PCChips M918 Motherboard Repair
ฝัง
- เผยแพร่เมื่อ 25 มิ.ย. 2024
- I have purchased a Socket 3 motherboard which shows "no signs of life". Will I be able to repair it?
Joyalens/Andonstar microscope review: • Joyalens JL249S-M / An...
Yamaha YMF719E Sound Card Repair:
Part 1: • Yamaha YMF719E-9 sound...
Part 2: • Yamaha YMF719E-9 sound...
@LGRBlerbs video on fake cache: • M919 Fake Cache and Re...
The RetroWeb page: theretroweb.com/motherboards/...
00:00 Intro
00:37 Overview
04:30 Jumper setting
06:57 Power up
09:21 Microscope trace repair
13:45 Power up 2
15:10 PCI troubleshooting
15:57 Chipset reflow
22:47 PCI Troubleshooting part 2
27:18 Testing a different PCI card
29:03 Battery holder replacement
29:30 Clock troubleshooting
35:50 So much time wasted!
37:12 It Lives!
37:48 Final testing
39:47 BIOS dump
40:40 Outro - วิทยาศาสตร์และเทคโนโลยี
I watched half the video knowing that the problem was the jumpers 😄 Back then, making a PC work was very difficult, compatibility between chipsets, CPUs, and various cards was a lottery. This video is fantastic, highly enjoyable, and perfectly understandable! Very informative!
Thank you Sir, indeed we forgot how cumbersome setting up a system was back in the days. Particularly with low quality hardware! Thanks for watching and for your kind words! They have a special meaning coming from you! (Che hai finito di taglia’ la scatola con il seghetto? 🙂)
those early motherboards looks deceptively simple.. but... one wrong jumper or component badly seated and is just silence or misleading beeps ;)
Kudos to you for not hiding your mistake! Just discovered your channel, watched a handful of videos and got to say I'm loving the content. Love the fact you're taking us all on your journey of learning.
Welcome! And absolutely, I'm not here to show off! I'm glad you like watching me breaking things :)
Years ago, I found the perfect use case for PC Chips motherboards. I lined my dumpster floor with them.
I'm not sure they would be useful for that purpose either :) Thank for watching!
Great Trubleshooting.
Wait a minute, I published this 5 minutes ago and the video is 41 minutes long! :) :D
Yay! New Tony's video! See one click one, its as simple as it can get.
I hope you like it! :)
The strand of shielding worked great to repair the trace.
Its alot cleaner then the enameled wire and I am certainly going to give that a go when ever I need it.
Very good fault finding again and I really enjoyed the video.
It wasn't my idea! :) But yes, much better than enamelled. Now, what do I do with those three spools of enamelled copper wire I got? :) Thanks for watching!
Time spent having fun or learning stuff is never wasted time.
Absolutely! :)
Great video. It is still good to see troubleshooting.
Thank you!
Tony , your videos are always so funny and inspiring! Thank you for showing us here all your triumps but also all you failures
Every time I atempt to repair something I ALWAYS go with the mind set that it is not working because I am doing something wrong and that I dont have enough information.
SO many times I've seen failures that pointed to "burnt chips" and "catastrophic failures" that were after all just some bad contacts or wrong jumpers
You're very welcome! Yes, I know that feeling well! Sometimes we overthink!
On the "it's all my fault" subject, I recommend this Shuttle motherboard repair video th-cam.com/video/pEeNqD2FOHQ/w-d-xo.html :)
Have a great day!
Oh, that was so familiar :) I spent so much time hunting phantoms just because I mixed the jumper orientation, or the documentation was wrong for particular mainboard revision.... Nice video!
So glad to hear I'm not alone here then! :) It was frustrating but also enjoyable! Thanks for the visit!
Good stuff Tony! Always great to see the full process!
Thank you, I'm glad you enjoyed the process!
Hi, Tony, thanks for another awesome video!
I was also puzzled by that 16.7 MHz output and thought it might be a failure one of the components those configure frequency in frequency generator circuit. Tend to overcomplicate things probably as much as you do :)
One positive outcome out of you misplacing the jumper is you now have some practice in drag soldering ;) Plus, those soldering joints were looking very suspicious anyway, won't hurt to get them resoldered. And you've got the experience. So many benefits out of one (not embarrassing!) mistake.
Great the repairing ended in a success!
Also, I remember back in the days my father taught me that before trying to solder the enameled wire you need to remove the coating with a [not too sharp] knife by stretching it between the blade and a thumb multiple times, turning the wire after a stretch. But it takes time, so, of course, the shielding from mic cable is way more convenient to use, good to know.
Absolutely, so not 100% wasted time! Still, it felt a bit silly, particularly because I moved that jumper to the wrong position myself :) Thanks for watching!
Thank you. looking forward to the next video.
Coming soon! Thank you!
Great video! Enjoyed this journey.
I'm glad you did despite being a silly mistake of mine :)
I hate those old jumper configs. I fried an AMD DX120 because I misread a jumper. I'm glad your problem wasn't with voltage like mine was.
