I've been struggling with BGA rework for a loooong time before I was able NOT to incinerate everything! :D Thank you for watching Adrian, I am honoured!
Your first two hours certainly weren't wasted. You learnt some things, as did we! Congratulations on a successful reflow. It's always satisfying when something comes back to life!
Great job, especially figuring out that the chipset needed to be reflowed! I never would have even thought about trying to push down on the chipset as a troubleshooting technique.
My Fujitsu Ergo Pro motherboard, is now deadlined post giving it a wash. I was gentle, but... It complains it has no RAM if there is none installed. And then I plug that in, and... nothing. So Tony, your video is certainly an inspiration to get tooled up, and to try and figure out where the issue is. In the meantime I've ordered two replacement DIMMS, just in case the original one went bad. Great video, please more of the same as I have learnt a lot from it :)
Amazing thank you! For your Fujitsu, if it was only washed I'd recommend some good contact cleaner on the RAM slots. I found that the one for switches - which leaves behind a thin layer of lubrication which enhances conductivity - is much better than the dry one. De-Oxit is probably as good. Good luck with your project!
I have ordered a POST analyser card, not needed one till now, but with zero output, its a must. Also ordered the Aishun soldering station you covered. I know its flawed, but it appears to be FAR supperior to the iron I have, especially the heat delivery. Anyway, Ill try the new DIMMS with the POST card installed, and take it from there. BTW, what oily contact cleaner do you use?? DeOxit is good, but its pretty expensive for its can size...@@tony359
In the Doom engine, the game simulation and the video frames are locked together, and each time the game simulation is advanced and a frame is rendered, it's called a tic. The game is designed to run at 35 tics (or frames) per second, so a tic is meant to be 1/35th of a second. When the game runs at this ideal frame rate, time in the game is progressing as fast as as real time. You have 1 gametic (1/35th of a second of in-game time) for each realtic (1/35th of a second of real time). If the computer isn't fast enough to keep up with that, the game slows down. There is more than 1/35th of a second of real time between gametics, so you have less than 1 gametic per realtic. If the computer is fast enough, the game doesn't try to run faster than 1, because then game time would progress faster than real time, so 1 is normally the maximum. But for the test, that limit is removed, and it tries to run as fast as it can, so you can get more than 1 gametic per realtic. So what that message in the video is saying is that it processed 2134 tics (60.97 seconds) worth of game time, but it only took 1459 tics (41.69 seconds) worth of real time to do so.
Well, that was a journey! Thanks for taking us along. After testing for shorts. I always plug everything in when testing, video, keyboard, the lot. So many boards just halt if they don't see what they want to see! 😆 You can make a great DOS machine out of that now, nice job!
Hello Tony - The Amptron MB brings back a lot of memories, the first PC I "built" featured the Amptron PM 7400 motherboard. I used it for years and years and still works (although it's not being used now). Never had any issues with the board whatsoever. thanks for your very thorough testing videos.
My goodness, you are the first hobbyist person on YT I've seen to this date use a PCB heating plate. I've been telling everyone for a while (even some of the larger channel hosts) to use one, because its not just for place and flow - it can enhance your desoldering efforts, enhance your regular soldering (metal-wetting and flow,) and even help make your joints more resilient to potential cracking resulting from cooling too fast. In some cases it might help with older PCB warping and separation. Good practice sir. Cheers.
ahah I'm glad you like my technique! I've read and experimented a lot to realise that pre-heating is more or less a must when working with SMD. I am too appalled when watching much more famous youtubers (sometimes professionals) blasting 500C at a multi-layer board to do the job. I have a video coming clearly showing why no preheating is bad! Thanks for commenting and for watching!
I did the same but it's really necessary - even for small jobs. Yes, you can set the hot air "at maximum" but that's not how it's supposed to work. If you end up with the one I got, please bear in mind it's now been banned from the UK because unsafe (no earth connection which you can fix yourself). My review is here: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Y3bPvus-_as.html It's not the best on the planet but it works. It's just very slow and ignore the temp readings it gives you :) Thanks for watching!
