

Gateway / 501s
April 23, 2010 at 15:39:56
Specs: Windows XP, Pentium 4, 1.5 GB mem
System takes 10 to 15 restart attempts to boot to windows menu. It hangs on the boot menu (F2 or F12) screen. When it finally starts it only stays on for a couple of hours & shuts it's self off. I replaced the hard drive, battery and reinstalled windows twice, no joy. Any suggestions would be much appreciated. I tried it w/o SP2 & SP3. Checked all cards and connectors. PCdoctor found no hardware issues? It doesn't freeze, it just turns off. Event log doesn't register anything until it boots to windows.

See More: win xp hangs at BIOS boot menu screen

Report •

April 23, 2010 at 16:02:59
This sounds like it could be two things, dodgy memory or the PC is overheating

For the first, if you can get a copy of memtest and let it 'do it's thing'. If it finds any problems you could try cleaning the memory contacts (carefully, Ram is VERY delicate if you dont know how dont try it) or replacing it

The second thing that could be causing this is that the computer is overheating and is shutting down to prevent damage.
If it is this, then just open up the side of your computer and clean out the dust. You must make sure that heatinks and fans (such as areas like CPU, Northbridge, Graphics) are all free of dust to help flow of air

Hope this advice helps!

Report •

April 23, 2010 at 18:14:26
Thank you for your quick response. I did remove & replace (clean) mem cards. I'll run the test again & copy the results. I thought about the over heating. I left the case open & both fans seem fine, didn't feel any unusal hot spots (is there a way to check that other than what I've done)? It does this shut down thing when I first start up (no heat at all). It shuts down repeatedly until 15 or 20 tries at rebooting & then stays on for awhile. I am wondering now about the power supply or a possible short?
I'm ordering a PCI Post Card to help diagnose. Could a Bios problem cause this?

Report •

April 24, 2010 at 02:44:29
Other things... Unlikely but not impossible... the PSU may be intermittent - assorted voltage rails not always as they might be.

Cracked or dry joints on the motherboard.

Possibly and a very long shot... the cpu has overheated somewhat in the recent (or not so) past...

An obsucre fault on the motherboard itself may be the cause too; though quite what etc...?

One item you might try as it would eliminate the HDD controller on the motherboard.; "borrow" an add-in HDD controller card and see it that resolves it;. Connect hard-drive etc. to that card - and thus bypass the on-board controller(s)... One might buy one to try etc... and if no better perhaps negotiate to return it?

Report •

Related Solutions

April 26, 2010 at 13:48:11
I do have a print out and description of everything inside my computer. I'm not sure what you mean by HDD controller card. I've identified the cards installed & don't know what you mean. Could you elaborate? It is possible it may have over heated at one point but I can't imagine when. I live in a rural area & we do experience brown-outs. I have always used a surge protector (but they can fail). I have access to another computer and could swap parts (if compatible). Let me know what the HDD card is & I'll try it. Thanks

Report •

April 26, 2010 at 15:10:38
HDD controller - Hard Disk/Drive Controller is usually built into the motherboard and all you can see (easily) of course is the ribbon connection points...

An external card is of course obvious - it sits in an expansion slot; and again has the usual ribbon connection points clear and obvious.

It is not usually an easy item (or recommended) to replace or repair the one on the the motherboard - if that one has a problem; instead one uses an add-in card...

So if the on-board HDD controller is failing - or has failed (or one has a suspicion...) then one can often (usually....) simply install an add-in card to replace its functions; there being nothing to do in the bios to enable it take over from the motherboard HDD controller..

Remove all power into the system; install card as per instructions that come with it; connect up the drive - as in remove the ribbon connections from the MoBo (motherboard) end and transfer/connect to appropriate sockets on add-in card... Restore power and start up again... (and maybe pray to the deity or spirit of your choice/preference...).

So if you can borrow one - even buy one and negotiate to return it if not needed... (one could do with Compu$a in the olde days... and possibly Frys, Best Buy or similar will allow it too) then likely it's an item to try and test out the system in that regard?

If problems persist then it's a reasonable conclusion that it's not the controller... If problem goes away (one lives in hope...) then logically it is an issue with the on-board controller and thus one keeps the add-in one?

