That "fault" sounds like the 50v pin next the GRD on the display cable IIRC. Can't remember the model but I do remember _someone_ ranting about it at the time. You are a good educator Paul and I like your work. 👍 I hope you are making a very comfortable living from your repairs because you deserve it.
Not comfortable yet but getting there. The fault you mention was on almost all the intel MacBooks starting in 2016, and again once apple released the m1 that one was definitely corrected. All the voltage pins are surrounded by ground pins now. We think these are stupid mistakes but I think apple knows exactly what they are doing.
Just did one today - fortunately repairable, though unfortunately it was fixed only 2 months ago; different cap this time but again 2V5 NAND :\ They're a regular occurence here along with the A1990's. EDIT: UPGRADE Your FlexBV! I'm almost certain you've got access to v5 now, and you'll get Teboview support too :D
Why didnt they put some sort of clamping diode on the 2V supply rail? I always do that when designing a board, I thought this was normal with low voltage electronics. And I am not even talking about chips that hold someones data. The idea is that the diode starts conducting from the overvoltage, gets overloaded and fails short. This obviously shorts the destructive voltage to ground and saves the IC.
What I have noticed about Apple design is that they never, never, think about failure modes. They put zero thought into dust infiltration, moisture, or liquid. All that they care about is that when you use the device in the manner and place they expect you to use it, it works. Anything else doesn't matter.
Another thing I have sometimes wondered is if it would be possible or advisable to install a 2.5V Zenner diode in parallel with 1 of the caps. That way, if the chip did go bad or corrode, it would either send the excess over to ground or just short bring the rail down. But it wouldn't send 12V to the NAND
Sadly it probably won't help much; the VREG is capable of enough current that it would overpower the zener and those zeners have terribly soft knees below about 5V6. Even if you put something like a crowbar circuit with a TL431 as the reference you'd probably find it still doesn't save the day every day. Best solution I see here is a physical redesign and keep the whole VREG controller away from the liquid :( Art of Electronics has a very good section on dealing with overvoltage situations and the pros/cons.
Happy New Year Paul. This atrocious design decision and makes FlexGate seem like a minor transgression. Apple, why??? Did you request Texas Instruments put the 2.5v NAND power rail right next to the ~12v PPBUS power rail, it wouldn’t surprise me at all.
For those wondering how dust can conduct electricity, Keep in mind that most of the dust is fiber, not dirt from outside, and you also get a lot of dander. Most of the fiber dust that gets inside has had contact with the skin. The skin has oil, and salt to keep it moist. This mixture of oil and salt also helps keep the dust moist. Think of how damp rid crystals (calcium chloride) pull moisture out of the air. By definition, a salt is an acid and base combined together. Sodium chloride, is sodium (base), and chloride (acid). Calcium and Sodium are both metals. The salt not only pulls moisture out of the air, but it also greatly increases the conductivity of water. As the solder dissolves, the conductivity goes up even higher. And then you also have the obvious situation where you get condensation from going from a cool environment to a warm steamy environment. Pure water has practically no conductivity, but a little salt, and metal oxides goes a long way.
Yea they fixed it, these chips are still used in newer machines but they are not near the edge. But they are still not underfilled or sealed in any way so still a danger.
I hate last 2 intel MacBook's, I think they are just waiting to be door step in some point, so much hardware problems If you replace SSD do you have to reprogram as well ?
Commented after his short introduction to the video, yes. Thought about this a few times over the years and wondered why you mostly have to DIY such a thing with consumer products. PS: Still have a fully functional 2012 MBP A1286 where even the original battery is still above 80 %. Am sad about this fragility in more current devices.
Why apple doesn’t use underfill in this case where would help but really often apply it for example under Tristar chips in iPads where is no moisture…the answer is: to make repair so hard that U gonna buy new device🤦🏻♂️
Sadly those chips are bad.. idk why and idk how but I've seen cases where there isn't any dust, liquid, corrosion or anything and they go bad even on other models than A2141.
That "fault" sounds like the 50v pin next the GRD on the display cable IIRC. Can't remember the model but I do remember _someone_ ranting about it at the time. You are a good educator Paul and I like your work. 👍 I hope you are making a very comfortable living from your repairs because you deserve it.
Not comfortable yet but getting there.
The fault you mention was on almost all the intel MacBooks starting in 2016, and again once apple released the m1 that one was definitely corrected. All the voltage pins are surrounded by ground pins now. We think these are stupid mistakes but I think apple knows exactly what they are doing.
