The format works. If you get some spools of tinned copper bus wire and some tiny heat shrink tubing to match, it makes jumper wires much easier to deal with, and gives cleaner looking results. The bus wire can be soldered to the first joint while still attached to the spool, then contoured, covered with shrink tube, cut to length, and soldered. I get tubing that is barely larger than the wire and don't bother heating it beyond just soldering it. The length doesn't change significantly either way, I'm just skipping the step because it's a waste of time. Alternatively when a point to point through hole wire connection is needed, use the heat shrink tubing cut to length, then add the bus wire to make perfect 90 degree through hole solder leads. The result is a perfect length, virtually heatproof, insulated wire. It's faster than cutting and stripping wires too. BTW, if you liked the silk screen with the pin numbers on this Amstrad, I have a github repo for classic dip chip pin labels you can print-cut-paste them on. There are around 300 part numbers available in pre-made PDF's that should be ready to print. The two GIMP xcf files I used to make everything are also in the repo, one for standard width chips and one for wide/oversized (like 68k), along with a full description of how I used them with the layers file organization structure and fabrication guide layers. I tried to explain it so that even someone unfamiliar with GIMP could figure it out. If you check out the repo, my labels are unlike any you'll find elsewhere. I started the label project after beginning Ben Eater's breadboard computer project and getting tired of looking at datasheets constantly for pinouts. Later, I also got tired of the nonintuitive pin nomenclature used on the various chip datasheets, so I started evolving my labels with lookup tables, logic gate graphics, and other (hopefully) more intuitive naming conventions. I'll skip adding the link here for YT reasons, but my github account has the same name as here, searching should find it easily. These labels might be a convenient filming alternative to the datasheet pinout overlays, although I didn't exactly design them for this, and the text sizes may be too small to be useful. I can also say, the labels are easy to add and remove using PVA glue (white glue like what is used in grade school), and alcohol to remove it without a trace. Most of my labels have full pin numbering, function naming, and a header portion with part number, manufacturer, and purpose. Most part numbers have several label versions, and I've made labels specific to the various package dimensions of my parts stock. I mean things like the logic gate chips that come in the short 1/2 end pin configuration with shorter length versus the ones that have overhang. I've made labels for both so my labels fit either version perfectly. I'm not monetized here in YT, and the labels are MIT licensed (AKA no warranty, do whatever you like with them, I don't care if it's commercial or otherwise). I share freely, no strings attached. Thanks for the upload. -Jake
I worked for Control Data in the late 60's early 70's and we used 30 gauge wire wrap wire to repair or modify circuit cards. The nice thing about wire wrap wire is you can route it between pins on the topside as well as solder it to pads or traces on the underside. A tiny drop of glue will keep long wires in place so they don't snag and get pulled off. We used to call the mods to existing boards "engineering change orders" so we often just called the wire ECO wire! I prefer videos under 20 minutes as well. Some of the channels with hour or longer videos are useful but hard to watch all in one sitting.
I usually like longer videos, but this format is perfect for covering just a single issue. Very informative and easy to follow troubleshooting with good explanations. Your channel is quickly becoming one of my favourite retro computer channels!
An 18 minute video is still a decent length. On top of that if you have something like this and you put it in a series of 2 or 3 videos on a full 6128 restoration then it's not likely to stand out for the particular fix you're doing. And an interesting fix like this does deserve to stand on it's own.
Thanks! Glad you liked it. This one ended up being longer than I intended, but the important part from my point of view is that it's quick to make 😃 It can easily take the slot of a mail day or something like that depending on what I run into. On the other hand, regular episodes can take MANY hours to plan, shoot, and edit.
Ok this is the second time I've watched this and now it makes more sense than ever, thanks Noel, now my Dandanator really fits the bill and I hope I can rescue my dead ram CPC 664 using these techniques.
