since a couple of days i have been having trouble with my heatbed, it doesn’t warm up propperly. It slows down the hotter it gets and eventually it crashes
Here is what i have done so far:
I checked al my wires and their connections, good.
I Replaced a part of my wires that had beed bent for a while, i assumed that might be the case (it was not). Futher more no weird looking or moving cables.
I don’t feel the cables getting hot at the connection to the heatbed.
And then it crashed at 64,9°C
Room temperature = 16,8°C
I have the heatbed hooked up to my motherboard (max 10A, peak 12A power). And the original powersupply
The original heating module for the Neptune 4 Max could deliver 25A. but i never had any issues with heating before, it is just a bit slower now.
No i have not because the values given are ‘normal’ at startup it matches with the 2 extruder temps and the values of the thermistor don’t act unusual, they slowly change value.
so that is why i did not check my thermistor
For me it looks like there is not enough power to heat up the bed quicker.
Maybe the cabling or even the PSU asks you for further inspection or care/maintenance.
its strange, i just heated my bed up to 75°C without error.
and if the power supply doesn’t show any sign of strange things i will see if i can put the old module back in with the use of the GPIO pins on the motherboard.
You can’t see from the outside when a switched PSU gets worn out.
Due to heat (in first place), capacitors can lose their capacitance and the PSU is not able any more to produce 100%.
A PSU should be able to deliver at least 10% more of the required power.
So, if the printer in all needs 260W, a 300W PSU is good to go.
If a PSU always runs on 100%, it will fail more quickly.
If a PSU always runs on 100%, it will fail more quickly.
I get that, but i am using less power then before. There used to go 25A (600W) to my heatbed (before I upgraded my motherboard), now it is only 10A(240W). this means my PSU should be fine, right??
could it be that my motherboard can’t handle it for longer periods of time?
I just read on my PSU 24v 17A max (408W) so not 25A, that is just the max of the original module.
The output of the power supply is determined by the drivers and not what you’re calling “internal resistance”.
You should assume that a power supply’s output impedance (what I think you’re calling “internal resistance”) is zero.
Sorry for jumping in here but I just wanted to get the full story. I’m been watching and things don’t quite make sense - I don’t think we have the full story.
When I looked up the Neptune 4 Max, the basic power supply is 408W (as @Jasper says).
Was the Manta M8P V2.0 running okay and then there was a change? Can you explain the change - how long did it take to heat up before (ie your expectation) and how long does it take to heat up now?
What is driving the heated bed? When I look at your klippy.log, you’re always specifying PF5 which is the Heater Bed transistor built into the Manta M8P V2.0 - this transistor (HYG017N04LS1C2) shouldn’t be a problem because it can drive up to 40A as per the datasheet: HUAYI HYG017N04LS1C2 Datasheet with a very low internal impedance (stated as 2.3mΩ). So you shouldn’t have a problem with it.
I’m confused as to where the MKS MOS25 comes in. Are you using it now or not and how have you wired it in?
If you’re using the MKS MOS25 - do you have any schematic or the N-Channel MOSFET part number? I can’t find any useful information online.
Why are you using watermark instead of pid control for the heater algorithm?
Not to hate on your math but you a sketching a different situation atm.
my bed isn’t power by an element, there is a module (post #1) and the maximum power that can run through that is 25A but the max my PSU can provide is 17A.
there are 2 wires connected to my bed to warm it up nothing more.
And then you also have the motherboard and motors so my bed will probably consume about 12A when using this module, 25A is impossible then my PSU would blow up.
but via my motherboard it can only draw 10A of current. it used to take about 10 minutes to warm up now it is closer to 15 minutes.
PROBLEM SOLVED:
i adjusted the “Verify_Heater” to different values and it was able to stay warm without any issues.
Yes it was running okay, the change was in the heating speed of the bed. it now takes about 15 minutes to heat up.
the manta M8P V2.0 has a heatbed driver built into it. you put the power in via a separate wire from the power supply and an output to the bed. and you can turn that on or of.
The MKS MOS25 was build in the original Neptune 4 Max, but since i swapped out the motherboard (and alot of other hardware stuff) i got rid of that too because the manta M8P had a built in switch for the heatbed an ouput for the MKS MOS25 like the original motherboard has.
But since the MKS MOS25 has a higher power rating than the Manta M8P V2.0 heatbed regualtor i was thinking of swapping it back in. (since i learned you can easily use the 40 GPIO pins provided on the board to toggle the power to the heatbed via the MKS MOS25.
nope, no schematics.
i don’t know why i am using watermark, it was always like that. Are there con’s to PID over watermark?
side note: the 10 minutes i was talking about was for my stock Neptune 4 Max. Not my modded version.
A possible reason for this ‘suddon’ trouble is because it is winter and maybe my room is just a little bit colder than usual. and the “Verify_Heater” configs were always on the edge and now it just tipped over it due to it being a bit colder.