In fairness I've only ever been aware of an on-board controller fail perhaps once... They are generally pretty solid devices.... but if you get brown outs and thus also likely surges too... who knows? Regardless it won't hurt to test out the system in that regard - if possible...; although being out in the sticks/country...???

Incidentally unless you have a high quality (professional standard, fast acting and high surge rated) surge protector... if it has been hit even once - and one hopes protected things against a serious power surge or lighting strike... - then popular wisdom is to replace it with anew one. And then if that gets hit... replace it too, and so on... Domestic and cheaper protectors are essentially regarded as one-shot devices from some by the reviews and advisories I have read in the not so recent past...

Report •

April 26, 2010 at 15:32:20
First run Memtest to test your memory for failure. Rerun if failure for each stick of RAM separately to find problem one.
Second, install a program called Speedfan. It is a program popular with overclockers, BUT it is just a tool and it can give you direct readouts on CPU core temperature, GPU (graphics) core temperature, at least 2 other motherboard temperatures, as well as voltage readouts at CPU and RAM at least. You can also set it to graph some of these over a period of time (like while a program is running).
Between these programs, you can determine whether the problem is Memory related, Heat related, or some of the possible power related problems without tearing anything apart or spending any money.
IMPORTANT: If you live in an area that is prone to brown outs, purchase a UPS. It is a battery back up system that will protect your computer from surges, power drop outs, black outs, and other related power problems that can cause many problems with your computer. It will sound an alarm when the power is off or low and has switched to battery, and will give you typically 10 to 20 minutes to shut down properly and the better ones come with software that can shut down for you at a specific % of remaining battery if you are not nearby to do this.

You have to be a little bit crazy to keep you from going insane.

Report •

April 27, 2010 at 13:52:13
> System takes 10 to 15 restart attempts to boot to windows menu. It hangs
> on the boot menu (F2 or F12) screen. When it finally starts it only stays on
> for a couple of hours & shuts it's self off.

I can think of 30 different reasons for this problem - because I also understand how to design this stuff. For example, you power supply could have always been defective. And the defect has now become severe enough to be detected by another power system part - the power supply controller.

One item that can make any other part look defective is the power system. Therefore informed techs use a 3.5 digit multimeter and one minute to obtain the necessary numbers. A tool sold in most any store that also sells hammers for less that the price of a good hammer. Or borrowed from a friend. Numbers that quickly identify the problem. Or that make possible replies from the few who actually know this stuff. Without those numbers, the informed will remain silent. Without numbers, you only answers will be speculation – not solutions.

Set to 20 VDC. Black meter probe attaches to the chassis. Red probe touches each wire where that power supply wire connects to motherboard (probe inside its nylon connector). Touch any one of each wire (red, orange, yellow, and purple) to read numbers are as the power switch is pressed. A three digit number for each will report (without doubt) the state of numerous power system components (including and not limited to the power supply).

Only then will other diagnostics accurately report on other system components.

Definitively not associated with your symptoms is memory or other motherboard semiconductor heat. All those semiconductors are at ideal perfect temperature even when the room is at 100 degrees F. Semiconductors heat does not explain your symptoms.

Provide numbers so that the few with superior knowledge can exonerate so many parts immediately (and without doubt). And then can list the next most likely suspects to move on to. Anything can appear defective if those voltages numbers are not known.

Report •

April 27, 2010 at 20:27:23
Thanks everyone for your help. I'm disabled & therefore have the time to research this problem while learning a great deal about the inner workings of my computer. I have felt since the beginning that it was somehow power related. Never bought into the overheating as once it booted (after many attempts) to the desk top I could run any program, shut down & restart with no issues. It would eventually shut off. I've inspected the board with a magnifier & can't see anything suspicious. It is very clean and dust free. I'm going to try the multimeter for a start (less scary). UPS backup sounds like an excellent idea. I can check the memory as mentioned. I don't believe it's the problem but it's easy enough to test. I'm checking into the HDD controller card as well. It's given me a lot to think about & I believe in starting with the basics first so multimeter it is. I'll let you know what I find. Thanks again.