I like the format, 18 minutes is not short... and you could make videos like this one about all kinds of "special" repairs. I'd love to see you do some collaboration with your peers on TH-cam, your channel deserves a lot more views. :)
Yeah, this particular one ended up being longer than I expected, but at least it was short to make, so it didn't take the spot of a full episode. I would also love to collaborate with other fellow retro TH-camrs! Hopefully it's something that's coming up in the near future! 😃👍
After using the desoldering gun, a little bit of hot air really does miracles during the actual IC removal. It works. No pin wiggling necessary. I never damaged a single pad anymore since I'm using that method.
The problem is that the hot air needs to be hot enough to melt the solder, right? I'd just be concerned about damaging the IC (assuming you want to save it).
@@NoelsRetroLab Thats what truly surprised me: 200°C and just around 5 seconds is enough. Put only a slight bit of tension under the chip with a tool an heat its legs. It will magically slip out. I don't really understand why this is. I suspect that if you already removed most of the solder with the desoldering gun, the little bit of solder left in the holes has such a little thermal mass compared to the IC legs that it will get soft -very- fast. Just try it next time.
very enjoyable and comfortable to follow, all coupled with your usual competence and clarity of explanation! No wonder, by now, as this has become the standard for most if not all of your videos. And great close-ups on the various board details! This really makes the difference and let us follow you so well! Thank you Noel!
Great video of troubleshooting on a damaged CPC. Thank you for explaning it in such a detail. 20 years ago I have fried a CPC while changing RAM chips and I put it away in storage. Maybe I can get it back to life with this video.
I find the fragile PCB traces are more a board manufacturer thing, as you will find boards made by a particular manufacturer are robust, yet the same board layout from a different supplier, ostensibly the same board base material, is more prone to either delaminating or breaking tracks. I got reasonably good at repairing them on some, as the traces would lift, so you had to glue them down with superglue, and then solder in a thin wire link to get connectivity, then glue the trace back down again where you soldered, as that are had debonded again. Then a thin film of conformal coat applied both to protect the board, and to hold down the other traces as well. Had to do a lot of really thin wires through some through hole traces to get connection to the inner traces, though they were only power and ground, so easy enough to simply glue down a wire to an alternate hole nearby for this, in addition to the original hole being repaired. Some of those boards the only thing keeping the traces on it seemed was the conformal coating, which was not meant to be solder through, but with a good extractor to get the fumes out fast you could solder through with no problems. Some though the coating would survive solder temperatures without passing it's glass transition point, so you had to use fine sandpaper on a stick to get through it to solder. Those boards you had a 50/50 chance to get the component off in many cases, as if there was anything under it trace wise you knew it was going to get torn off with the faulty component in most cases, and you would have to do a whole bunch of fine repairs. Those boards though were bleeding expensive as spare parts, so you did try your best, the price was more than the mass in gold.
Thank you Noel. I have never had a problem with any of my 6128's. Then again, I don't supply the 5V rail with 12V but if any repair is needed in future I will be aware. I can't understand why the tracks need to be so thin. After all the blank PCB's are fully coated with copper before etching so that rules the cost saving out.
Noel, this format is excellent. As soon as you could produce some more, it will make a guide of usual/unusual repairs. A very straightforward method to test the shared pins in the bank memory!. Another Noel's fine video to index and record in my personal wiki !!!!
I do like this format for both its "quick fix" content and running time. Personally, I usually prefer retro repair videos under 20 minutes -- multiple parts are fine but 20-ish minutes suits my free-time requirements the best. Wow, those are thin traces. It is no wonder why they are so delicate. I have used the same probing technique when testing C64 RAM faults; it's quick and methodical. Great work again with this video -- very informative. 👍
Hi Noel! Very nice that you are showing the things that come all right but also those that went not so good like those traces. Me watching the close ups, couldnt see signs of damaged tracks but they were there. Very nice bodge wire work. It is good that all chips will have a socket. In the future fixes will be easier.
I have worked on all of those machines and I believe that the the 8bit line and especially the 800xl without the Freddy chip has the best board to work with. C64's(and Amiga 500) ground and 5 volt planes are so massive and the absorb all the heat that even changing caps becomes a pain, Speccys are really delicate as you said and I was fortunate enough to watch this video before digging in my CPC! Last month I worked on an Atari 130 XE. I removed all the ICs , installed standard sockets... and didn't work. I remove all the sockets since I don't have have diagnostic tools, installed machine sockets and it worked perfectly. I even de-soldered some sockets for a third time (really late arrivals of machine sockets) and no tracks were broken. I can not say the same for an 65XE though!
