PC@LIVE wrote on 2026-02-22, 13:02:

Minutemanqvs wrote on 2026-02-21, 16:05:

PC@LIVE wrote on 2026-02-21, 15:33:

Following your instructions 🧭, I checked through datasheet the voltage regulator chip 😬, it seems to be very common in PC P3 and Athlon, well it seems that the initial problem, of failure to start, could be connected in some way to this IC, so the subsequent problem, could be due to the definitive breakage of this IC, or something connected to it.
I tell you that the problem could be another, but in my opinion, it's probably him who broke 😑, looking at the pdf file, I extracted an image, where you can see the individual PINs to what they are connected to, I don't know if you are equipped for removal and verification, that is, you have specific tools, but most are quite easy to verify, using a simple multimeter with diode test or in OHm 🕉.
Also from that image you can understand which PINs to check, clearly if you make suspicious readings, you could be close to identifying the problem.

I will have a look at it tomorrow, I have a hot air gun since 4 days so it's absolutely time to do dangerous things to an innocent PCB! These components seem to be available easily on aliexpress for 2-3€ so it might be worth trying to replace it if I measure something suspicious. I'm doing this mostly for the fun of reviving the board, not that I really need it. And I also have a good coworker/friend equipped for micro-soldering under a microscope if everything goes wrong 😉

Interesting 🧐

Keep me updated on the progress and work you will do, I would like you to be able to fix it, I will cheer for you.

So I finally got a replacement SC1155CSW chip and yesterday I did the exchange on the K7T Pro 2. And...unfortunately no difference, so it wasn't the source of the problem despite it getting very hot. At this point I'm calling it a day, this board will go to the "for parts" bin. Really thanks for your guidance and support until now!

Minutemanqvs wrote on 2026-03-08, 11:06:

PC@LIVE wrote on 2026-02-22, 13:02:

Minutemanqvs wrote on 2026-02-21, 16:05:

I will have a look at it tomorrow, I have a hot air gun since 4 days so it's absolutely time to do dangerous things to an innocent PCB! These components seem to be available easily on aliexpress for 2-3€ so it might be worth trying to replace it if I measure something suspicious. I'm doing this mostly for the fun of reviving the board, not that I really need it. And I also have a good coworker/friend equipped for micro-soldering under a microscope if everything goes wrong 😉

Interesting 🧐

Keep me updated on the progress and work you will do, I would like you to be able to fix it, I will cheer for you.

So I finally got a replacement SC1155CSW chip and yesterday I did the exchange on the K7T Pro 2. And...unfortunately no difference, so it wasn't the source of the problem despite it getting very hot. At this point I'm calling it a day, this board will go to the "for parts" bin. Really thanks for your guidance and support until now!

Hello, well it seemed that there could be some possibility, that by replacing the SC1155CSW chip, it could work again, unfortunately it didn't work, but finding where the fault is, without adequate professional equipment, also requires a lot of luck, so in short it can be done, but only if you have free time.
So in short, using it for spare parts, to fix some other motherboard, could be a good idea, but for some components after removal you should make sure they work, if they don't go you would have found what was broken!
Anyway no problem, I can help you if you have other motherboards with problems, just ask if you need, and thank you for updating me on the progress, lately if you have read the last pages, I too have had cards that no longer work, one day maybe I will understand exactly why!

For the motherboard PCCHIPS BKi810 V3.3A, or equivalent with another brand, with S.370 and Intel i810 chipset, unfortunately will have to wait quite a while, because checking the electrolytic capacitors, I don't have the 1500 uF 6.3V replacement ones, and soon I don't have any orders scheduled, we'll see maybe in a few months, but having suspended the purchase of another old HW, to date I have no reason to buy spare parts.

But anyway I have a more than enough supply of motherboards to repair, I don't know exactly how many there are, and also including those that are working but with small problems, for example the defective L2 cache, I have a lot of work, for a few years, since I do these jobs in my free time.