Report •

April 27, 2010 at 20:57:32
It sounds like a good plan. The numbers CAN be very useful if you know what is correct and what is not (though I do not think that the very superior attitude of that respondent is in any way helpful. I have seen the same post, probably saved and pasted, before. I do not know whether he actually did a follow up with detailed advice (though I hope so), since I was so turned off by it that I did not bother to look back at it again). What would have been helpful would have been a chart or list of correct number ranges and key indicators.
I would not purchase a hard drive controller card unless you had an indication that this is the specific direction to look, or you already eliminated power, memory, and heat problems first since they are pretty reliable. Know that power surges and even sudden voltage drops can have lasting effect on some internal components, including the power supply, memory, motherboard, and hard drive.

You have to be a little bit crazy to keep you from going insane.

Report •

April 28, 2010 at 10:15:09
I suggested a HDD controller test if possible (test by substitution) as it's a kwik and easy test? And is fair to say they seldom fail; but not entirely unknown... If one is in the major cities etc. it's usually easy to get an add-in card (fits into a standard card-slot, and ribbons go to that instead of to motherboard) and which can be returned if not required? However out in the sticks... may not be so easy?

Our "self-proclaimed Expert" above is correct in that voltage measurements can help determine what is amiss - if it is in fact related to the power-supply (psu) - but often sed volts can be within tolerances and thus not tell you anything as result. Also that it might be the psu itself and may have been less than purrfekt from the start... But that would be well nigh impossible to prove (duff from the start); although it could have started to fail over time of course? Equally it could be just one part of the voltage distribution that is faulty; OK leaving the psu - but not so further on; a voltage check can reveal this. After-which one has to determine the what/why etc...?

Likewise it could be several other items too - viz:
failing capacitors - as any "expert" will know (and usually the rest of us mere mortals too...) if a system will appear to work OK after being power-cycled a few times - often just once or twice this "can" (but not always) be a pointer a leaky or failing capacitor (usually of the tubular etc. style). The application of volts provides a momentary surge of power through the component and thus "heat" is generated (albeit not much - but enough) and the leaky item reseals itself - for a while; then starts to leak again and the fault(s) will re-appear...

voltage-dependent capacitors can do somewhat similar too; appear to be OK (for a while) after being hit as it were with applied volts etc..., and then start to fail.

"some" - but not all - semi-conductor devices can/will exhibit similar effects too... As can RAM chips on occasion (I put these apart from standard semi-conductor devices as they are usually in an IC (integrated circuit) form rather than discrete etc. components of a semi-conductor nature...

logic-ccts (the actual chips themselves) can also produce similar effects; and these often do not show up any significant voltage errors. A logic-probe is then useful here so as to check the "0"s and "1"s (the on and off points and outputs etc.).

cracked resistors - hair line cracks usually invisible to naked eye - can also produce similar effects to the above.

dry joints (poorly soldered connections) or hair-line cracks on motherboard circuitry - the actual copper/alloy traces on the motherboard and component solder points thereon can also produce the same effects...

the psu can exhibit all manner of faults that embrace any and all of the above; not the least the leaky/failing capacitors; and also the solid-state voltage regulators...

Really the list is almost endless....; and I mention the above so you have some idea of what can be involved in "searching at component level for a fault...

" Voltage testing is no more than a starter... It may be a simple cause - the psu is duff (faulty).; it may be something much more obscure...? Certainly test (measure volts on) the psu - on load and off-load; i.e. nothing connected to it and then with everything connected...

Having also designed (and serviced) a wide range of kit over 50yrs or more... starting with valves (tubes in USA/Canada/Japan etc.) and working through transistors (never really feel comfortable with their theory...) and then onto digital (logic ccts etc... for a while); and that kit being anything from ham radio, domestic radio/tv - broadcast and AV kit; and a decent spell with IT kit... all of the above "possibles" have been within my experience...

It is I think fair to say that most IT servicing is usually done on the basis of replace suspected components - i.e. plug-in items (assorted cards etc.); removable on-board components/chips; the psu etc... even interconnecting ribbons and the like... This as usually it is real pain and not cost effective to go hunting for obscure faults at component level. If testing components by substitution - one by one with know to be good items - doesn't reveal the cause then usually it's the motherboard - (and the psu, hard-drive and RAM are also part of the test by substitution route); and this again can be proven by trying another one (if one is fortunate to ahve one to hand..) and re-installing all the original add-in components (cards, RAM etc.) that don't come as part of a new board?