The shape of the track->pad connection on the C64 is called "teardrop" ( en.wikipedia.org/wiki/Teardrop_(electronics) ) and it's made specifically to prevent the stress point where those breaks happen. It's much less of a concern on modern PCB process, but still used regularely.
Thanks! I didn't know that name, but I definitely saw that the C64 has that shape. This explains exactly why you'd want that! Now, was it more expensive to design PCBs with that shape? I wouldn't think so, right? I wonder why they didn't use that in these boards.
Mmm, I don't see why the PCB production would be more expensive with that shape, it's just the photo mask that changes. Now it's possible that whatever process they were using to create the photo mask master didn't support it ...
Great subject about the broken connection, and the video was just the right length, 20 min and i was ready to clean up after eating my dinner. Keep em comming.
Excellent way to explain this, it's amazing that on the empty bank, you could not see any problem. Thank you! Hopefully I sort this cpc 464 that thinks all it's ram is bad. Take care, and I look forward to the next one :)
excellent repair there. my 6128 though ive made a lot worse today 😢 i had changed all of the ram for ram sockets.. 7 repaired traces later i go to connect the cables to the board without the keyboard just to test and i had connected the 2pin connector of the 12v in into the 2 pin connector on the top left of the board... i turn on the amstrad and got a black screen initially, but after checking continuity again on the ram and the 400025 chip it booted to a black border screen with a solid pink sort of colour. im going to buy a dandanator if i can find one in stock to see what it could be. i wish i hadnt have messed with desoldering all of the ram!
Thanks a lot for the extremely systematic repair video. The insert graphics are very helpful. I learned quite a bit. As far as I am concerned length is perfect.
Nice format and easy to comprehend explanations. Did you ever try to use the wiring pen for repairing the tracks? Works much better than the wires in plastic insulation. I once had to fix 4 factory-drilled holes on the RAM placeholders of the CBM-4008 motherboard. Those were drilled to prevent RAM upgrade. Using the wiring pen was the obvious choice. You can even do the cable management with the special plastic holders.
I enjoyed the video, very informative as always and something that will definitely be useful knowledge for me in future. You must have a good eye to have spotted this issue! 👍 Also congrats on the PCBWay sponsorship, I've used them for my own projects in the past and they've always been great! I think they're a very good fit for your channel.
Glad it was helpful! That's the thing, I spotted the damaged track of the ROM, but I was blown away with one of the ones of the RAM (the one in the video thumbnail). I can't find anything wrong for the life of me just by looking at it! It must be a crack between the edge of the eyelet and the track that's pretty much invisible. Kind of crazy!
the tracks on that 6128 look great compared to the ones on the 464 board i have..almost looks as if they are just soldered tracks.. bubbley and wavey in places and way to easy to bridge the tracks! nice one Noel.. keep it up!
Hi, these bubbly and wavey tracks are just fine. Back than the HASL process ( hot air solder level) as it is done nowadays wasn’t esteblished that widely and ENIG (electrodes nickel immersion gold) is expensive in this low budget market for large boards so they simply put the bare boards through a wave solder bath a - for protection and b - for added current handling capabilities on thin traces. Those old HMOS and NMOS chips are much more power hungry than more modern CMOS variants.
Thanks for the video. Been looking at Anstrads but so far just looking. Guess the takeaway for me is to just do a continuity test every time you replace something. Pretty sure I was taught that, but you know how that goes. But with the amount of stuff you pulled I can see how you could really get behind the curve. Wonder if using some hot air and a solder sucker may have been less abusive. I find that just because I have a tool that is fast I tend to use it. Sometimes fast just turns into more work in the end. Anyway thanks. Great content.