I even have some motherboards to try, which could be working, some I repaired a while ago, others instead are to be tried and see if they work, probably in the next few days I could start to see if the doubts (if they work?), trying some, in particular I would try a LS S.462, an Acer AM2, and an ASUS 775, if I wanted I would have BTX motherboards, but for these I don't have any case 🏘 suitable, so for the moment I prefer those with traditional form factor (ATX).

Having the chance (time), I could try some of my older PCs, in particular from 286 onwards, just the 286 had a problem with the power supply years ago, which failed, I'll see if I can fix it, the problem I think depends on the electrolytic capacitors, because they are very old, it does not have a linear delivery, but the current oscillates between a normal voltage and a low one, then who knows there could be something else, but it is not possible to replace it, because it has a particular shape (very compact), and the connector is not the same as the P8 - P9 of the AT, I think it is XT type.

As anticipated, I will try some motherboards again, most of them I have never tried, and I don't know if they work or not, some I know that in the past were (working), and that have been disassembled to update the PC with something more modern.

The first one I will try is an ASROCK, model 775Dual-VSTA REV. G/A 1.02, already has the CPU installed, I think it's a P4 of about 3.0 GHz (or something similar), in any case at least for now I don't care about the type of CPU installed, but rather that it works.

The main purpose of this test is to make sure that it works, and that there are no various problems, then eventually a reflection will be made, on the final configuration, and on any RAM and CPU upgrades, as far as the video card is concerned, since there is a PCI-E 16X slot, just choose a VGA that is too cheap or low level, in short, nothing too expensive and powerful, since this PC is quite dated.

I would avoid, the use of VGA AGP, which here was a 256 MB ATI 9550 AGP 8X, nothing special, but for a possible CPU update, with the consequent increase in processing power, it could be a little slow, clearly this 9550 would go on other dated PCs, which unlike this one, do not have PCI-E 16X slots.

For RAM, I have the possibility to use both DDR and DDR2, but not at the same time (be careful!), I usually go directly with the faster ones, in this case the DDR2, and I will necessarily use two RAM, to activate the Dual Channel support, as you can read directly on the motherboard, we also read the Dual Core CPU support (Conroe Presler) with FSB 1066, all quite interesting, which should provide a nice increase in performance.

I will update you as soon as possible, with the results of the tests that I normally perform, I have to do some preventive checks, to verify the absence of shorts, but overall I would say that it is very likely to start and work, the motherboard was delivered to me, because it was no longer used, and there was a long time left to get dust.

Here I am, the start-up test of the ASROCK 775Dual-VSTA was performed a little while ago, and the result was what I had imagined.

For the motherboard test, I added a 512 MB DDR2 533 RAM, the ATI Radeon 9200 AGP video card used in the Gigabyte GA-7VAXP, the usual ATX power supply (provisional), PS/2 keyboard, and LCD monitor, plus the usual ISA PCI post card.

I had no problem at the first start, the codes displayed are gradually changed, until a Beep that signals a checksum bad error, probably due to the battery (?) Or to the changed configuration, compared to the previous one.

Let's be clear, I didn't doubt that the motherboard worked, because it had been put away working, but I knew that after several years of inactivity, problems could arise, and currently the only one that could be solved, is regarding the configuration, hardly a PC like this could be used with Windows, except for the more dated ones (let's say up to XP), let's see why:

The 3.00 GHz Intel Pentium 4 CPU is a single-core, even if there should be HT to improve performance, clearly if you want to make this PC usable with the possibility of connecting to the internet, you need an upgrade with a dual core, possibly something that is fast enough, I don't think you can use a quad core CPU, which would have been perhaps better.

The RAM must absolutely be brought to at least 2 GB, or even 4, in both cases the dual channel is activated, but only in the case we use two identical RAMs.

Regarding the VGA, I had already written that you can put a PCI-E 16X, a medium-low level model I think can work properly, of course then you can temporarily use an AGP, but in my opinion it is better to use the PCI-E directly.

For the hard disk, having two SATA ports you can use those, instead of the normal ATA 40 PIN, as far as Windows is concerned, instead I would go from 7 up, or you can choose an alternative OS, if you are familiar with Linux for example, it could be a valid alternative.

