If the power supply was failing which would be reduced, voltage or amperage? 🤔
Can't measure >10A with my multimeter, but the 12.08V output of the power supply are absolutely stable when the temperature begins to drop uncontrollably. Meanwhile the MCU reports it tries to give full power to keep temperature.
In my very limited understanding of electronics this indicates that my power supply degraded and doesn't provide all of its 350W anymore?
#3DPrinting #electronics #repair
Questa voce è stata modificata (1 settimana fa)
Technology Channel reshared this.
vk3kri
in reply to Natasha Nox 🇺🇦🇵🇸 • • •If it was the power supply failing I'd expect to see the voltage sag. As you you indicate the extruder and plate both aren't keeping temperature , I'd check the wiring from the PS to whatever controls the temperature. Measure the 12V there to see if there is a voltage drop in the wiring .
0.01 Ohm at 10A gives a 1V drop.
Norris Dale AC1LP
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Bill
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Bill
in reply to Bill • • •Bill
in reply to Bill • • •Natasha Nox 🇺🇦🇵🇸
in reply to Bill • • •Natasha Nox 🇺🇦🇵🇸
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Bill
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Peter_Panther
in reply to Natasha Nox 🇺🇦🇵🇸 • • •@FloppySalmon
4censord
in reply to Natasha Nox 🇺🇦🇵🇸 • • •what you can do about measuring the amps: put a resistor in the thingy that should make it draw 10amps. then measure voltage at that.
if its degraded, it should not keep up the voltage
you probably need to put a bunch of resistors in parrallel, and even then they will only last a short amount of time unless you have some that are actaully rated to the 350w
so it'd be prep verything, plug in, measure, unplug, put out fire
Sebastian Lauwers
in reply to 4censord • • •4censord
in reply to Sebastian Lauwers • • •Natasha Nox 🇺🇦🇵🇸
in reply to 4censord • • •Chuck
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Fritz Adalis
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Sebastian Lauwers
in reply to Fritz Adalis • • •@FritzAdalis I’m also thinking this (or something similar to this) rather than the failing PSU. Initially I was thinking the heating element, but considering there’s something related to the motors running in your tests Nat, I don’t think that’s realistic.
Just to isolate: what happens if you let the printer warm up and then try to increase temps (leaving the motors disconnected/disengaged)?
Natasha Nox 🇺🇦🇵🇸
in reply to Sebastian Lauwers • • •@teotwaki @FritzAdalis Mmh… this actually makes a lot of sense. I'll check that tomorrow, now it's too late.
The original maximums for the printer were 260/70°C. Given good experiences with the smaller but very similar i3 Mega (running that one at 280/110°C max) I've raised the max for this bigger one to 270/90°C. This might've been too much for one or multiple of those FETs given the 50% bigger build plate, my mind was pre-occupied with power requirements and fire safety.
Natasha Nox 🇺🇦🇵🇸
in reply to Natasha Nox 🇺🇦🇵🇸 • • •@teotwaki @FritzAdalis Just to make sure, you guys were thinking about these, right?
Do I have to unsolder them for testing? They seem to be glued with adhesive thermal paste or pads to the PCB for heat dissipation, would probably be somewhat ugly to rip them off.
To state the obvious, I never tested FETs before. 😶🌫️
Fritz Adalis
in reply to Natasha Nox 🇺🇦🇵🇸 • • •@teotwaki
Oh, that's solder holding them on. You'll be sad if you rip them off the board.
Do you have an oscilloscope? If so measure the output and see if the voltage drops when the problem starts. A multimeter might not be that accurate since the output is pulsed. Worst case it looks like the board has leds that pulse when the heaters are on, you could put a light on the output and see if the pulses match.
Also see if the fet gets hotter than you'd expect. If you have a thermal camera that's ideal.
Is it the hot end or the bed that's the problem? Can you tell your firmware to use the other heater port if you're not already using it?
Natasha Nox 🇺🇦🇵🇸
in reply to Fritz Adalis • • •I neither have an oscilloscope nor thermal camera. :/ I could use the other hotend port for testing indeed, I only noticed when taking the pictures that there's a second one. There's not a second port for a bed heater though, so if it's one of the FETs I hope it's specifically the one for the hotend.
How could I even prevent this from happening again if it's that part? Putting a heatsink on it…?
Edit: Also no car to get it to the hackspace, nor money for equipment…
Fritz Adalis
in reply to Natasha Nox 🇺🇦🇵🇸 • • •If the part's gone bad from too much current your best bet may be to switch to 24v (assuming you're at 12v now). You could upgrade to a higher-current part but then you have to worry about if the rest of the design can take higher current. Like I blew out the fan fets on my board and upgraded them, but they're not 60A parts.
Natasha Nox 🇺🇦🇵🇸
in reply to Fritz Adalis • • •@FritzAdalis @teotwaki Oh god, no. The Trigorilla board we're looking at could take 24V, but I'd have to replace PSU, hotend, probably the whole bed, the bed control board (there's a daughterboard in this printer for the bed)… for that effort & price I could just grab a new Anycubic Kobra 2 printer. Which I can't, because of the price (even though it's just 149€). 🫤
If the part has gone bad I'd have to get a replacement and limit the bed to 70°C once more to not damage it again.