I’m always scared about that, I always check the voltage at first power up but yes, hate that too 😂 thanks for watching!
Yeah everyone has been bitten more than once with those jumpers. Happens to me all the time! But at least you know more about PCI slots now
Exactly! So it wasn't all time wasted in the end! Thanks for watching!
I made a 100 of these kind of errors, and I know I am not an exception, just like you are not. What does help in a situation like this, is to have a friend around.
It always helps to have a shoulder to cry on indeed :) Thanks for watching!
Great info and video, I have an M915 PcChips board with the most peculiar errors that I've been unsuccessful with. This is giving me incentive to give it another try. Thanks.
I'm sure you can fix it, good luck and thanks for watching!
@@tony359 Grazie
Great work 👌👏👍
Thanks a lot 😊
@@tony359 you welcome 🤗
Important lesson, that's for sure :)
I'm confident this is not going to be the last time it's happening! :) Thanks for watching!
great work, still entertaining.
Thank you!
Great video Tony👍🏻 Luv it ❤… you just show how to trobleshoot the jumper setting problem if we don’t have any referrance for the motherboard.. so thank you..👍🏻
Glad you enjoyed it, thanks for watching!
I love videos like this
Thank you!
Nice work Tony and we can forgive a little bit of pilot error!
Thank you Sir - frustrating but it was fun :)
That was a great video. I love learning about how to approach troubleshooting. Even if it's something else in the end. I mean... even Saint Adrian had an Amiga that diagnosed the heck out of, only to find out that all it needed was reseating the chips, just last week.
Saint Adrian! I love that! I watched that video and I enjoyed it too! I would have done the same! Thanks for your comment and for watching!
@@tony359 Yes, I call him Saint Adrian the Resurrector, because he pulled off some absolutely incredible repairs.
@@catriona_drummond oh well Adrian’s videos are the next level indeed 🙂
I knew you set JP12 wrong from the beginning of the video because I have this board and just recently set it up. But I did not know JP12 set a 1/2 PCI divider, so thanks for determining that!
Yea felt like a waste of time but I also learned something so happy days 🙂 thanks for watching!
@@tony359I pulled out my M918 today and experimented with this a bit. I found that the system will work either way (1:1 or 1:2 PCI) but the jumper JP18 *must* also be set appropriately. If you select 1:1 with JP12, JP18 must be open. For 1:2 JP18 must be closed.
Ah the old days when you can misconfigure a jumper and declare the board as broken and buy a new one.
Ahahah 😂, definitely an option!
Wire wrap wire makes good bodge wire.
The sound driver may be related to CPU instruction set.
Early 486 boards has a 2GB hard drive limitation, you may also need to enable large LBA or similar setting in bios chipset settings.
Thank you for your comment!
Well, setting jumpers can be tricky. Today we have Internet but I remember a story from the back of the days when I was forced to configure 486/586 board without the manual becouse it was missed. It was trial and error excersise but finally I have managed!
Btw PCChips before Pentium wasn't that bad. Funny thing is that I just received 286 PCChips board as a gift :)
Great gift! Well I am told the 918 was also available with fake cache so not sure about the "pre-pentium" era! :) But if you get one with proper cache and proper chipset, well they work :) Setting up a Mobo with no manual is an interesting task :) Thanks for watching!
@@tony359 Well PC Chips engineers were able guys :D. Regarding jumpers - on many motherboards it's not only clock-divider-voltages but internal cache as well (WT/WB). And jumpers can be placed in both directions (90 degree one to other). So yes, with no manual it was interesting task. More over - I had no oscilloscope and multimeter. But I had something what do I don't have now - planty of time.
Nice video Tony. :-) I am a programmer, sometimes it takes several days to find out why the code is not working, and the reason, missed comma:-) Go ahead! And good luck!
And, it is not wasting time, it is experience.😉
Thank you! I do remember those missing semicolon when doing a little bit of Turbo Pascal at school LOL! Thanks for watching!
doesn't matter. it's still a great video. Thanks😉
Thank you!
Great video Tony. They do say the jouney is just as important as getting there :-)
The jumper markings a really terrible on that board.
From what I can see, there is a line between pin 1 and 2 under the plastic of the jumper.
How hard would it have been for them to just print a 1 or a dot there?
Keep up the good work. I look forward to your videos.
There is a line indeed! Granted, I should have invested a few more seconds at the beginning - maybe the manual was giving some clues - but I tend to be impatient and then I just got in the troubleshooting rabbit hole LOL! Thanks for watching!
"you wasted your time" was PcChips trademarked marketing slogan.
ahah - I see why! :) Thanks for watching!
@@tony359 Just finished watching the video. Great content as always! Those pins might have worked, but they were definitely on their way out, slight motherboard flex and we end up with intermittent fault, the worst kind.
@@rasz oh yes, they were a ticking bomb indeed so not all time was wasted 🙂
I encountered my first pc chips board last week. Very cost down lol.
PCChips boards were notorious for being a PIA
Yeah, I can see that! LOL! Thanks for watching!