Before putting away the 3.00 GHz ASROCK 775Dual-VSTA and Pentium 4 HT 520, I wanted to make some benches, including those of Phil's, which then also serve as a reference for the upgrades that I will do later.

Yes because I don't need another 3.00 GHz P4 HT, rather I prefer to change the CPU with something better, and that it is a dual core, in short, a more usable definitive configuration, even if it is not the best for that socket (T-LGA 775).

I proceed to connect the usual IDE CF adapter, with CF and DOS memory card installed, I start with Phil's benches, some results are quite impressive, especially when compared to those of rather dated PCs, but it could be done even better, since I used an ATI Radeon 9200 AGP, and here is the possibility of using a PCI-E 16X video card.

But we will be able to see this later, when I do all the upgrades I have in mind, which are mainly three, VGA CPU and RAM.

For now, therefore, I have written down the results, I will publish them when I have performed the updates, and redone the same benches, so as to see how much they have individually engraved, I will start with the RAM, then with the CPU and finally the VGA.

I don't know if I'll be able to get an increase of around 50%, at least this is the value I would like to approach, of course if it was higher for me it would be fine, but if it were just above 30%, it wouldn't be bad, but I hope it can be done better.

In conclusion, I already have another motherboard to review, this time it's an ASUS (!), with socket 478, to be precise a P4S800-MX, uATX format with SIS chipset, in the past I tried it but it didn't start, or rather there is something that could prevent it from starting, I want to try again to see if I find the right combination, maybe I hadn't used RAM or tested and working CPUs, we'll see if I'll succeed, and if I'll be able to make it work.

Coincidentally I could have found the problem of the ASUS P4S800-MX, it would be in the PIN of the Winbond W83697HF chip, which would then be the I/O chip, well I have not yet seen the pdf with the pin out, but some PINs from no. 1 to no. 5 or 6, must be straightened, also right among the last, it seems that there is another pin stuck, that is, it seems that it is a metal piece coming from another board.

You also need to check the final PINs of the chip, because you can see the sign of a bump, maybe some are to be spaced out, because I see them a little too close.

Surely this problem can cause problems of BIOS failure to start, or partial boot, for this reason, all boot tests must be postponed, that is, they are postponed to the end of work on the Winbond chip.

At the moment, the motherboard shows few codes on the post card, this suggests that the problem may be only that, and it will be useful for other motherboards that have similar problems, in those cases I would check the I/O chip, which could be suspected of not starting the PC.

If after fixing the PINs, I would have solved it, and I would get the BIOS screen on video, I would have solved it without having to replace anything, but looking better at the embedded piece, it looks like a tiny SMD capacitor, maybe it's the C506 that came off, and went to get stuck between those pins, I'll check with a magnifier, and in case I have to re-sald that tiny SMD.

As far as I can see, what I had written in the end was right, basically the tiny SMD capacitor (C506), came off and stuck between two pins of the Winbond W83697HF I/O chip, this must have created some problems and prevented the correct start of the motherboard (?), momentarily I removed it and kept it, but I will re-solder it as soon as possible, as for the PINs, the neighboring ones I spaced them as much as possible, certainly it could be done better, but I think this is enough to create a sufficient space between one PIN and the other.

The biggest problem now is welding the SMD capacitor (tiny), given the size it would be impossible if it were next to others, here fortunately it is in an isolated position, and easily accessible, it will not be easy, but I could succeed, even if I predict that it will be quite ugly aesthetically speaking, as far as the values and voltages are concerned, I don't have to worry about anything, as I'm sure it's him, I don't know the chances that by disconnecting it can fit (without getting lost), it shouldn't be something that happens often, in fact I think it's very difficult, moreover you have to Consider that he managed to make the trip inside the box, and that he could detach at any time during transport!

The only thing, which would have been better to avoid, was to try the motherboard, but years ago I did not see this problem, I also hope that the previous startup attempts did not damage the I/O chip, theoretically the capacitor should not create problems, different speech if it had been an inductor, but maybe a look at the PIN OUT of the Winbond chip, could provide useful information, on those PINs and if power supply current arrives.