@@tony359 oh i see why you put off buying it.. its a pretty damn expensive investment to fix a $20 board (that might be worth $100 when fixed).. youd have to fix 4 or so boards/cards to make it worth it
I have the same Pre heater. My ground wire on my device was added abouve paint and didn’t make connection. You probably should check if your ground is good if you not already have done it. 👋
Thank you! I had the same issue, you can see me rectifying it on the video review I made here: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Y3bPvus-_as.html I also published a follow up where I add some extra earth connections to be on the safe side. This pre-heater has also been banned by the UK government for this very reason.
Dear Tony, when i see that motherboard? I got so many flash backs of my old job. Did repair those kind of motherboards like Asus, Abit, Aopen (did like that brand back then), MSI. Wish i could turn back the clock and repair them again for the 1st time. Great video Tony, watching this video gives me a good feeling from the old days.
ahah thank you! To be honest, I'm not sure either! I think it's a mixture of nostalgia but also an easier way to learn electronics: much easier to probe a large chip with big pins than a modern PCB where you struggle to see the components :) Thanks for watching!
Good job sir! This kind of problems are typical for the crappy boards from ECS, Zida, Biostar and of course PC Chips who is original manufacturer of this board probably.
If you look at the data of the flash bios showing at 22:23, you'll see at 0001FFFF0, "EA 5B E0 F0 00", which is the reset vector, the first instruction executed by the processor after it comes out of reset. It's a jump to F000:E05B, down lower in the Flash ROM, where it starts to execute code.
Shorting electrolytic capacitors will generate heat and noise on the power line. When you check to see if anything is getting hot, you should check the tops of the electrolytic capacitors as well... but be aware, when they go full short they can get extremely hot!
Thanks very much for your videos. I have several socket 7 boards needing repairs, and thanks to your videos I now know some of the basics to check in regards to this platform. Isn't communal learning wonderful?
oh, I didn't consider that, thank you! I should have touched the chips to check they were floating but my hands were not cooperating... :) Thanks for the tip and thanks for watching!
@@tony359 Even in my younger days I'd probably not have trusted my hands not to shake. I certainly wouldn't trust them now - nor my eyes. Stereoscopic magnifiers are your best friend 😁
Thank you for that great Video. I have found a PC on the Streets, it is an Desktop PC with an Intel Pentuim MMX with 166 MHz, 16 MB RAM, 1,6GB HDD IDE, 52X CD Drive, it has a VX Pro II Chipset, and it runns Windows 95, that pc was almost bever used. The Windows Installation is from 1997 als the Whole PC. Special for me is, it has a AMI Bios called Winbios, and a Souundblaster card. (Vibra_16) ISA I just replaced the CMOS Battery.
I like these lower speed Pentiums. With 486 machines being extremely expensive, a P100 can be under locked to a P33 with jumper settings it runs nearly all 486 based games and Windows 3.1 very well. I find 75mhz is the sweet spot.
Just FYI, back in the day it was always recommended to populate the first PCI with the video card. Not sure what difference it actually makes in real like. Probably to give the video card resource preference.
That’s a good tip thanks! Now you mention that, it rings a bell. Do you know which one the first slot is, the one on the PSU end or ISA end? Thanks again
It would tend to place the video card closer to the North Bridge and therefore electrically better signal quality, but the main reason may have been that each PCI slot has a unique configuration space address, decided by the IDSEL pin, and I think the first slot may have been more consistent in being a particular device ID. It may also have meant a better bus priority from the bus arbiter, as you mentioned.
This motherboard is very similar to the PC-Chips M571. It has integrated VGA, there's a 15 pin connector on the left of the PCI slots, where you can add a VGA header. You can confirm that by looking at the bios, where you can select the frame buffer size up to 4mb.
oh, I was wondering what that header was. I'll check in the BIOS but if it has integrated VGA, do you know which one the VGA chip is? I cannot find any info about the AliM1531 (TX TWO) having an integrated VGA inside.
@@tony359 Edit: that's actually a pin header for USB/ PS2 ports, using an ATX form factor card, check out this video: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-49nIzruszhI.html Now why did they add a Frame buffer option in the BIOS, that's beyond me.
@@juniorbcm5375 ah ok! Some web pages say the chipset is an Ali M1531 and I could not find any reference of a VGA. It would have been cool though! Thanks for your feedback!
