I’ll post the information here, because not everyone looks at Reddit. Posted a few minutes ago and, as always, please keep in mind: Nothing has been verified.
Edit: Not mine, the information came from Reddit.
1 Like
Good to know, and seemingly a favorable development. Did you purchase yours direct from Bambulab?
BestBuy briefly had it back in stock, but now it’s sold out again.
Not mine, the information came from Reddit.
I’m guessing a replacement board would cost less than $30, even with a 50% gross margin built-in for Bambulab. So… I went to look for one on the US Bambulab store, and I don’t see it anywhere:
How weird is that? If it were simply out of stock, there’d still be a listing for it.
Rather than all this angst, worry, and indecision, I’d be willing to buy it as cheap insurance for $30 out of my pocket and just be done with it and move on without wasting more time over it and never looking back.
Seems like Bambulab could have added a copper pour to a PCB layer as a heat spreader and kept the NTC if they had wanted to, making an incremental improvement to cover an at-the-margin issue, if that’s what it was. But from appearances they decided to replace it with wire instead, if that’s what it is. I checked to see whehter it might theoretically be some kind of wire that acts like an NTC, and found that such wire do exist:
but I can’t tell from looking at just a photo whether the replacement wire is something fancy like that or just a plain old piece of wire. Why might this be plausible? Because if you compare v4 against v2, they removed all the landing pads for the components they excised from the board, EXCEPT for the NTC landing pad.
From:
To now:
If it’s just a plain piece of wire, why do that? Why not remove the NTC but leave the landing pad, to be crossed with wire, rather than just fix the PCB not to require that wire at all as you update the board anyway? Or is Bambulab leaving an option open to install a better NTC in the future, if it ever finds one, and doing it this way is simply a stopgap? I’m guessing not. So, if the wire is not an NTC wire, could it be a particular kind of wire that’s functioning as anything else, like maybe a fuse?
Anyone understanding what I’m trying to say here? There’s probably a more elegant way to say it.
Anyway, I feel like I’ve done my share. Time for someone else to pickup the torch and carry it. I’ve timed out on this one and need to switch gears to other matters.
1 Like
Search Amazon, there are tons of flame retardant pads and papers and none are concrete blocks or ceramic tiles.
Well, one last thing I just now noticed (and sorry if anyone already pointed this out, but if so, I must have speed readed past it):
The boards appear to have a date code (DC) printed on it, in plain text, staring us right in the face as bold as brass. By my reckoning, this one would be week 38 of the year 2023:
Version 2 (above) would be week 26 of year 2025, and version 4 would be week 32 of year 2025. Because of the date codes on some of the other boards in the earlier photos, I suspect it’s the manufacturing date of the board rather than a design date, but I could make up plausible sounding stories where it could be a marker for a minor revision, so go figure.
Actually, this board:
in week 46 of 2024 looks like a partial step in component populating between board red circle number 1 and board red circle number 2, even though all three are version 2, assuming we’re correctly inferring the meaning of _V2.
It is not listed for probably the same reason as in the last recall: the end user is not actually "allowed" to change the circuit board for legal and insurance reasons, because it is part of the main power supply.
“Consumers should immediately stop using the recalled 3D printers, and contact Bambu Lab to arrange either (1) return of their A1 3D printers for a full refund, or (2) for free replacement of the heatbed and cable by a local electronics repair facility. Consumers requesting a full refund will be sent a prepaid shipping label to return the printer. Consumers should not attempt to fix the recalled product themselves. Bambu Lab has contacted all known purchasers.”
Now, of course, the question arises as to why some power supplies are available as spare parts. Well, I suspect it’s because this circuit board has open pins and is not a closed system.
Either way, this circuit board must be replaced for legal reasons by a trained electronics repair technician. I assume that must support asks beforehand, “Are you familiar with this?” in order to be on the safe side legally.
However, I also find it absurd. Either such printers are intended for experienced users and all replacement parts are available, or the printers are toys and the user is not trusted, and therefore no replacement parts are provided.
TL;DR - I don’t think the reason can be explained logically, but rather it’s a combination of the legal situation and the fact that Bambu itself probably knows that the A1 series in particular is often seen as a toy? A mixture of everything?
2 Likes
I’ve skimmed this post trying to wrangle the information out but I can’t sleep and it’s 4am. So that means my dumb is in the red zone.
This is going from a reddit post-polish users comments that this is happened?