And if one is going to hunt around for voltages various... very useful to know exactly what they ought to be wherever one checks. Thus one really needs to correct service manual as a starting point and reference for the journey...

The typical psu produces a few fixed voltages; and these can appear OK initially when system is cold; then start to wander as it were when it warms up... This could be normal - within reason; could also be indicate a faulty psu itself - the voltage regulation is failing, or smoothing ccts. are going down; or there is something on the motherboard or add-in components drawing more current than "it orta" and thus pulling down the volts on that rail...

Again one needs to know what "ought to be there on each psu voltage out point; and at appropriate places on the motherboard. So again one really needs the manual - or decent repair guide that will usually list the standard volts on most boards...

What is the make/model of your system; and specifically - if possible - the actual motherboard?

Scott Mueller has good books; as does Mark Minasi; both have enjoyed a high reputation earned over years of practice and professional expertise...

Incidentally I know of a million $ broadcast vision-mixing (switcher) system (brand new) that had a habit or freezing and/or rebooting itself every midnight or thereabouts - all on its own. The company who designed/provided it (a major player in the field) tested by substitution - after reading assorted logs generated by the kit; couldn't resolve it that way; finally replaced the whole main desk system. If that hadn't resolved it then the associated computers would have been next. Fortunately it was within the main-frame of actual desk-panel/switcher itself as the replacement unit is OK....

Television switchers (vision-mixers) are essentially bog standard computers; usually running on XP-Pro, or Linux, or - until recently - W2K; with a fancy set of controls to use/activate/run predefined software routines...

And finally (as per the late Walter Cronkite) like many here I do not claim or profess to be an expert in any field. Like most I have been around and done my journeys, travels various, and some of them include the above... Just adding and sharing some of my experience; and open to modest debate/discussion so as to learn more...

Report •

May 13, 2010 at 15:37:02
Problem solved! Power supply to blame. Evidently it was failing, SLOWLY. Since my first inquiry on this site, I could no longer get it to boot for more than 20 -30 seconds & then it would shut off. I could only test the 9 pin, purple wire with power off but plugged in. It should've registered 5v, instead I got 4.77. Took it to a guy locally to swap P. supplies. With a new Power supply installed all is great. I only wish it would have occurred to me earlier to have it checked. Six months is a long time for a power supply to slowly die, isn't it? Thanks everyone for your help! I've learned from all your input.

Report •

May 14, 2010 at 06:59:30
Thanks for posting back; so seldom happens.

Don't beat yourself up over it; unless a psu is seriously underrated for the load on it..., as in not able to supply the current demanded, it's not often they fail - but sometimes they just do...

A psu can slowly die for numerous reasons; many listed in an earlier post.

As before, typically it can be the smoothing ccts. (slightly larger than elsewhere capacitors within the psu - from the diodes to the actual output - start "to break down and leak"); the actual voltage regulator (a chip) itself being unable to sustain the current drain/demand on it (semi-conductor devices can often have a gradual failure - almost like a slow but persistent breakdown overall. That latter can also go down for other reasons too...

The actual rectifier diodes in the early part of the psu cct. likewise can start to fail - almost like a small leak or phsyical breakdown within themselves and it gradually becomes more than just a wee leak...

Rectifier diodes as used in alternator ccts on cars/automobiles can show a similar effect too...; which is why one "may" start to see the charge/discharge light for the battery starting to (initially) glow dimly... then increasingly brighter...

Overall what was amiss with your psu can only be determined by closer inspection - at component level etc... And for many that is a non-starter; which is why increasingly "test by substitution" is often easier (and in industry - cheaper/more efficient overall).

Report •

May 14, 2010 at 09:15:35
> Evidently it was failing, SLOWLY. ...
> I could only test the 9 pin, purple wire with power off but plugged in. It
> should've registered 5v, instead I got 4.77.

A 4.77 volts on the red or purple wires suggested the supply was completely defective. However numbers from other wires were necessary to say with certainty.