In conclusion I would say that, as far as I'm concerned, I would be moderately optimistic about the possibility that this motherboard will work again, in that case, it would still be a great result, and it confirmed to me that you can often solve the problem, only after carefully looking at the motherboard, unfortunately the most recent motherboards are so miniaturized, that it is almost impossible to find missing SMD without the help of a magnifier.

Recently I had to take care of something else, and the time to repair and try some motherboard, unfortunately zeroed 😣
Starting from the last motherboard, ASUS P4S800-MX I plan to finish the repair as soon as possible, re-seld that SMD capacitor, it will be difficult 😩 but I hope not impossible, I have already spaced the individual PINs of the I/O chip, and this should be enough to avoid contact between the adjacent ones, apart from that, there should be no problems, but a lot depends on when that SMD came off?
In my opinion it was in its place during the trip 🧳, and it came off during the delivery phases, I saw that packages are often dropped (or even thrown), and I think the package took a hit, and something came into contact, causing the damage with migration of the SMD capacitor, which by chance got stuck between the PINs of the I/O chip.
So, in conclusion if the card was working there are excellent possibilities of being able to fix it, if instead it was broken you have to understand what the problem is and if it is possible to fix it, in any case I can hope that everything goes well 😊
I previously wrote about an ASUS ATX P4 motherboard with some missing SMDs in the back chipset, if I remember correctly it was a P4S800, in this case it would be compared with the P4S800-MX, maybe you can be lucky 🍀 and find correspondence between the missing SMDs, the ideal would still be to find an equal motherboard or a manual service, but I have to look at the back of both motherboards, if they are completely different (circuits), I can't take it as a reference.

I go back to examine and compare two ASUS motherboards, on which I worked recently, they have in common the SiS Chipset, in one some SMDs are missing on the back of NB and SB, let's see in detail:

The ASUS motherboard with missing SMD, is a P4SDX 478 ATX format, I kept it almost as a scrap motherboard, and only recently I realized that I could try to fix it, the biggest problem is those missing SMD capacitors, I don't think it's anything else, because they are usually capacitors, however I don't have useful information about it, I can only compare them by eye 👁, and try to replace them with similar ones.

The other motherboard is a P4S800-MX, uATX SKT 478 chipset SiS format, maybe the NB is slightly different, but the SB is the same, here I have to re-seld a detached SMD capacitor, not a simple operation given the size, but I will try to do as soon as I can.

Starting the comparison, I notice on the back of the two SB, the presence of two SMD capacitors, I think the circuit is very similar (?), unfortunately on the uATX there are only those two, while in the P4SDX there is a third SMD (missing here), maybe the missing one needs another chip, and so I have no idea if it was the same as the other two.

Looking at the back of the NB, I would say that it is different, they do not look alike, but I can see that the missing ones would be two, one is parallel to another, this makes the search easy, the other instead is connected to a large track that goes around the RAM, here I can see on the other board how it is, and I see that it is the same as the other neighbors, this suggests that they are SMD capacitors, with the same dimensions, and therefore they are probably the same both in capacity and voltage.

At this point, I would have to look for the same SMD capacitors in some recovery board, it may not be easy to find three, while as far as welding work is concerned, they are in fairly free positions, and therefore it should be relatively easy to weld them.

After putting those three missing SMDs back, I will have to prepare the motherboard for startup, but first I have to put its BIOS chip and all the jumpers, for the BIOS no problem, for the jumpers instead, I would have to borrow them from other failed motherboards.

Continuing with the observation of the ASUS P4S800-MX motherboard, I notice that it has the possibility of two additional USB ports, and that right next to the available ports (USB56), there is the predisposition for two other USB ports (FUSB12), it would not be at all complex to add the missing components, because they are the same as the USB port present.

But what should I recover?

The blue 9 PIN plug, and another black 3 PIN plug for the jumper, then some SMD components, the same as the nearby ones, here for some it is very simple, they have numbers printed on it, a couple instead are capacitors and inductors, here in this case, I should go and recover these from a scrap motherboard, then to finish I should follow the tracks to see if something is still missing ️.