So these are all stock A1’s? Purchased at the same time frame or over a period? Running in the same environment? Stable power grid?
Seems like if these questions alone can’t be answered for each one there’s no way to begin trying to figure what’s happening out?
My bias is for the truth, not made up madness because of said youtuber’s own bias. How many times do you need to be lied to before you stop believing what the liar said.
I do apologize for my wording. It is phrased as if I don’t believe that there is an issue because of the source of the video. I really dislike youtubers, people like him.
Like I said above. Are they overheating? Evidence says yes. Why are they overheating? What is the same and what isn’t? Like I said. I need that info and tea with the Queen of Printers to even listen.
2 Likes
The ones I linked just now all come from Bambu subreddit with users from across the world. Previously I shared photos from Polish discord & facebook. This means the issue is spread worldwide. I don’t know about those on reddit, but the grid in Poland is pretty stable, those that talked about their issues bought from various places (bambu store and official resellers), all are “stock” and all have the same component failing in same manner - the one identified by “biased” youtuber. Would be cool for some “non-biased” ones to pick that up, but there’s no such thing.
I see.
So what about when these were manufactured? Anyone have any ideas other than the versions of the board? I’m more curious than anything. I’d love to help but I like to know more than “mine burned too”. Sure it did, but what’s in common with the others.
Sort of off topic-There’s a saying. “Sun shines up a dogs a** once in awhile”. Means sometimes even a idiot can be right. Doesn’t change the fact that he’s biased, so the quotes aren’t necessary. He is.
This guy claimed to have 3rd party evidence (insinuating he was involved to some extent) that X-P series where phoning home all the juicy bits. You know, super secret stuff like your logs, your files, your ip, etc. There was proof, there was evidence! That is until someone started questioning it. Suddenly what he was claiming didn’t make sense, and best yet he wouldn’t share the evidence. Didn’t stop him from posting about it on Redit or his channel. It only took a few hours to debunk this, but it had spread like everything does on the internet.
Nathan Builds Robots took up the “security breach” cause and spread it while shaming BL. Problem is there really was nothing. Nothing. It was all made up just to make BL look bad. BL even responded, but you know…they lie and all. NBR whined when BL wouldn’t send him a printer to “review” afterwards. Was pathetic.
To my knowledge 3D never apologized. Just tried to fade back into the woodwork. He has a history of boosting Prusa and forgiving their mistakes while being super critical of anyone that approaches Prusa as a competitor. You know that’s fine if you are the random guy on a internet board of forum. Not so much when you want to be taken seriously as a content creator.
So yes, I believe that there is a problem that needs addressing. Love to know how and why it happening and how it could be addressed. I’ll be damned though, if I take a word this tool says as truth on anything. I’d like to see posting of facts and with desire to help those affected and BL being held responsible if need be.
EDIT. I have some things wrong (details about what the claim was) so here is one of the original Redit threads about it.
https://www.reddit.com/r/3Dprinting/comments/18nzf1v/setting_the_record_straight_bambu_labs_response/
It’s important to be fair and make a clear distinction here. Nobody is denying the problem. It’s good that it’s being highlighted. However, what you’re leaving out is this:
This particular YouTuber is being thoroughly roasted on Reddit Bambu sub for spreading misinformation. Time and time again. This has been a recurring theme in his videos for years. Take a look at Reddit, where there are thousands of comments criticising his videos - and rightly so.
What you picked out from the Bambu Reddit sub was very selective. There are just as many heavily commented topics that clearly distance themselves from this YouTuber.
It’s not rocket science, and even here in this thread, it has been pointed out repeatedly that this YouTuber is only interested in clickbait. The problem in this case is not the message, but the fact that he is the messenger.
3 Likes
An NTC inrush limiter is only experiencing a high current for a short time, normally (which it also limits due to not having heated up yet). Looks like typical power is about 100W when printing. So roughly 1A of current in USA (half of that in Europe). The voltage over the NTC reduces as it heats up, so Voltage (U) X Current (I) = Power (W) wouldn’t be very high.
Has anyone actually measured the voltage across the NTC?
If anyone has the code for that NTC we can easily do the calculations.
Yes, it was on Reddit in the Bambu sub, but I don’t think I could find it again off the top of my head. The problem with Reddit is that funny memes about the A1 get now pushed to the top and real solutions to problems get completely buried.
According to his measurements, it is only devices operating at 230V that are at their limit, not those operating at 110V. His measurements and calculations appear to be correct. He analysed this extensively.