That 4.77 number is within written tolerances. But when including how meters work, then that number says a defect is identified. Meter is the fastest path to a solution because it reports so much information in one number.

That number also implies a defect existed many months ago. Defect may have been observed even when the system was first constructed. And slowly got worse over six months. Your symptoms are typical of a supply that was defective when first built. And that slowly gets worse to create a failure even one year later.

Multimeter numbers are definitive. For example, 4.80 volts says a defect exists - without doubt - even though that computer booted. Those numbers contain information that others do not realize. And why better shops always measure those voltages before a system goes out the door. One failed item identified by a bad multimeter number, for example, is a failed capacitor. That or many other problems could have caused your failure. One number definitively said something - a tiny list of supects - is defective. Pointed right at the problem without speculation. Hopefully the shop confirmed that by measuring those voltages after a new supply was installed.

UPS to protect hardware is nonsense. A power supply must provide a constant and rock hard 5.0 volts even when incandescent lamps dim to less than 50% intensity. A functioning power supply makes such AC power variations irrelevant. Protecting unsaved data from blackouts is the only purpose of a UPS. A UPS does not protect hardware. It only protects data. A UPS would have done nothing to avert your failure or correct it.

Report •

May 14, 2010 at 11:06:39
Agree that in the hands of those who know how to use one and interpret/understand what they measure etc. a decent multi-meter "is" a definitive way to go... However not "everyone" is suitably savvy there...; even if they have one.

A kwik test by substitution is often a very easy way to determine if a psu is duff... (providing one has one to hand - or ease of access of same...).Most IT service outlets go that path; kwiker etc. than poking around inside "some" systems.

Once it is known to be duff, dying - even near unto total death... it is of limited academic interest only (and to a small number usually) to know the what/why etc. But it it is also true that for that "special few" it can be a cheaper option to to repair it, and well within their remit; but not for all.

A ups would be useful in a unstable mains voltage supply; as some kit just doesn't like being surged up/down and down/up... If those surges (drops) are "significant" the psu may not be able accommodate them and thus delivered psu volts out may drop too low and things become very unhappy inside the system...; This just as a major surge/increase could do some damage - depending on quality of components/tolerances etc within the early stages of the psu...; these days some capacitors especially have very small tolerances etc. compared to days of olde...; and some voltage stabilisers are less than they might be (some kit "is built very cheaply"...) A ups at least helps to reduce or minimises those side effects; and does allow a chance of safeguarding a system - proper shutdown and - as stated in posts above - of course data safeguards...

If this wasn't the case then why do many major IT facs have them - not just to preserve data only... I know of a brand new HP/Compaq workstation (not on a ups mains feed) that died a death after an improper shutdown - actually due to failed standby power (mains to diesel generators) switch over test... Our IT depot advised that the HD did not take it kindly and neither did the MoBo either - an odd combination of failures nonetheless... Was this a duff system from the start; or was it the improper shutdown - a crash due to loss of incoming/correct mains volts (on this occasion - there weren't any!).

And there are still some bits of broadcast and related comms kit that really "don't" like brown-outs and up-surges in general... In the past it was often cameras themselves; but today they (usually being ccd/cmos etc.) are generally (but not completely) less vulnerable in that regard... but the list does still exist...

That the psu was possibly duff at start is of course a very distinct possibility; but obviously impossible to verify at this time. Equally it may have been inadequate for the demands made on it; either with system as shipped, or possibly if other kit was added to load it... Overtime "the strain" might have been too much for it (it's regulators etc.) and it simply gave up the ghost... Dell made one model of Dimension a little while back that had a totally inadequate psu for systems as shipped. Frequently (but not always) they died as they were... But if one added even more drive (HD, cd/dvdrom) it was almost certain to kill the psu... They later up'd the specs for subsequent psu for that model - most of them going out as replacements... There was whole slew of posts on forums (fora?) various about that model and its psu. Didn't do Dell any favours either as they effectively denied there was a problem; and yet they up'd the specs. for the replacement psu...

The saving grace on desktops and similar systems is that often one "can repair" a duff psu; certainly it can be easily replaced - with a standard pre-built unit. Whereas with many laptops it's not really practical - too time consuming, less than easy access etc if there major parts on the MoBo; and of course the usual "mains-adapter" is a sealed/potted module increasingly...; and thus effectively throw away... (not good environmentally...)