Fritz Adalis
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Yeah, that's reasonable. Sometimes I lose sight of the goal, my hobby is 3d printers, not 3d printing.
Natasha Nox 🇺🇦🇵🇸
in reply to Fritz Adalis • • •Fritz Adalis
in reply to Natasha Nox 🇺🇦🇵🇸 • • •That's okay, I'm still using an over-upgraded Ender 3 pro.
Sebastian Lauwers
in reply to Natasha Nox 🇺🇦🇵🇸 • • •@FritzAdalis Could you swap the board with something else? In other words, do you have another board lying around that could drive the motors and heated bed? This would let you know if the issue is on the board or the loads side.
Maybe reach out to local friends if anyone has an oscilloscope. You can probably borrow one from a makerspace if you leave your kidney or something. Schools/unis tend to have stocks of them lying around gathering dust.
gafu
in reply to Natasha Nox 🇺🇦🇵🇸 • • •@teotwaki @FritzAdalis
Dead MosFETs typically melt inside and then no longer be able to switch off the current.
If you can heat up with usual speed and then it does not run away hotter and hotter the mosfet will be fine.
Do you have stable hotend temperatures when *not* moving the printer head after heat up?
The typical 6mm heaters in the hotend do have a crimp connection close to the heater in the glassfiber tubes, maybe a wire broke feom constant movement, or one of the crimps is going bad.
Natasha Nox 🇺🇦🇵🇸
in reply to gafu • • •@gafu @teotwaki @FritzAdalis Okay, sooo… I connected the hotend to the other output (which was never used before). It behaves exactly the same. However… I noticed it's perfectly capable to keep the temperature the moment I stop the damn parts fan.
before you ask, I of course do have the silicon sock on the hotend.
I could lower the cooling ring a few millimeters, but aside from that… it's not like I used an over-the-top cooling solution, just a standard 5015 like I have on the other (green).
Fritz Adalis
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Can you swap the fans and/or ducts, or is that a hassle/impossible?
Natasha Nox 🇺🇦🇵🇸
in reply to Fritz Adalis • • •gafu
in reply to Natasha Nox 🇺🇦🇵🇸 • • •@FritzAdalis @teotwaki
I would glue on a thin metal piece to each side to deflect the air just below the heater block.
I am lazy^^
Natasha Nox 🇺🇦🇵🇸
in reply to gafu • • •Natasha Nox 🇺🇦🇵🇸
in reply to Natasha Nox 🇺🇦🇵🇸 • • •@gafu @FritzAdalis @teotwaki Sooo, now that I lowered the part cooler by 3.7mm (which helped a tiny bit) and, uh, raised the PSU voltage to 13V (helped a ton) the hotend is *just* able to hold the temperature with the part fan going at 100%.
The printer blessed me with a whole new problem. ❤ That might just be wet PETG though, this spool by now was in upper corner of the ceiling (not too far from the bathroom door) for weeks. Probably just throwing bubbles. But so far so good. 😀
Natasha Nox 🇺🇦🇵🇸
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Natasha Nox 🇺🇦🇵🇸
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Fritz Adalis
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Nice!
Natasha Nox 🇺🇦🇵🇸
in reply to Fritz Adalis • • •Momo
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Your multimeter can only measure the current that is momentarly drawn from your power supply. But unless you short out the output, this is no indicator for how many Ampere your power supply can deliver max.
However, if more current is drawn than your power supply can handle, the voltage begins to drop. And that is your indicator that your power supply is forced to punch above its weight. In your case the voltage level is stable, so your power supply is fine.
(The math behind it is P=U*I where P=350W max. So at U=12V your current I can be 29,16A max. If you start to draw 30A for example, the voltage U drops to 11,66V to keep P at 350W.)
Momo
in reply to Momo • • •Natasha Nox 🇺🇦🇵🇸
in reply to Momo • • •@momo Printy McDerpface. It had googly eyes on its print head before I upgraded that one. 😁
Got it, thanks! If my PSU is fine it really seems to be a board issue.
Momo
in reply to Natasha Nox 🇺🇦🇵🇸 • • •gafu
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Stable voltage indicates power supply delivers drawn currekt without problems.
Highest current demand is at heatup also.
Mawoka
in reply to Natasha Nox 🇺🇦🇵🇸 • • •Natasha Nox 🇺🇦🇵🇸
in reply to Mawoka • • •@Mawoka I'm pretty positive I've figured it out. 🙂 Guess it's a design flaw in the head which I simply didn't catch until now.
chaos.social/@Natanox/11514105…
Might still check the heat element's voltage, thank you!
Natasha Nox 🇺🇦🇵🇸
2025-09-03 15:25:17
Peter_Panther
in reply to Natasha Nox 🇺🇦🇵🇸 • • •