You can also put a fuse on the power line before the zener. That way when the voltage goes over the zener's nominal value the zener will sink current to ground blowing the fuse and no voltage will go to the power rail.
I like that. It would be a matter of picking the right fuse, right? Probably something like 2A slow? The biggest problem is mounting it neatly on the board somewhere.
A signal level mosfet, zener, two bjts and a few resistors can make a proper over voltage cut off circuit. Using SMD stuff it can be made pretty small.
@@NoelsRetroLab search 'pico fuse', choose a current value that's 135% of max load (don't forget cart & other peripherals that load the power supply). This circuit is better than nothing but the diode you are using is intended for spike protection not continuous over-voltage (maybe a few in parallel would be better). But you would be better off using an SCR crowbar circuit, it's only a few parts. (prototype tip: use a 4.8-6.3v bulb instead of a fuse to test trigger voltage)
This is fast becoming my favourite retro hardware repair channel :-) Clear influences from Adrian Black but that's no bad thing. Keep up the great work. Adiós!
@@NoelsRetroLab Yourself and Adrian are my two favourite channels these days. Both informative and not overly pretentious :D Thank you for the entertainment :)
Thank you for showing this P6KE TVS diode! It's quite more resilient in terms of power handling than I expected, it should certainly blow a carefully picked fuse before blowing up itself from overheating.
I think others have written comments along this line already but my problem with this is that while the zener diode is sinking current for a while it will probably die eventually. If it dies short you'll get a nice burn mark somewhere, if it dies open you're back to square one. This seems like a good place to get on lcsc (jlc pcb's component arm) and see if there is some weird chip out there that integrates the over voltage detection and a MOSFET to switch off the supply.
You're totally right. I thought I read in the datasheet for that diode that it'll fail to short, which is better than open. Hopefully at that point the power supply will fail in some way. But I agree a better way would be to have a fuse or some other fancy component to stop the current. I've gotten a couple of recommendations so I might look into that. Cheers.
"Off with their heads". Hahaha! Nice quick video! Lots of retro repair UA-camrs' videos have been creeping up to the 60 minute marks -- which isn't bad!! -- It just gets hard to find quick videos to watch during lunch or breaks. That sure was a quick fix!
Glad you liked it! I'm enjoying a mix of shorter and slightly longer, but I really try hard not to go past 25-30 minutes as it starts getting too long. I guess that's why I end up stuck doing multi-parters for some repairs, but I think they call for it. Anyway, glad to see this format is also working. Cheers.
What a handy diagnostic tool :-D Breaking the power + line from the power socket and bridging it with a fuse+holder would make sure the psu did not stay shorted for too long, and the diode would likely survive.
@@NoelsRetroLab Perhaps the screw together type of holder with a wire from each end could be tied to other wiring using small Cable ties or Zip ties. Or get some pcb fuses, they often look like a small resistor body type. Worth a bit of research so you know they exist :-D
@@NoelsRetroLab The remaining RAM might be OK! It may just be that over voltage in that case! I have left other MT RAM chips on boards when I've found one chip faulty and they are still running. That cart is great! I need to get one of those!
I think it is a bad place to mount the zener diode due to track, leading to the positive lead of zener, is too thin. So, this track could be burnt and zener diode will be ignored. Obviously, it depends of external PSU power. BTW, I'd better use TVS (transient voltage suppressor) with appropriate fuse instead of zener, placed in reliable, "powerful" part of the board (to thick tracks or even polygons, if any). p.s. Thank you for your content. I really like your channel.
Interesting. I didn't think about the thickness of the tracks leading up to it. I don't know if that would break in case of a short, but it's certainly worth considering!
Yes, so I've been told. I haven't had a horrible experience with them on the CPC, that's why I didn't rush out to repair them. Besides, the ones left standing after that must be tougher right? 😃
I have an Amstrad PCW with these symptoms. It just beeps and displays a garbled screen when booted. I also suspect the RAM. I must get around to swapping the chips for some new ones. PCW RAM is usually very reliable and this is the only faulty board I’ve found.
Great video! I will need to make one of those adapters. Seems like it really shortened your trouble shooting. The diode looked like a good idea. Too bad the manufacturer of the connector didn’t notch it so that it would be keyed and impossible to reverse.