Don't feel bad about the jumper. It was a common issue with 486 & socket 5 systems. That's why the industry pushed for jumperless configurations.
I can see tons of computers being returned because "faulty" - and damage done - indeed. It's in fact amazing where we are today in terms of plug an play. Thanks for that and for watching!
I still look for the board diagram to make sure I’m getting the jumpers the right way today. Even when working on modern boards.
Theres an anti glare light for microscopes. Might help. Northridge Fix uses this all the time.
It’s the ‘side light’. They have pros and cons. In fact I discuss that on the Joyalens video. The ring light I have is so convenient I’m struggling to justify another (expensive) item on my desk. But thanks for mentioning and for watching!
Nice thorough video. Can i ask your oscilloscope, where do you attach the gnd lead? Is it the keyboard connector. Thanks
Also what purpose is there for 16mhz on a pci slot?
Usually the keyboard connector yes as it has a nice place where to clip. I need to make sure it’s grounded first. Could that be that the clock generator is referenced to the CPU clock so that 16.7MHz would output 33Mhz with a different CPU? Is that generator just a multiplier? Thanks for watching.
I ended up buying a PCChips M918 board with fake cache.
It had a broken transistor (the big one) for the 3.3V power output, so I had to find a similar transistor to replace it with.
I then removed all the fake cache chips and put in real sockets. Getting the cache to work was a real pain in the butt, ended up using oscilloscope to buzz out all the address and data lines on the cache chips until I found a small trace break in the cache jumper selection block where I damaged a trace from removing the fixed wires to install selectable jumpers. A small bodge wire and she is fully working with cache.
So yeah, a little of my own soldering error there.
I learnt just yesterday about of the existence of "fake cache" M918! I'm glad you managed to swap the chips for real ones, I'm sure the board feels much better! And well done for finding the broken trace! Must have felt good when the board finally booted up with some working cache!
@@tony359 It is a bit faster with the added L2 cache ... but this board is already pretty quick and has a decently fast memory bus, so not as much difference as you might first expect.
The M918 will also take EDO RAM in addition to Fast Page RAM and gets some pretty decent memory performance speeds out of EDO RAM. This is kind of rare for 486 based motherboards, as EDO RAM was really more of a Pentium chipset thing.
As much as PCchips boards feel a bit like junk, they must have done something right because so many of their boards are still ticking along some 30 years after manufacture.
@@tony359 Oh, yes, it was very much a good thing to find and fix the fault. Even if I did create that fault myself.
Brilliant people can make silly mistakes like the rest of us .. I’m pretty sure it’s our subconscious just having a laugh. Mine is a bloody comedian.. lol.
LOL! Thanks for watching!
very interesting video - thank you. Can I ask what microscope system did you use in the video for SMD/examination?
Thank you! I can actually show you! th-cam.com/video/2Ag7E7KMR6s/w-d-xo.html
Much appreciated@@tony359
I think if you look carefully at those 3 pin jumpers from the side/on an angle you will see that pin 1 is boxed on the silkscreen.
I read your comment and I had to go and check!!! You are right but I have to say it's barely visible, the pins are covering it 99.9%! They could have made a thicker line or anything. Good hint though for my next time! Thanks for watching!
@@tony359 I caught only the tiniest glimpse of it on the jumpers with the red links.
Could definitely be more obvious! Or maybe it is in the manual.
Love your work. 😄
@@TomStorey96 From the manual "We intentionally hid pin 1 locations to make your days much more exciting" - thanks for your comment, appreciated! :)
5:20 is the exact moment where you messed up
YESSSSS!!! I did it all by myself LOL!
Amazing, after you remove the adesive from the EEPROM put it back or the light can delete the data in the EEPROM
Absolutely! Thanks for watching!
@@tony359 thank you!
That CMA8819A produced by C-Media Electronics Inc. but unfortunately, i also not find any more information about this clock generator.
Thanks for looking!
😂😂😂😂😂 eu ri no fim do vídeo com aquele jumper
Yes, better to have a laugh... 😂😂
Hi tony359, I have a motherboard like this but missing 2 transistores, Q6 and Q7. Do you have information about it to obtain and complete board? Thanks
Q6 is TL431, Q7 is 2N3904. Good luck :)
@@tony359 many thanks, I will try to return back to life. I am follower of your videos and I love your technique to solve issues. Continue like this!!!
For the sb live to work you need emm386.exe RAM in config.sys that was your problem i think.
Good suggestion, thanks! I cannot remember if I had it running - I likely didn't. Thanks for watching!
@@tony359 no , thank's to you tony
How did u get old pc board
Ebay :)
@@tony359 good job, even those board more old than my age 🤣🤣
Maybe this board last 100 years
Maybe. It may become Skynet and take over the world.
Then the trace I fixed will break and the world will be free again.
As soon as you hear PC Chips or Matsonic, you are losing your time. I remember they were horrible back in de days…
ahah yes, I know! But hey, they keep me busy :) Thanks for watching!