Checking every single connection you unsolder is ideal but a bit unrealistic when you do massive ones like in this board. And really, 99.9% of the time I either have no problems, or I'll see that I pulled up a track and I'll fix it. What shocked me here is that I honestly thought they were fine and the damage was tiny.
@@NoelsRetroLab Yeah I couldn't see it on the video. I was kinda shocked too because everything I've seen on Amstrads say they are built really tough and are one of the more reliable of the 80's 8 bits.
Well done , another good machine coming back to life :-D When i removed solder, i would make sure each pin could be waggled easy, then check the top of the pcb for the pin shoulders that may still be soldered. If you need really thin enameled copper wire, find a faulty battery clock, there is a solinoid full of really fine wire :-D
That's also what I normally do (I don't usually show that step in the videos because it's boring once you've seen it once). Normally I never have this happen, so there's something very fragile about these tracks. Weird. The enameled copper wire would be for those track repairs? I should look into it because I don't like this small wire. Very easy to melt the plastic away.
@@NoelsRetroLab The enameled wire in battery clocks is like hair, you can make a link and it's almost invisible. Thicker enamel wire can be found in dead car relays or small motors/stepper motors/solinoids. Worth having a look around :-D
@@NoelsRetroLab No, if you apply an iron and solder to the end, the laquer burns off while you tin it. The clock wire is like a hair so you would never be able to scrape it. If thin wire is a nightmare to you, take a small mains transformer apart and use that laquered wire, an old phone charger thats. (low current means thin wire.) Anything that uses inductors, a speccy 5 to +/- 12v bobin coil. have a go :-D
Do it! It's really fun learning about an alternate tech history you didn't grow up with. Just the same way I've been enjoying diving on the TI99 or the Apple IIe.
17:20 That ground lead looks to be shorting against a pin on an IC down to the right there. It even looks like the machine reset at that point. Great and informative video though! :)
@@NoelsRetroLab In the country where I live I can not buy the thinnest shrink tube in stores. The alternative is to order from Ebay or buy a cord that can withstand more heat.
The format works.
If you get some spools of tinned copper bus wire and some tiny heat shrink tubing to match, it makes jumper wires much easier to deal with, and gives cleaner looking results.
The bus wire can be soldered to the first joint while still attached to the spool, then contoured, covered with shrink tube, cut to length, and soldered. I get tubing that is barely larger than the wire and don't bother heating it beyond just soldering it. The length doesn't change significantly either way, I'm just skipping the step because it's a waste of time.
Alternatively when a point to point through hole wire connection is needed, use the heat shrink tubing cut to length, then add the bus wire to make perfect 90 degree through hole solder leads. The result is a perfect length, virtually heatproof, insulated wire. It's faster than cutting and stripping wires too.
BTW, if you liked the silk screen with the pin numbers on this Amstrad, I have a github repo for classic dip chip pin labels you can print-cut-paste them on. There are around 300 part numbers available in pre-made PDF's that should be ready to print. The two GIMP xcf files I used to make everything are also in the repo, one for standard width chips and one for wide/oversized (like 68k), along with a full description of how I used them with the layers file organization structure and fabrication guide layers. I tried to explain it so that even someone unfamiliar with GIMP could figure it out. If you check out the repo, my labels are unlike any you'll find elsewhere.
I started the label project after beginning Ben Eater's breadboard computer project and getting tired of looking at datasheets constantly for pinouts. Later, I also got tired of the nonintuitive pin nomenclature used on the various chip datasheets, so I started evolving my labels with lookup tables, logic gate graphics, and other (hopefully) more intuitive naming conventions. I'll skip adding the link here for YT reasons, but my github account has the same name as here, searching should find it easily.
These labels might be a convenient filming alternative to the datasheet pinout overlays, although I didn't exactly design them for this, and the text sizes may be too small to be useful. I can also say, the labels are easy to add and remove using PVA glue (white glue like what is used in grade school), and alcohol to remove it without a trace. Most of my labels have full pin numbering, function naming, and a header portion with part number, manufacturer, and purpose. Most part numbers have several label versions, and I've made labels specific to the various package dimensions of my parts stock. I mean things like the logic gate chips that come in the short 1/2 end pin configuration with shorter length versus the ones that have overhang. I've made labels for both so my labels fit either version perfectly.