@@NoelsRetroLab maybe that's just me, but i get the feel of being in a doctors office listening to you speak. nothing more significant or solid than that. it's not bad or anything.
It's funny you mention that because that's already in my list. It should be a quick/shorter video, so I'm hoping to make it soon. The spoiler is to use a Meanwell power supply with +5 and +12 and some kind of enclosure.
I am wondering about the utility of the diode overvoltage protection. As it has no indication of overvoltage and a missing 12v rail wouldn't be apparent until disk access, meaning the diode has to last for "a while" without blowing. And once it blows, the 12v will be dumped into those poor ram chips. A fuse or a crowbar circuit would be better in my opinion, if a lot harder to make and fit.
Question for you: If you plug 12V on the 5V connector with the Zener installed, the zener will dissipate 5V (12V - 7V that you measured). Depending on the PSU being used, the Zener should be able to dissipate that power or it will burn. How do you account for that? Thanks for the video as usual!
The diode will dissipate anything over 7V in the form of heat, so it will eventually fail. This could be really quickly depending on the voltage and current rating of the power supply. From the datasheet, it looks like it should fail to a short, so that will pass on the problem to the power supply and have that fail (or, if it's a decent one, blow a fuse there).
20 years ago I tried a RAM chip change and I accidently did this. 12V on the 5V line. I thought CPC is dead because no video out signal anymore. I thought all the ICs will be dead when they get shocked by 12V. Can this be or just like your case here only some are affected ?
Thanks! This adapter is the same thing, just in PCB form and much neater. Dave Anders made that PCB after watching the video and sent it to me. It's very cool. You can even toggle the banks with a couple of jumpers.
This may work if the 12V power supply actually sources 400 mA maximum, like the limit value you have set on your bench supply. If the power supply can source let's say 2A, your poor zener diode will dissipate 7*2=14W and probably it will die right away.
True. But the diode will short when it fails (or so says the datasheet) which will still protect the rest of the board (and possibly blow up the power supply).
@@NoelsRetroLab You are right, the initial failure mode is usually short, but then everything can happen. Perhaps it goes a little bit beyond the scope of this video, but if you are curious you can see this for example: www.vishay.com/docs/88440/failurem.pdf
@@lorenzol.8798 Awesome, I'll check that out for sure. Yes, this was definitely just a "quick fix" video. There are much better ways to prevent overvoltage. I'll probably end up revisiting it sometime soon going over the different options and implementing something safer. Cheers!
That's a good question. I think it's a good idea, and it would generate better voltage levels on the Amstrad. The problem is that you need to change the input voltage to be at least +2V more, so +7V. At that point I guess you might as well make it center negative and use a ZX Spectrum power supply. It's not a bad idea... The only problem is potential confusion and then getting used to it and plugging that power supply on a new Amstrad that doesn't have the mod and BOOOM! 😃
@@NoelsRetroLab I forgot to calculate the voltage drop, and if the power supply is good enough, there’s some very tiny buck converters which can output constant 1Amp, but I have no idea since my home city Hong Kong almost have no Amstrad for sales back in 80s, at that time Apple II clone dominated the computer market.
A linear voltage regulator requires the input voltage to be higher than the output so it would require injecting more than 5V for it to work properly. One could, of course, make a mod to take the 12V input, use the regulator to reduce to 5V and use that to power the logic board but that would require a different power supply and I think the idea is to keep the whole setup as original as possible.
@@adilsongoliveira if the power supply output is high enough, is it possible to use 12v rail to split a 5v by buck converter? I have no idea the output number of the power supply.
@Mr Guru yes you’re right, old 7805 generate too much heat and I bought some switching replacement. The specification is quite nice and should work well but not sure if any trouble in this kind of circuit since switching type have higher noise.
Hi! No, I haven't. I know it's possible and I think it involves making a whole new SD card with the set up. I own one but I'm not a fan of the M4 for the most part I'm afraid 😃
Hi, I just made a mistake and hooked an external 12v connector into the keyboard and now the computer don"t run anymore when hooked properly. Should I follow your exemple? If so, what/where can I buy the needed remplacement & tools? Great video thank you.
Yes, you should do a similar diagnostic and fix it. Hopefully it's just a blown RAM IC or two and nothing more serious. Check out the Amstrad diagnostics ROM if you have a way to use it. Good luck!