Report •

May 16, 2010 at 09:13:23
No multimeter numbers means only the least technically informed can reply. When the multimeter is used, then the most informed can reply. And reply with definitively - reply without doubt or speculation. In his case, the multimeter identified the problem immediately - in less than a minute. Multimeter provides the least informed tech with knowledge from the best.

All electronics must love being "surged up and down". That is called normal power.

Voltage never goes too low except when incandescent bulbs are at less than 40% intensity. As the electrically informed know, if voltages are dropping that low, then electronics are happy. And other appliances - refrigerator, dishwasher, air conditioner - are at risk. This paragraph contradicts popular myth. But then even A+ Certified Computer techs need no electrical knowledge to pass that test.

Serious locations such as Telco COs have a UPS because power loss is not an option. Too many 'so called' professionals are only programmers. Too many 'programmers' never learn what hardware - i.e. a UPS - does. UPS located at the service entrance typically does more than just provide backup power. The plug-in UPS adjacent to electronics only provides backup power. Anything else it might do is already inside the electronics. If in doubt, then post specification numbers. If making those claims, then you first learned from those numbers. Post them.

A UPS recommendation was obviously useless to the OP when first posted. Obvious to those with basic electrical knowledge. Easy for most anyone to understand. Are incandescent bulbs dimming to 50% intensity? If not, the UPS would do nothing useful. If not dimming, the UPS leaves a computer connected directly to AC mains where it still 'suffers' from 'surges up and down'.

Yes, when not in battery backup mode, a computer is connected directly to AC mains. And when in battery backup mode, a computer must 'clean' some of the 'dirtiest' electricity it will typically see. More numbers: this 120 volt UPS in battery backup mode outputs two 200 volts square waves with a spike of up to 270 volts between those square waves. Which electricity is 'dirtiest'? That from the UPS. And made completely irrelevant because all power supplies 'clean' even the dirtiest electricity from a UPS.

See that 200 volt square waves and up to 270 volt spike? Potentially harmful to small electric motors and power strip protectors. And perfectly ideal power for all computers.

Testing by substitution - also called shotgunning - is a symptom of techs that never learned how electricity works. Even automobile mechanics find themselves quickly unemployed if shotgunning. Substitution is replacing many perfectly good parts until something works. Takes longer. Violates a well proven rule: "work smarter, not harder".

Any properly trained computer tech should know how to use a tool that I even give 12 and 15 years olds to use. A meter. A tool so pathetically simple that it is sold even to K-mart shopper. Sold in most stores that also sell hammers. Feared only by those with so little technical grasp as to shotgun - and therefore take longer to fix things. Shotgun. Therefore not eliminate defects until AFTER those defects cause unnecessary failures.

OP touches a multimeter meter to his power. In less than one minute, the defect was found. He only swapped a defective part - nothing more.

We have a serious problem. Many techs have virtually no hardware knowledge. Even fear a tool sold in Wal-Mart and Lowes. Foolishly use power supply testers. Never learned why so many Indian and Chinese immigrants are now hired in the Silicon Valley - because so few American computer geeks can even use a multimeter.

A consumer magazine created simple problems in computers. Then took those computers into shops to be repaired. Immediately, many techs would replace the power supply, etc. Shotgun. Had no clue how to find a problem before fixing it. Did what only the least trained would do – shotgun. Too often never solved the simple problem. We need more immigrants.

OP quickly identified his failure faster because he used a multimeter.

Report •

May 16, 2010 at 10:26:28
We could keep this going forever...; but it isn't worth it.

I agree totally with the comments re' consumers magazine tests and the like. (they apply to more areas than just IT as well...) Having had to deal with the aftermath of some of those outfits I know what it's like.. And also that many today feel that servicing is simply switch out parts/pull and replace cards various - until you find the duff one...

A discussion with a Bell Canada/Teleglobe senior supervisor some year back in the Laurentians (N. of Montreal) was about just that. He was having to teach his "engineers" how to solder properly; let alone do anything more that switch out a card or three... But that is the approach most outfits now use; it's cost effective; and you need fewer "experts" around.. And the few you do have can do the bench/component level repairs later on the kit/items removed...; and those "experts" usually cost more too...