I'm not monetized here in YT, and the labels are MIT licensed (AKA no warranty, do whatever you like with them, I don't care if it's commercial or otherwise). I share freely, no strings attached.
Thanks for the upload.
-Jake
Thanks for the feedback. And that sounds like a good idea. I'll try that. Cheers.
I worked for Control Data in the late 60's early 70's and we used 30 gauge wire wrap wire to repair or modify circuit cards. The nice thing about wire wrap wire is you can route it between pins on the topside as well as solder it to pads or traces on the underside. A tiny drop of glue will keep long wires in place so they don't snag and get pulled off.
We used to call the mods to existing boards "engineering change orders" so we often just called the wire ECO wire!
I prefer videos under 20 minutes as well. Some of the channels with hour or longer videos are useful but hard to watch all in one sitting.
I usually like longer videos, but this format is perfect for covering just a single issue. Very informative and easy to follow troubleshooting with good explanations. Your channel is quickly becoming one of my favourite retro computer channels!
An 18 minute video is still a decent length. On top of that if you have something like this and you put it in a series of 2 or 3 videos on a full 6128 restoration then it's not likely to stand out for the particular fix you're doing. And an interesting fix like this does deserve to stand on it's own.
Thanks! Glad you liked it. This one ended up being longer than I intended, but the important part from my point of view is that it's quick to make 😃 It can easily take the slot of a mail day or something like that depending on what I run into. On the other hand, regular episodes can take MANY hours to plan, shoot, and edit.
Ok this is the second time I've watched this and now it makes more sense than ever, thanks Noel, now my Dandanator really fits the bill and I hope I can rescue my dead ram CPC 664 using these techniques.
I like the format, 18 minutes is not short... and you could make videos like this one about all kinds of "special" repairs. I'd love to see you do some collaboration with your peers on TH-cam, your channel deserves a lot more views. :)
Yeah, this particular one ended up being longer than I expected, but at least it was short to make, so it didn't take the spot of a full episode. I would also love to collaborate with other fellow retro TH-camrs! Hopefully it's something that's coming up in the near future! 😃👍
After using the desoldering gun, a little bit of hot air really does miracles during the actual IC removal. It works. No pin wiggling necessary. I never damaged a single pad anymore since I'm using that method.
The problem is that the hot air needs to be hot enough to melt the solder, right? I'd just be concerned about damaging the IC (assuming you want to save it).
@@NoelsRetroLab Thats what truly surprised me: 200°C and just around 5 seconds is enough. Put only a slight bit of tension under the chip with a tool an heat its legs. It will magically slip out. I don't really understand why this is. I suspect that if you already removed most of the solder with the desoldering gun, the little bit of solder left in the holes has such a little thermal mass compared to the IC legs that it will get soft -very- fast. Just try it next time.
@@Shmbler Interesting. I will try it next time. Thanks for the tip!
very enjoyable and comfortable to follow, all coupled with your usual competence and clarity of explanation! No wonder, by now, as this has become the standard for most if not all of your videos. And great close-ups on the various board details! This really makes the difference and let us follow you so well! Thank you Noel!
Wow! Thanks so much! I'm glad you're enjoying them 👍😃
Great video of troubleshooting on a damaged CPC. Thank you for explaning it in such a detail. 20 years ago I have fried a CPC while changing RAM chips and I put it away in storage. Maybe I can get it back to life with this video.
Glad it was useful. Amstrads are easy to repair. Give it a try!
I find the fragile PCB traces are more a board manufacturer thing, as you will find boards made by a particular manufacturer are robust, yet the same board layout from a different supplier, ostensibly the same board base material, is more prone to either delaminating or breaking tracks. I got reasonably good at repairing them on some, as the traces would lift, so you had to glue them down with superglue, and then solder in a thin wire link to get connectivity, then glue the trace back down again where you soldered, as that are had debonded again. Then a thin film of conformal coat applied both to protect the board, and to hold down the other traces as well.