Ah, good point! I didn't think much of it because of how this board was fried (by swapping that Molex connector) so it got a dose of +12V. But the polarity of the barrel jack of is actually center negative. So if you did manage to connect it with an adaptor to the 5V jack, the diode would act as a short right away (and probably fail within seconds, but hopefully to a short).
@@NoelsRetroLab The reason I asked is I have a CPC6128 but the drive didn't work when I got it. As I don't have the one of the monitor I was trying to figure out what polarity the 12V was. I thought I had killed it because I though I got the polarity wrong but I am sure I had it center negative. I have two drives one dead and terrified I am going to kill the second because I don't know how the 12V power is to be delivered. I do have a Gotek for it but have yet to try it.
Strict 5v, there’s no voltage regulation on the circuit board, so if the 5v supply is too far out of spec (too high, too low), then the ICs would either fail to function correctly, or will permanently fail
No, I'm using this one: www.cpcwiki.eu/forum/amstrad-cpc-hardware/quick-and-dirty-ram-test-for-cpc/ I'm planning on covering the Dandanator in a future video and I'll talk about how to add it (hint: It needs to be padded to 16KB exactly).
@@NoelsRetroLab Thank you for your reply :-). I will wait for your video on this as I do not have a clue on how to pad it to 16kb ^-^. Looking forward for you explanation on a future video release! Great channel by the way :-D
I use this type of transient voltage suppressor also in my circuit ( bitbucket.org/UweGottschling/oak65c02/src/master/Oak65c02_KICAD/65xx-CPU/6502.pdf ). This diode is a simple surge protection, but not entirely perfect. The tolerance range of the diode is large and in the worst case the voltage is 7.14V. The absolute maximum rating for the Z80 is around 7 Volt.
Easy. The ribbon cable is identical, just connect to the DD expansion edge connector. However, you will only get the same disk capacity as with the 3inch disks, because the CPC can only use 40 tracks and can not switch sides, and you cannot flip the disks over. You will need to get an alternative DOS to use all 720kbytes, and a side switch to access the second side. The side switch connects the side select line of the ribbon cable to 0volts for side one and 5volts for side two. Unfortunately I can't remember offhand which line it is! As far as an alternative DOS, I wrote one that was sold by Phil Craven's Microstyle under the name MS800. It was favourably reviewed by Amstrad Action . There is a ROM-based DOS called Parados that is still available
I haven't, but like Ewen mentioned, it's definitely possible. There should be quite a few resources out there. It's not something I've done because I love the 3" disks because of nostalgia 😃
That RAM test shows one bar for every bit in the data bus. Starting at the bottom with the least-significant bit (Gerald, the author explained it here: www.cpcwiki.eu/forum/amstrad-cpc-hardware/quick-and-dirty-ram-test-for-cpc/ ). Then you need to look at the CPC schematic and find out which bit corresponds to which IC.
That one is a bit more complicated. It has several internal voltage regulators, but in theory yes. It's just a matter of finding the right place to put it.
That would be a good start. Or even use even more different connectors. But if you have a ZX Spectrum around, it's always a possibility that you connect the wrong supply on the DC jack.
I do enjoy the shorter video fixes. The content is great but the 5min timer dont like. Would never encourage hobbisist to fix a problem fast. It invites error. Thanks
You can also put a fuse on the power line before the zener. That way when the voltage goes over the zener's nominal value the zener will sink current to ground blowing the fuse and no voltage will go to the power rail.
I like that. It would be a matter of picking the right fuse, right? Probably something like 2A slow? The biggest problem is mounting it neatly on the board somewhere.
That's the right solution, maybe even use a polyfuse instead of a conventional fuse.
A signal level mosfet, zener, two bjts and a few resistors can make a proper over voltage cut off circuit. Using SMD stuff it can be made pretty small.
@@NoelsRetroLab search 'pico fuse', choose a current value that's 135% of max load (don't forget cart & other peripherals that load the power supply). This circuit is better than nothing but the diode you are using is intended for spike protection not continuous over-voltage (maybe a few in parallel would be better). But you would be better off using an SCR crowbar circuit, it's only a few parts. (prototype tip: use a 4.8-6.3v bulb instead of a fuse to test trigger voltage)
@@NoelsRetroLab Pin the original comment please.
It's very informative.
Cool stuff. Thanks Noel.
Glad you liked it!