Theres no point of using it.. Back it the days it was a low cost solution.. So dont expect high performance of it.. Better go with a faster PCI solution instead.
Thank you! The 866II does that as well. All I need to do is to load the BIN file and then click "verify". The "verify" button compares what is read from the chip with what is on the "workspace". Thanks for watching!
Sir, you just got a new subscriber! :) I have myself an Amptron PM8600 which is of course not working properly... when I enable the onboard L2 cache it's just random stuff on the screen after post. Will do the same as you did. Will look for detached pins between cache and pcb... as a first step toward a fully working mobo.
Please keep me posted! Also it seems that AMPTRON is another name of PC-CHIPS and apparently they used to manufacture boards with FAKE cache chips! I know, it sounds impossible but there is an LGR video showing that! Make sure your boars has good ones - this is LGR's video: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-mZnS-whVuDI.html Thanks for watching!
@@tony359I totally gorgot about this. Yeah i checked the pins of the cache but they are all ok, so i guess it's a bad chip. Meanwhile I got another pm8600 and this one is perfect. Maybe I could try and swap chips, but not really seeing the point, since the first board is in a poor shape anyway... Burnt usb traces (i guess someone connected the usb ports the wrong way) 😅 I also read about the fake pc chips cache yes, but fortunatelly these boards have working chips, usually they fakef the amount (showing 1mb, when there was only 512kb)
Seems like this board had integrated sound on board, because they had the jumper pin for the bracket. But not the sound pro chip. Anyone who knows about this board , because the TX Pro chipset mainboards mostly (not all) had a sound pro(CMI-8330) sound chip onboard.
I think I had one of these boards for a cyrix build I did in 1997. IIRC it failed shortly after my trident video chip blew up, if I can find it I'll have to look at those BGAs. I was under the impression that boards started going solder free much later than 1995, I wonder if I'd kill fewer boards with a bga reflow attempt if using a preheat station.
Doesn't surprise me that the problems on this mainboard were that of a bad soldering, Those mainboards with TX Two or TX Pro chipsets were the cheapest out there. Imagine a mainboard for $35. So their build quality wasn't really up there. They got used in the cheapest PCs, also I remember that they weren't that stable
@@tony359 No problem. If you have some time, check out the Red Hill Hardware Guide, which has a lot of the hardware info of the late 1980s and 1990s. PC Chips made those fake cache boards. I watched about a dozen of your videos, and I like them
Flux. It prevents the solder from oxidising. This way the solder melts better and flows better. It's totally needed for soldering, maybe not 100% for de-soldering but it never hurts. Thanks for watching!
WTF is that chipset? TX TWO? Never seen those in my life, and trust me when I say I worked on a TON of PC's during those days. Who the heck made those?
That's something PCChips did for their mainboards, they had ALi, SiS and VIA supply them re-marked chips. The different iterations of "TX Pro" and "VX Pro" were from each of these companies. SOYO did the same but rebranding various chips to ETEQ. Amptron was a brand of PCChips.
amptron was another division of that crappy company pc chips the company who was known to sell motherboards with fake cache chips. amptron had awful c/s i had one of there socket 462 boards back in the day that failed and not only did i have to pay to send the board in at my cost they made me pay them $30 to repair the board even though it was under warranty they said this was a fee the charged everyone for warranty funny now other manufacture had such a charge. then after all that they sent a board back with dead ram slots.
Ah, amazing! I've heard of PC Chips only a few times and every time it was such a negative feedback! How can you possibly sell boards with fake chips on it??? Lol Thanks for watching!
@@tony359 ya it was back around the 486 era the funny thing to the fake chips are soldered to pads that don't even have traces connected to them so u couldn't even put real chips on them. i guess they also did some tricky in the bios so the system would even claim there was cache when there wasn't. pc chips lasted till around 2005 and then they were bought out by ecs who are still around they are another low end motherboard company. a few other utubers have done videos showing off those boards with fake chips.
@@tony359 idk there is a lot of scummy companies out there who have no trouble with lying to there customers for profit i guess. in more recent times look at gigabyte with there exploding psus that they tried to sweep under the rug rather then just do the right thing right way they had to be called out by multiple large tech utubers before they recalled them.