Had to do a lot of really thin wires through some through hole traces to get connection to the inner traces, though they were only power and ground, so easy enough to simply glue down a wire to an alternate hole nearby for this, in addition to the original hole being repaired. Some of those boards the only thing keeping the traces on it seemed was the conformal coating, which was not meant to be solder through, but with a good extractor to get the fumes out fast you could solder through with no problems.
Some though the coating would survive solder temperatures without passing it's glass transition point, so you had to use fine sandpaper on a stick to get through it to solder. Those boards you had a 50/50 chance to get the component off in many cases, as if there was anything under it trace wise you knew it was going to get torn off with the faulty component in most cases, and you would have to do a whole bunch of fine repairs. Those boards though were bleeding expensive as spare parts, so you did try your best, the price was more than the mass in gold.
Thank you Noel. I have never had a problem with any of my 6128's. Then again, I don't supply the 5V rail with 12V but if any repair is needed in future I will be aware.
I can't understand why the tracks need to be so thin. After all the blank PCB's are fully coated with copper before etching so that rules the cost saving out.
Yeah, it's puzzling. Especially having great examples like the C64 boards.
Noel, this format is excellent. As soon as you could produce some more, it will make a guide of usual/unusual repairs. A very straightforward method to test the shared pins in the bank memory!. Another Noel's fine video to index and record in my personal wiki !!!!
Thanks! Yes, there are more to come for sure 😃
I do like this format for both its "quick fix" content and running time. Personally, I usually prefer retro repair videos under 20 minutes -- multiple parts are fine but 20-ish minutes suits my free-time requirements the best.
Wow, those are thin traces. It is no wonder why they are so delicate. I have used the same probing technique when testing C64 RAM faults; it's quick and methodical. Great work again with this video -- very informative. 👍
Hi Noel! Very nice that you are showing the things that come all right but also those that went not so good like those traces. Me watching the close ups, couldnt see signs of damaged tracks but they were there. Very nice bodge wire work. It is good that all chips will have a socket. In the future fixes will be easier.
Thanks 👍
I have worked on all of those machines and I believe that the the 8bit line and especially the 800xl without the Freddy chip has the best board to work with. C64's(and Amiga 500) ground and 5 volt planes are so massive and the absorb all the heat that even changing caps becomes a pain, Speccys are really delicate as you said and I was fortunate enough to watch this video before digging in my CPC! Last month I worked on an Atari 130 XE. I removed all the ICs , installed standard sockets... and didn't work. I remove all the sockets since I don't have have diagnostic tools, installed machine sockets and it worked perfectly. I even de-soldered some sockets for a third time (really late arrivals of machine sockets) and no tracks were broken. I can not say the same for an 65XE though!
The shape of the track->pad connection on the C64 is called "teardrop" ( en.wikipedia.org/wiki/Teardrop_(electronics) ) and it's made specifically to prevent the stress point where those breaks happen. It's much less of a concern on modern PCB process, but still used regularely.
Thanks! I didn't know that name, but I definitely saw that the C64 has that shape. This explains exactly why you'd want that! Now, was it more expensive to design PCBs with that shape? I wouldn't think so, right? I wonder why they didn't use that in these boards.
Mmm, I don't see why the PCB production would be more expensive with that shape, it's just the photo mask that changes.
Now it's possible that whatever process they were using to create the photo mask master didn't support it ...
Great subject about the broken connection, and the video was just the right length, 20 min and i was ready to clean up after eating my dinner. Keep em comming.
Great to hear!
Excellent way to explain this, it's amazing that on the empty bank, you could not see any problem. Thank you! Hopefully I sort this cpc 464 that thinks all it's ram is bad. Take care, and I look forward to the next one :)
excellent repair there. my 6128 though ive made a lot worse today 😢 i had changed all of the ram for ram sockets.. 7 repaired traces later i go to connect the cables to the board without the keyboard just to test and i had connected the 2pin connector of the 12v in into the 2 pin connector on the top left of the board...
i turn on the amstrad and got a black screen initially, but after checking continuity again on the ram and the 400025 chip it booted to a black border screen with a solid pink sort of colour. im going to buy a dandanator if i can find one in stock to see what it could be. i wish i hadnt have messed with desoldering all of the ram!