This is fast becoming my favourite retro hardware repair channel :-) Clear influences from Adrian Black but that's no bad thing. Keep up the great work.
Adiós!
Thank you! Adrian was a huge influence on my starting this channel, so that's not a coincidence 😃 Glad you're enjoying it. Cheers!
@@NoelsRetroLab Yourself and Adrian are my two favourite channels these days. Both informative and not overly pretentious :D Thank you for the entertainment :)
Excellent video!
Glad you liked it!
Thank you for showing this P6KE TVS diode! It's quite more resilient in terms of power handling than I expected, it should certainly blow a carefully picked fuse before blowing up itself from overheating.
You're welcome. And you're right about having a fuse somewhere (in the power supply, or on the board itself) to make it as safe as possible.
I don't know why youtube recommends your channel, but here I am and I really liked the content. Congratulations! Greetings from Brazil!
Awesome! Thank you!
nice quick fix noel
I think others have written comments along this line already but my problem with this is that while the zener diode is sinking current for a while it will probably die eventually. If it dies short you'll get a nice burn mark somewhere, if it dies open you're back to square one. This seems like a good place to get on lcsc (jlc pcb's component arm) and see if there is some weird chip out there that integrates the over voltage detection and a MOSFET to switch off the supply.
You're totally right. I thought I read in the datasheet for that diode that it'll fail to short, which is better than open. Hopefully at that point the power supply will fail in some way. But I agree a better way would be to have a fuse or some other fancy component to stop the current. I've gotten a couple of recommendations so I might look into that. Cheers.
Nice video. Thanks. These"Quick Fix" videos are addictive.
Glad you like them!
Great job yet again Noel!.. Thanks :)
...easy to get it done so quick when you have all the right tools at hand!
That's true! Without the tools it would have been a much bigger job.
Watching your videos makes it look so easy...i wish i had half of your skills in that! Keep up the good work!!!
"Off with their heads". Hahaha! Nice quick video! Lots of retro repair UA-camrs' videos have been creeping up to the 60 minute marks -- which isn't bad!! -- It just gets hard to find quick videos to watch during lunch or breaks. That sure was a quick fix!
Glad you liked it! I'm enjoying a mix of shorter and slightly longer, but I really try hard not to go past 25-30 minutes as it starts getting too long. I guess that's why I end up stuck doing multi-parters for some repairs, but I think they call for it. Anyway, glad to see this format is also working. Cheers.
Fine work, Noel!
Fit Bwack Savers. I did this on most sensitive machines. Plus a PFET before it to prevent RP.
What a handy diagnostic tool :-D
Breaking the power + line from the power socket and bridging it with a fuse+holder would make sure the psu did not stay shorted for too long, and the diode would likely survive.
Agreed. That'd be great. The only complication is finding the place to put the fuse securely on the board.
@@NoelsRetroLab Perhaps the screw together type of holder with a wire from each end could be tied to other wiring using small Cable ties or Zip ties.
Or get some pcb fuses, they often look like a small resistor body type.
Worth a bit of research so you know they exist :-D
Great video =D I would replace all that MT RAM in that case! It fails a LOT in C64 boards!
Thanks! Yes, so I've heard! People seem to hate the MT RAM!
@@NoelsRetroLab The remaining RAM might be OK! It may just be that over voltage in that case! I have left other MT RAM chips on boards when I've found one chip faulty and they are still running. That cart is great! I need to get one of those!
Those TVS diodes should also have the nice effect of protecting the rails against static discharge and fast transients.
Oh interesting. I hadn't though of that!
Another retrocomputer saved. Good work! :)
Yes! Thank you! 👍
Excellent video again! thank you!
My pleasure!
Almost at 10k subs! Still loving your content...
Thank you! I see you've been keeping close tabs on the subscriber numbers 😃 Yeah, hopefully in a couple more videos I'll break through 10K 🎉
Cool T-shirt....
I think it is a bad place to mount the zener diode due to track, leading to the positive lead of zener, is too thin. So, this track could be burnt and zener diode will be ignored. Obviously, it depends of external PSU power. BTW, I'd better use TVS (transient voltage suppressor) with appropriate fuse instead of zener, placed in reliable, "powerful" part of the board (to thick tracks or even polygons, if any).
p.s. Thank you for your content. I really like your channel.
Interesting. I didn't think about the thickness of the tracks leading up to it. I don't know if that would break in case of a short, but it's certainly worth considering!