He also pointed out that it’s not only the new circuit board that will help. For older devices, updating the firmware to control the heating bed in stages could also be an effective solution. While this reduces initial power consumption, it does mean that it takes a little longer to warm up.
Maybe I’ll come across the topic over the next few days, but as I said, I don’t think it received so few likes. The sub was spammed with so many memes and panic-mongering.
OK, I’ve found the calculations at least. I’ll post them as an additional comment, hoping I don’t exceed the character limit. These aren’t my calculations. Thanks go to Jusanden from Reddit. Slightly shortened by me, but no sentences or content changed.
=== Part 1 ===
Couple other important specs to mention: Its rated for 6A max steady state current and it has a dissipation factor of 20mW/C (data sheet uses MW/C but I assume this was a typo). The former basically states how much current can go through during a “long” operating cycle. My definition here is usually a couple seconds but it’s very much a rule of thumb. The latter tells you how much hotter the device will get if it dissipates a given amount of power. Importantly, this is temp rise, so the temp of your surrounding matters, your heat sinking matters, and the NTC characteristics also completely ruin my day when it comes to the math here so I didn’t bother and just assumed best case scenario to see if this was plausible.
=== Part 2 ===
Okay enough setup get to the point. Well… some quick googling shows that the A1 uses a mains connected bed. That means it’s a big resistor and V=IR, or more importantly, I=V/R. At 220V, the resistor will dissipate 4x the amount of heat than at 110V. (Pdiss = V2/R)
The question is now how much power is that? Well, the spec sheet claims 1300W max current for 220V input with everything heating up at once. This is 5.91A through the resistor, just shy of the 6A limit. This is not accounting for ANY tolerances and is, in my professional opinion, cutting it pretty freaking close. I like my margins too much.
Crunching the numbers a bit further, it’s also about 3.9W dissipated on the resistor (P=I2R). At dissipation factor of 20mW/C this is a temp increase of just shy of 200C. This is assuming the minimum resistor value at hot. Starting from ambient, this results in around 220C on the component and you are starting at likely 50C or higher. This is frankly, absurdly hot, not to mention out of spec. I’m hesitant to call it a mistake outright, but lets just say its a bit sus.
=== Conclusion ===
Is there an issue? Well, all the reports point towards yes, even if you didn’t bother reading through my wall of text. Is the issue steady state current on the inrush current limiting resistor? If you‘re on 220V, the math definitely has not exonerated it. Again… hesitant to draw hardline conclusions given I don’t have the sucker in front of me, but it definitely seems plausible. The temp rise on the component is probably the most concerning part of this whole wall of text.
One unpopular, but immediate, action Bambu can take is to duty cycle the heater and prevent it from running on full blast throughout the entire heating up period. I suspect this is already happening to some degree, but doing so may help prevent that resistor from crossing the line between hot and flaming hot at the expense of longer heat up times.
Bambu’s solution on the new AC board rev of just ditching the stupid NTC may be valid. They also removed some of the caps, and as you’ll recall caps are what required the inrush current limiter in the first place. Worse power filtering, but hey, if they can pass EMI/EMC requirements without it, then there’s not really an issue here.
You’ll also perhaps notice I didn’t do any analysis at 110V inputs. It’s because I don’t think there’s a point and I’m way too lazy. At a quarter of the heat dissipated, it seems unlikely to still be causing issues and this is getting long enough as is.
=== Afterword ===
And IMO, this is just me griping a bit, all of this wall of text should have been done BEFORE a video was made on the topic.
1 Like
Just ask them what they thought about the melty qidi plus 4 SSR, and youll see the bias.
Anybody that reads this forum every day and has a decent memory of who says what, would find it hilarious
1 Like
He seems to have milked that “repair” vid for all he could. 4 hours?
Er, yes they do, they declare compliance with EN IEC 62368-1:2020 as shown in the EU Declaration of conformity posted by RetroSharky.
Section 6 of this safety standard covers “Electrically-caused fire” and section 6.3 covers “Safeguards against fire under normal operating conditions and abnormal operating conditions”. Unfortunately, this safety standard is not freely available.
1 Like
Are you insinuating that this board, the users here are biased?
The shame! The shame!
Or, now hear me out, could be the cited source outside of reddit is as much a troll as some “unbiased” users here are.
Thank you for pointing that out. I’m not familiar with that standard. Are you? Given the current situation, do we know enough to say whether the printer is or isn’t compliant?