You can also just burn the diagnostics ROM to a 27C128 (or 256) and use that, but you'll have to desolder the ROM IC for that.
Thanks a lot for the extremely systematic repair video. The insert graphics are very helpful. I learned quite a bit. As far as I am concerned length is perfect.
Great to hear!
Great work Noel. I have no issues with this format. Infact I dont mind any format , its all good ! . Thanks Noel.
Good to hear!
Nice format and easy to comprehend explanations. Did you ever try to use the wiring pen for repairing the tracks? Works much better than the wires in plastic insulation. I once had to fix 4 factory-drilled holes on the RAM placeholders of the CBM-4008 motherboard. Those were drilled to prevent RAM upgrade. Using the wiring pen was the obvious choice. You can even do the cable management with the special plastic holders.
I enjoyed the video, very informative as always and something that will definitely be useful knowledge for me in future. You must have a good eye to have spotted this issue! 👍
Also congrats on the PCBWay sponsorship, I've used them for my own projects in the past and they've always been great! I think they're a very good fit for your channel.
Glad it was helpful! That's the thing, I spotted the damaged track of the ROM, but I was blown away with one of the ones of the RAM (the one in the video thumbnail). I can't find anything wrong for the life of me just by looking at it! It must be a crack between the edge of the eyelet and the track that's pretty much invisible. Kind of crazy!
Surprise NRL video! I always feel happy once I hear you say "Hello and welcome to Noel's Retro Lab" :-)
Haha, thanks! Yes, I don't often post on Mondays, but I had enough videos saved for a Monday-Thursday week, so double whammy! 😃
Same here :D
nice diagnostics Noel great episode learned a lot from it mate
Thanks 👍
the tracks on that 6128 look great compared to the ones on the 464 board i have..almost looks as if they are just soldered tracks.. bubbley and wavey in places and way to easy to bridge the tracks!
nice one Noel.. keep it up!
Hi, these bubbly and wavey tracks are just fine. Back than the HASL process ( hot air solder level) as it is done nowadays wasn’t esteblished that widely and ENIG (electrodes nickel immersion gold) is expensive in this low budget market for large boards so they simply put the bare boards through a wave solder bath a - for protection and b - for added current handling capabilities on thin traces. Those old HMOS and NMOS chips are much more power hungry than more modern CMOS variants.
Great info. I have a poor 64c board I have restored. I use nylese 30 gage enamel wire and squared routing in the NASA fashion
Yeah, I just got some enameled wire as well for similar track restoration projects. Pretty useful.
Hello! Congratulations for your great work. I have a question. Is there a way to measure or to check the ram chips when they are out of the board ?
It did'nt even look damaged on the close-up. Just goes to show you should always use a continuity tester. Great vid.
Right? Those tracks were crazy.
I like both short and long videos!
Wow, that is interessting, thank you for sharing! Gonna have a closer look upon traces next time I work on PCBs :-)
Glad you liked it!
Really informative video. I guess Alan Sugar, ever the shrewd businessman, decided to save a few bob by using thinner tracks on the PCB. 😊
Who knows! This might just be a manufacturing issue without having much cost-savings associated with it. Not sure.
Thanks for the video. Been looking at Anstrads but so far just looking. Guess the takeaway for me is to just do a continuity test every time you replace something. Pretty sure I was taught that, but you know how that goes. But with the amount of stuff you pulled I can see how you could really get behind the curve. Wonder if using some hot air and a solder sucker may have been less abusive. I find that just because I have a tool that is fast I tend to use it. Sometimes fast just turns into more work in the end. Anyway thanks. Great content.
Checking every single connection you unsolder is ideal but a bit unrealistic when you do massive ones like in this board. And really, 99.9% of the time I either have no problems, or I'll see that I pulled up a track and I'll fix it. What shocked me here is that I honestly thought they were fine and the damage was tiny.