Fuse on input, then a TVS diode.
Over curent cuts the fuse, protect the board and the suplly
Yes, that seems the way to go. Mounting it on the board is trickier, so that might need its own mini-pcb unfortunately.
I would have replaced all the MT rams..... they are bound to fail sooner or later. Nice use of Voltage diode.
Yes, so I've been told. I haven't had a horrible experience with them on the CPC, that's why I didn't rush out to repair them. Besides, the ones left standing after that must be tougher right? 😃
@@NoelsRetroLab Well.... for how long though? ;)
I have an Amstrad PCW with these symptoms. It just beeps and displays a garbled screen when booted. I also suspect the RAM. I must get around to swapping the chips for some new ones. PCW RAM is usually very reliable and this is the only faulty board I’ve found.
I don't know much about PCW repairs, but what you're saying sounds about right. We'll see what happens when I finally get around to restoring my PCW!
Great video! I will need to make one of those adapters. Seems like it really shortened your trouble shooting. The diode looked like a good idea. Too bad the manufacturer of the connector didn’t notch it so that it would be keyed and impossible to reverse.
Which adapter, the Dandanator? Yes, those are awesome! You can use any ROM board that lets you replace the lower ROM, but I particularly like those.
Use to call them idiot diodes can use a diode to prevent wrong polarity as well
🤣
Great video! Exactly what I'm looking for. Where do you get your testing equipment and what are they called please?
I plugged in the power upside-down on a disc drive on my Amiga, it blew the disc control chip with a flash an a bang. LOL.
Ouch! That must have been a lot of current or higher voltage I imagine.
@@NoelsRetroLab Yeah. It frightened the life out of me. I never bothered fixing it, it's an A500, I use A1200's now. :)
Thanks, Noel :)
Welcome!
you speak with the cadence of a medical doctor.
Whoa, that's a new one to me! 😃What exactly does that mean? Not offended at all, just very curious.
@@NoelsRetroLab maybe that's just me, but i get the feel of being in a doctors office listening to you speak. nothing more significant or solid than that. it's not bad or anything.
@@slipcurve1410 Haha fair enough! 😃
Hehe editing it down to 5 mins doesn't count :)
Haha, true, it's not 5 minute fix. It's a 5-minute video of a quick fix 😃
Presss F to pay respects. These RAMs our finest.
I'd like a video suggesting good psu s for the cpc 6128. Both 5 and 12v is hard to find in one psu
It's funny you mention that because that's already in my list. It should be a quick/shorter video, so I'm hoping to make it soon. The spoiler is to use a Meanwell power supply with +5 and +12 and some kind of enclosure.
I am wondering about the utility of the diode overvoltage protection. As it has no indication of overvoltage and a missing 12v rail wouldn't be apparent until disk access, meaning the diode has to last for "a while" without blowing. And once it blows, the 12v will be dumped into those poor ram chips. A fuse or a crowbar circuit would be better in my opinion, if a lot harder to make and fit.
Question for you: If you plug 12V on the 5V connector with the Zener installed, the zener will dissipate 5V (12V - 7V that you measured). Depending on the PSU being used, the Zener should be able to dissipate that power or it will burn. How do you account for that? Thanks for the video as usual!
The diode will dissipate anything over 7V in the form of heat, so it will eventually fail. This could be really quickly depending on the voltage and current rating of the power supply. From the datasheet, it looks like it should fail to a short, so that will pass on the problem to the power supply and have that fail (or, if it's a decent one, blow a fuse there).
@@NoelsRetroLab ahhh - I did not know it would fail to a short. That makes total sense then. Thanks!
nice amstrad (amsoft) t-shirt!!!
20 years ago I tried a RAM chip change and I accidently did this. 12V on the 5V line. I thought CPC is dead because no video out signal anymore. I thought all the ICs will be dead when they get shocked by 12V.
Can this be or just like your case here only some are affected ?
Hi i have an egg incubator which is 110 v and put it in 230 v..so is it repairable?
Excellent video, is there a reason why you didn't use the adapter you made a while back to force the second bank of RAM? I'm just curious ;)
Thanks! This adapter is the same thing, just in PCB form and much neater. Dave Anders made that PCB after watching the video and sent it to me. It's very cool. You can even toggle the banks with a couple of jumpers.