You also need a 1000uF 6.3V electrolytic capacitor, and you need to clear the holes to solder the two sockets and the electrolytic capacitor, but all this work will be completely useless, if the board is not working.

For this reason, I will have to start by re-soldering the missing SMD capacitor (C506), and do a boot test, if the card should work again, in that case I could make this addition, for two additional USB ports.

Another thing to add that could be useful, is the buzzer, here it is prepared but it is not present, for this I think it is quite simple, there are two PINs (+ and ground), and the continuity must be checked with a tester, between the PIN of the buzzer and those of the speaker, usually they are 1 and 4, if we have the direct connection, this would be enough, and just add it!

Another thing is missing, is the second serial port, but here I think it's much more complex, and so I can't say if it's better to add it, because I think we need other SMD components, among which it looks like an IC (probably a buffer), which is located between the two PCI slots, with high welding difficulties.

I have temporarily postponed to the next few days, the repairs of the motherboards, because at the moment I am taking care of something else (including a couple of video cards), but comparing myself with other motherboard owners, in particular the ASROCK 775 Dual VSTA, it emerged that the CPU used for the upgrade is an E6600, here is this point it is not very clear to me if it is a Pentium Dual Core or a Core2 Duo, apparently they look very similar, but they work at different frequencies.

For the upgrade unfortunately I don't have any 6600, but I have both 6700, the C2D E6700 didn't work in the past, but I didn't understand if this depends on the lack of support (on the part of the BIOS), or the CPU is not working (?), the Pentium DC E6700 instead, I haven't tried it yet, and therefore it could be a valid alternative, go on more powerful CPUs like C2D E7X00, maybe it would also be possible, but only after adding the support of these CPUs (in the BIOS), but I don't think we can go further, I would exclude the FSB 1333 CPU support, so The maximum FSB would be 1066, which then in overclock could be brought to 1133 or even 1200 MHz.

Regarding the video card, I'm finishing the repairs on a PCI-E GeForce 7300GS 256 MB DDR2, I'm having trouble finding an SMD inductor, for the VGA output, while for replacement electrolytic capacitors, I just have to decide whether to put something identical as values, or use others of the same size but of slightly different values.

I have not overlooked the work to be done on the other two MB, ABIT S.462 and ASUS S.478, I will see to finish them as soon as possible, then when the work is finished, we will be able to see on the screen if everything went well, we will see the BIOS screen on video, in addition to the post codes scroll on the PCI post card.

Waiting to complete some repairs by this weekend, I saw some videos on YouTube channels on the topic of oldware!

Here I don't watch all the videos, and sometimes it happens that I miss some interesting video, like this one:

https://m.youtube.com/watch?v=_hemUMLIhWw

I didn't understand from the title that it was one of the last motherboards for S.7, and not having S.7 boards with SiS chipset, I had never seen it, but now I can re-evaluate it, and above all I can re-evaluate this S.7/SiS paired, well I knew that those like the 530 were not very performing, selecting 100 MHz, in reality the frequency is only 90 MHz, so in the past I preferred the cards with Ali and VIA chipsets.

But as he explains in the video, the performance depends a lot on the speed of the memories, this is because increasing the frequency increases the transfer speed (MB/s)!

This motherboard was ready for the K6+ CPUs with FSB 133, but they were never seen on the market, if they were out we would have exceeded 600 MHz from the factory, and theoretically the maximum frequency of 800 MHz could have been reached, in any case nothing prevents using the K6+ with FSB 133, using lower multipliers, this makes the K6+ faster at the same frequency.

Well unfortunately there were no further developments, it would have been interesting if by overclocking, you could raise the FSB to 150 MHz or even to 166 MHz, to note also, that what looks like an AGP slot, in reality it is not, but it is a proprietary slot that I did not know existed!

Unfortunately, I don't think that today it's easy to find a similar MB, I'm sure that years ago they were included in cheap batches of motherboards, just to get rid of them, today the prices of a single motherboard, could reserve surprises (not really nice!), I don't think there are any now, because the vast majority have been replaced by more powerful systems, and except for strokes of luck, today it's not very easy to find PCs or motherboards from the late 90s or early 2000s.