Nonetheless - useful to remember that most - but not all - who come to forums such this have little or no bench/component level knowledge of electronics. And many fear attending to anything that has possible need to go into the innards of even a basic computer (whatever style).

Similarly there are "many" well/highly trained techies who are more than able to do the test meter and other test kit approach on all manner of kit.. But when time is of the essence - and in many arenas today that "is" the main consideration - get whatever it is that's down back up and running again - NOW - is the attitude... I know that from daily experience...

Thus today many highly qualified experts will resort to test out by substitution of possible parts; and later will investigate and repair the failed item. This is standard approach in many telecom, broadcast - and IT - areas - today. It may not be the "ideal" approach favoured by some "experts"; may not satisfy the purist (and I can be one too...) - but... It is cost effective; reduces down-time etc. Get the ship sailing again "now"; fix the duff part later when you have time and the pressure is off; and then you have a spare unit again...

Again many here are not really comfortable with testing with meters (and logic probes and the rest..); thus they are happier and more comfortable with a test by substitution approach - when it's evidently easier and more practical for them. Which is why frequently is suggested in forums/fora like this. If one is out in the sticks etc., or it's out of hours or whatever - then yes one is more likely going to have go inside and measure, check etc.. and even then one may not find it easy... Having repaired a conference (recording) tv system in the early 70's in Austria - around midnight - and I had no recourse to a "decent" volt meter, the one to hand (from a local villager) had an appallingly low impedance, was very cheap and known to be seriously less than accurate - and thus its measurements were very suspect...) - the only recourse was to test by substitution. This by borrowing parts from other kit that would work (and that kit wasn't required just then). I know what is ideal, what is possible - and what is often practical... Also fair to say that I was drawing on a lot of previous experience too; and thus was more than able to ascertain what was up with the kit - even without a decent meter at least.

If you were at all familiar with certain makes of tv camera in the 70's you knew instantly what was duff in them when they went down; they were known for it - no meter or test kit required... It came under the heading of "experience and known defects..."

So again please try to accept that while "you" may feel very superior and very happy to tell others to "use a meter and other kit" - first.. not all of the punters "out there" are happy to do so. They are happier with a simpler - building blocks/test by substitution approach... It's easier for them - and they feel safer. Who knows - in time - they may even feel OK to go the more technically correct approach; but give them the benefit of doubt and also time to learn?

As regards voltage surges...; I can recall when (in Europe/UK at least) anything more than a 1-2 volt drift could and would cause all manner of kit to become very unhappy... even with stabilised psus on board the kit.Voltage (mains/line) surges often are accompanied with a frequency shift too; and if you were to chat to RCA and Ampex VT engineers from the late 50's into early 60's they would tell much more of that. That kit was very dependent on stable main-volts "and" stable mains frequency; and even a minor drift in frequency "could" have serious consequences... And with a drop in frequency (sudden or not - and sometimes they were "rather sudden") came a drop in volts... As one from that period I can speak from personal (and group) experience...

That our poster managed to use a meter to good effect is fine and excellent for him. Equally it might well have been that he didn't succeed or feel comfortable with that approach; so then what would he have been expected to do? Pray to Whomever, whistle dixie, weep, hope any one several masked/caped heroes came by - all of whom"were" experts with test kit - which they also carried with them? I doubt it; likely he would have acquired a known to be good part to test with - if at all possible?

And finally - it sad that so many open the case on a PC (and other kit) and start removing parts - with the mains cable still attached and "powered up" at the wall... They forget (if they ever knew) that on post AT systems - the on/off switch/button at the front "doesn't" kill the mains power into the box. For the benefit of those who might read this in the distant future...; on post AT boxes unless one removes the mains power entirely from the back or side of the box... there are still volts circulating around the MoBo etc - as the psu is still active... Sadly so few ever bother to remind posters here (and elsewhere) to first remove all mains-power (and battery power too on a laptop etc.) before pulling stuff out or adding things in... Often they get away with, sometimes not...

Time for dinner...

Report •

Ask Question