This may work if the 12V power supply actually sources 400 mA maximum, like the limit value you have set on your bench supply. If the power supply can source let's say 2A, your poor zener diode will dissipate 7*2=14W and probably it will die right away.
True. But the diode will short when it fails (or so says the datasheet) which will still protect the rest of the board (and possibly blow up the power supply).
@@NoelsRetroLab You are right, the initial failure mode is usually short, but then everything can happen. Perhaps it goes a little bit beyond the scope of this video, but if you are curious you can see this for example: www.vishay.com/docs/88440/failurem.pdf
@@lorenzol.8798 Awesome, I'll check that out for sure. Yes, this was definitely just a "quick fix" video. There are much better ways to prevent overvoltage. I'll probably end up revisiting it sometime soon going over the different options and implementing something safer. Cheers!
I wish I could replace chips that fast
Me too! 😃😃
Why not add a fuse as well as the zener, say 2A? So when the zener conducts, the excessive current should blow the fuse and cut the power completely.
Yes, that'd be a good idea. It's just a matter of finding a good place to add it securely to the board.
Thanks for the video =)
My pleasure!
Question: Why not put a 7805 (or switching 7805 equivalent) so the 5V rail will always remain at 5V or just kill the 7805 and stop damaging the board?
That's a good question. I think it's a good idea, and it would generate better voltage levels on the Amstrad. The problem is that you need to change the input voltage to be at least +2V more, so +7V. At that point I guess you might as well make it center negative and use a ZX Spectrum power supply. It's not a bad idea... The only problem is potential confusion and then getting used to it and plugging that power supply on a new Amstrad that doesn't have the mod and BOOOM! 😃
@@NoelsRetroLab I forgot to calculate the voltage drop, and if the power supply is good enough, there’s some very tiny buck converters which can output constant 1Amp, but I have no idea since my home city Hong Kong almost have no Amstrad for sales back in 80s, at that time Apple II clone dominated the computer market.
A linear voltage regulator requires the input voltage to be higher than the output so it would require injecting more than 5V for it to work properly. One could, of course, make a mod to take the 12V input, use the regulator to reduce to 5V and use that to power the logic board but that would require a different power supply and I think the idea is to keep the whole setup as original as possible.
@@adilsongoliveira if the power supply output is high enough, is it possible to use 12v rail to split a 5v by buck converter? I have no idea the output number of the power supply.
@Mr Guru yes you’re right, old 7805 generate too much heat and I bought some switching replacement. The specification is quite nice and should work well but not sure if any trouble in this kind of circuit since switching type have higher noise.
I'm getting a surgeon-vibe.
Hey Noel, have you tried the RAM check ROM with the M4?
Hi! No, I haven't. I know it's possible and I think it involves making a whole new SD card with the set up. I own one but I'm not a fan of the M4 for the most part I'm afraid 😃
Hi, I just made a mistake and hooked an external 12v connector into the keyboard and now the computer don"t run anymore when hooked properly.
Should I follow your exemple? If so, what/where can I buy the needed remplacement & tools? Great video thank you.
Yes, you should do a similar diagnostic and fix it. Hopefully it's just a blown RAM IC or two and nothing more serious. Check out the Amstrad diagnostics ROM if you have a way to use it. Good luck!
keep up the video's :)
Just a quick question. What is the polarity of the 12V barrel jack?
Ah, good point! I didn't think much of it because of how this board was fried (by swapping that Molex connector) so it got a dose of +12V. But the polarity of the barrel jack of is actually center negative. So if you did manage to connect it with an adaptor to the 5V jack, the diode would act as a short right away (and probably fail within seconds, but hopefully to a short).
@@NoelsRetroLab The reason I asked is I have a CPC6128 but the drive didn't work when I got it. As I don't have the one of the monitor I was trying to figure out what polarity the 12V was. I thought I had killed it because I though I got the polarity wrong but I am sure I had it center negative. I have two drives one dead and terrified I am going to kill the second because I don't know how the 12V power is to be delivered. I do have a Gotek for it but have yet to try it.
Hello! I managed to source only 5.5v power adapter that enough amperage.
Would 5.5V damage my CPC 464, or I have to look for strict 5V?