Just did one today - fortunately repairable, though unfortunately it was fixed only 2 months ago; different cap this time but again 2V5 NAND :\
They're a regular occurence here along with the A1990's.
EDIT: UPGRADE Your FlexBV! I'm almost certain you've got access to v5 now, and you'll get Teboview support too :D
Don’t delay, upgrade your FBV today
haha@@MrPnew1
This just makes me mad. Borderline criminal negligence to design it that way! Horrible to loose data :(
Think different indeed...
@JimAllen-Persona Not from me :)
Interesting to know. Thanks Paul.
This is how to teach people, thankyou
Your thumbnail look so much like Luis old video thumbs. 😅
Why didnt they put some sort of clamping diode on the 2V supply rail? I always do that when designing a board, I thought this was normal with low voltage electronics. And I am not even talking about chips that hold someones data. The idea is that the diode starts conducting from the overvoltage, gets overloaded and fails short. This obviously shorts the destructive voltage to ground and saves the IC.
What I have noticed about Apple design is that they never, never, think about failure modes. They put zero thought into dust infiltration, moisture, or liquid. All that they care about is that when you use the device in the manner and place they expect you to use it, it works. Anything else doesn't matter.
Another thing I have sometimes wondered is if it would be possible or advisable to install a 2.5V Zenner diode in parallel with 1 of the caps. That way, if the chip did go bad or corrode, it would either send the excess over to ground or just short bring the rail down. But it wouldn't send 12V to the NAND
Sadly it probably won't help much; the VREG is capable of enough current that it would overpower the zener and those zeners have terribly soft knees below about 5V6. Even if you put something like a crowbar circuit with a TL431 as the reference you'd probably find it still doesn't save the day every day. Best solution I see here is a physical redesign and keep the whole VREG controller away from the liquid :(
Art of Electronics has a very good section on dealing with overvoltage situations and the pros/cons.
Happy New Year Paul. This atrocious design decision and makes FlexGate seem like a minor transgression. Apple, why??? Did you request Texas Instruments put the 2.5v NAND power rail right next to the ~12v PPBUS power rail, it wouldn’t surprise me at all.
For those wondering how dust can conduct electricity, Keep in mind that most of the dust is fiber, not dirt from outside, and you also get a lot of dander. Most of the fiber dust that gets inside has had contact with the skin. The skin has oil, and salt to keep it moist. This mixture of oil and salt also helps keep the dust moist. Think of how damp rid crystals (calcium chloride) pull moisture out of the air. By definition, a salt is an acid and base combined together. Sodium chloride, is sodium (base), and chloride (acid). Calcium and Sodium are both metals. The salt not only pulls moisture out of the air, but it also greatly increases the conductivity of water. As the solder dissolves, the conductivity goes up even higher.
And then you also have the obvious situation where you get condensation from going from a cool environment to a warm steamy environment. Pure water has practically no conductivity, but a little salt, and metal oxides goes a long way.
Hi Paul, great video. Did Apple rectify this issue in later models? (change the chip or moved it)?
Yea they fixed it, these chips are still used in newer machines but they are not near the edge. But they are still not underfilled or sealed in any way so still a danger.
I hate last 2 intel MacBook's, I think they are just waiting to be door step in some point, so much hardware problems
If you replace SSD do you have to reprogram as well ?
Would a service adding protection like conformal coating to new, properly working devices be a stupid idea?
See 21:40 that is exactly what Paul does. Did you comment before watching all the video??? 🤣
Commented after his short introduction to the video, yes. Thought about this a few times over the years and wondered why you mostly have to DIY such a thing with consumer products.
PS: Still have a fully functional 2012 MBP A1286 where even the original battery is still above 80 %. Am sad about this fragility in more current devices.
Why apple doesn’t use underfill in this case where would help but really often apply it for example under Tristar chips in iPads where is no moisture…the answer is: to make repair so hard that U gonna buy new device🤦🏻♂️
Sadly those chips are bad.. idk why and idk how but I've seen cases where there isn't any dust, liquid, corrosion or anything and they go bad even on other models than A2141.
you can't save them All 😞
My track record for A2141 is terrible. 75% of them are all dead like this.
@@A2000MHz Not good At least you tried to say people's data Unlike apple They should have a 20 year Recall On this