@@NoelsRetroLab Yeah I couldn't see it on the video. I was kinda shocked too because everything I've seen on Amstrads say they are built really tough and are one of the more reliable of the 80's 8 bits.
Really loved this format video, very informative, thanks Noel.
Glad you enjoyed it! Thanks for the feedback.
Disfruté el formato de este video y me gustaría ver más. I enjoyed the format of this video and would like to see more.
Short video also good, but no problem with longer one, just depends on how common the problem was, personally I don't mind longer video.
Good to hear. Thanks for the feedback!
Great format, thanks!
Thanks! More to come!
And this is why we try to keep up with commodore instead. Because they are keeping up with us :P
A brave format! Great job!
Continuity testers are your friend.
welcome to the quack bats !
Great video as always and yes to the format...
Awesome, thank you!
Well done , another good machine coming back to life :-D
When i removed solder, i would make sure each pin could be waggled easy, then check the top of the pcb for the pin shoulders that may still be soldered.
If you need really thin enameled copper wire, find a faulty battery clock, there is a solinoid full of really fine wire :-D
That's also what I normally do (I don't usually show that step in the videos because it's boring once you've seen it once). Normally I never have this happen, so there's something very fragile about these tracks. Weird. The enameled copper wire would be for those track repairs? I should look into it because I don't like this small wire. Very easy to melt the plastic away.
@@NoelsRetroLab The enameled wire in battery clocks is like hair, you can make a link and it's almost invisible.
Thicker enamel wire can be found in dead car relays or small motors/stepper motors/solinoids.
Worth having a look around :-D
@@zx8401ztv You still need to strip the enamel to solder the ends, right? How do you do it? Just with a knife or some scissors I take it?
@@NoelsRetroLab No, if you apply an iron and solder to the end, the laquer burns off while you tin it.
The clock wire is like a hair so you would never be able to scrape it.
If thin wire is a nightmare to you, take a small mains transformer apart and use that laquered wire, an old phone charger thats. (low current means thin wire.)
Anything that uses inductors, a speccy 5 to +/- 12v bobin coil.
have a go :-D
@@zx8401ztv Cool! I'll definitely look into it. Thanks!
What damaged it? Bad caps in the monitor supplying the power?
Also very cool with the short repair video. :)
Glad you liked it!
Nice video Noel
Absolutely. Nice blerb ;-)
These videos make me want to buy some British/European computers.
Do it! It's really fun learning about an alternate tech history you didn't grow up with. Just the same way I've been enjoying diving on the TI99 or the Apple IIe.
Muy instructivo!!! 👏👏👏👏
Excellent!
17:20 That ground lead looks to be shorting against a pin on an IC down to the right there. It even looks like the machine reset at that point. Great and informative video though! :)
"How do fix it? We just need to make a connection between pin 12 and the via."
Then he solders a wire to pin 11.
Noticed that too...
15:57 I suspect break is next to red pokey thing. Small dark line in circuit.
Yeah, it has to be there, but it's looking more to me like a manufacturing error that the track can "crack" away from the eyelet so easily.
Another great video thankyou short and sweet you got straight to the problem thankyou
I like the format, thumbs up from me
to prevent the cable from melting, use a crimp tube
I'm not sure I have them thin enough, but that's a good idea.
@@NoelsRetroLab In the country where I live I can not buy the thinnest shrink tube in stores.
The alternative is to order from Ebay or buy a cord that can withstand more heat.
Por mi perfecto, este formato está muy bien.Buen video.
very interesting video
Glad you liked it!
When I track a board I make the tracks as big as possible. I hate tiny tracks, there is no point for using them if you have space.
More videos are more gooder.
Sei Italiano Noel? Quando visito la tua pagina Patreon è in italiano. Tuttavia non è un problema per me.
No, sono spagnolo. Forse la pagina di Patreon è stata tradotta. 😃
@@NoelsRetroLab Sí, eso es lo que ha sucedido, aunque no estoy segura de por qué.
Why not desolder a socket and make a fix where it was broken? It was doable. I know, some struggle with desoldering.