Strict 5v, there’s no voltage regulation on the circuit board, so if the 5v supply is too far out of spec (too high, too low), then the ICs would either fail to function correctly, or will permanently fail
@@bruce_just_ Too late. AY-3-8912 is an expensive bugger. 😥
What memory test rom are you using with the Dandanator? Is it the one that comes already by default?
No, I'm using this one: www.cpcwiki.eu/forum/amstrad-cpc-hardware/quick-and-dirty-ram-test-for-cpc/ I'm planning on covering the Dandanator in a future video and I'll talk about how to add it (hint: It needs to be padded to 16KB exactly).
@@NoelsRetroLab Thank you for your reply :-). I will wait for your video on this as I do not have a clue on how to pad it to 16kb ^-^. Looking forward for you explanation on a future video release! Great channel by the way :-D
I use this type of transient voltage suppressor also in my circuit ( bitbucket.org/UweGottschling/oak65c02/src/master/Oak65c02_KICAD/65xx-CPU/6502.pdf ). This diode is a simple surge protection, but not entirely perfect. The tolerance range of the diode is large and in the worst case the voltage is 7.14V. The absolute maximum rating for the Z80 is around 7 Volt.
Have you ever tried to install a 3,5" Floppy to the Amstrad CPC ?
Easy. The ribbon cable is identical, just connect to the DD expansion edge connector. However, you will only get the same disk capacity as with the 3inch disks, because the CPC can only use 40 tracks and can not switch sides, and you cannot flip the disks over. You will need to get an alternative DOS to use all 720kbytes, and a side switch to access the second side. The side switch connects the side select line of the ribbon cable to 0volts for side one and 5volts for side two. Unfortunately I can't remember offhand which line it is! As far as an alternative DOS, I wrote one that was sold by Phil Craven's Microstyle under the name MS800. It was favourably reviewed by Amstrad Action . There is a ROM-based DOS called Parados that is still available
I haven't, but like Ewen mentioned, it's definitely possible. There should be quite a few resources out there. It's not something I've done because I love the 3" disks because of nostalgia 😃
What is that tool you used to remove the bad RAM DIPs?
It's a desoldering gun. This one is a ZD-915. Highly recommended!
Ive got a 6128 green screen with black boarder would like to test the ram but the Dandanator CPC is out of stock, Anyone in uk selling them?
You could just write the test ROM to an EPROM and put it instead of the Amstrad ROM. I does require desoldering the ROM though.
@@NoelsRetroLab Thanks for the reply I will try this. I'm new to repairing computers. Just got hold of a 6128 and I determined to get it fixed :)
How did you know from looking at the screen which ram chips were bad?
That RAM test shows one bar for every bit in the data bus. Starting at the bottom with the least-significant bit (Gerald, the author explained it here: www.cpcwiki.eu/forum/amstrad-cpc-hardware/quick-and-dirty-ram-test-for-cpc/ ). Then you need to look at the CPC schematic and find out which bit corresponds to which IC.
@@NoelsRetroLab Thanks for the info.
@Mr Guru Oh wow! Grrrr... UA-cam! Thanks for the heads up. Fixing it now.
Does that mean that if/when the diode does pop due to over voltage/ over heating the board with get the full 12v?
My understanding is that when the diode fails, it shorts, so the circuit will still be protected (but the power supply won't be happy).
Would a similar application work for gx4000?
That one is a bit more complicated. It has several internal voltage regulators, but in theory yes. It's just a matter of finding the right place to put it.
The real fix is to only ever use nylon Molex connectors.
That would be a good start. Or even use even more different connectors. But if you have a ZX Spectrum around, it's always a possibility that you connect the wrong supply on the DC jack.
Crowbar circuit would do a better job, won't overheat!
True. I was just looking for a bandaid, something really simple to implement (and hopefully something that will never be needed 😃).
"5 minute fix" No, it wasn't. IT was only five minutes because it was edited. Not real time.
Yes, clearly. I'm not THAT fast 😃 But it was 5 minutes worth of a video I hope. Nobody would have wanted to sit through the 30 minutes it really took.
@@NoelsRetroLab I am silly enough to sit through a 30 min video lol
I do enjoy the shorter video fixes. The content is great but the 5min timer dont like. Would never encourage hobbisist to fix a problem fast. It invites error. Thanks
The timer was a bit of a joke, but fair enough 😃 Clearly it too me a lot longer than 5 minutes of real time (probably about 30 minutes for real).