I’m happy that you have things working but I’m trying to understand what you’re saying above and I’m a bit nervous that you’re making some erroneous assumptions.
Are these two wires running from the Manta M8P V2.0 or the MKS MOS25 module? This is not clear in what you’re writing.
Could you draw out your power supply wiring with the heated bed?
I have no idea what you’re trying to say here. In the first paragraph you are implying that the Manta M8P V2.0 is drawing 5A (assuming a 17A power supply and the bed drawing 12A) and in the second you are saying that it can only draw 10A of current.
Where do you get the idea that the “motherboard” (which I assume is the Manta M8P V2.0) can only draw 10A - you also state that in the original post. I just checked one of my Manta M8P V2.0 boards and they have three 15A fuses in them - one for the motor, one for the board electronics and one for the heated bed. So, at the very least you can provide up to 15A to the heated bed.
Just FYI: power supplies don’t “blow up”. They provide the maximum of what they’re designed to provide - good ones maintain the voltage level up to this point while the voltage output drops off with the more current drawn from them.
Did this just happen suddenly or was there a change (hardware, printer.cfg or Klipper version) that preceded it?
As I indicated above, the N-Channel MOSFET used in the Manta M8P V2.0 can handle up to 40A - why would you put in the work to replace it with one that handles a maximum of 25A?
Watermark is best suited for a mechanical system or control in which rapid turning on and off would wear it out quickly. At best, you get a temperature range (called the “comfort zone” because it was first used in home thermostats) of a few degrees.
PID provides you with more continuous control and temperature stability (on a heated bed) of 0.1C or so. You’re using it with your extruder.
This is actually the big question I’m trying to get the answer to - was your original printer heating up the bed in ten minutes and then, when you swapped out the main controller board for the Manta M8P V2.0, you saw that the heating time increased to 15 minutes?
Two final questions:
Could you measure the resistance of your heated bed? Disconnect the two heater wires and measure the resistance with an ohmmeter. It would be helpful if you gave the resistance of the test leads (just touch them together with nothing in between) as well.
Are you heating your extruder at the same time as the heated bed? If you are then you might want to change your Gcode so that the extruder heats up after the heated bed. This will provide maximum power to the heated bed so it will get up to temperature as quickly as possible and then it just takes a minute or so for the extruder.
After answering @mykepredko s questions, could you please upload a current klippy.log or at least you current printer.cfg?
Something is still odd here and I doubt that you have fixed the real reason that caused the issue, if it really worked way better before.
You just changed the important boundary checks for an unstable system, what can be dangerous.
The initial klippy.log shows a lot Transition to shutdown state: Heater extruder not heating at expected rate and later the same for the bed, what is not normal.
The later screenshots at least show that the extruder seems to heat up properly.
Just to be sure: the heat bed is still the original one or did you change/mod anything hardware related on that printer?
The heating curve of the bed looks fine except for the time it takes.
Though the heating bed with only 320 W is quite weak for that big dimensions (delivers only about 0,18 W/cm²).
The reason, as Myke stated already and I was stating in my first post here, is either not enough power being provided by the PSU (e.g. the extruder is heating up at the same time) or a higher resistance between PSU and heating bed.
Important side note: Elegoo specifies the heat bed for a maximum of 85°C only but you set max_temp = 130.
However Elegoo has this in its own config files:
At least the bed thermistor differs from the one you set in your initial klippy.log.
Here you can see that the bed is considered as very weak, as they expect it to increase its temperature by only 2 K within 300 seconds at least…
If you changed those values to be even more conservative then you almost don’t need a heated bed.
But here you can see that PID is being used for the bed too. This should eliminate the swinging you can see in your second screenshot once being set up properly.
This swinging is not good and can/will produce Z artefacts. It is no magic to get thet bed curve flat.
Did this just happen suddenly or was there a change (hardware, printer.cfg or Klipper version) that preceded it?
The stock version took about 10 minutes to warm up (with the MSK MOS25)
The current version takes about 15 minutes. (With the Manta M8P V2.0)
why would you put in the work to replace it with one that handles a maximum of 25A?
as said a couple lines before, the manta can only output 10A to the heatbed.
PID provides you with more continuous control and temperature stability (on a heated bed) of 0.1C or so. You’re using it with your extruder.
changed it to PID!
Could you measure the resistance of your heated bed? Disconnect the two heater wires and measure the resistance with an ohmmeter. It would be helpful if you gave the resistance of the test leads (just touch them together with nothing in between) as well.
I can’t atm i am running a print but i will try when it is done.
Are you heating your extruder at the same time as the heated bed? If you are then you might want to change your Gcode so that the extruder heats up after the heated bed. This will provide maximum power to the heated bed so it will get up to temperature as quickly as possible and then it just takes a minute or so for the extruder.
When starting a print, the nozzle heats up to 140°C i about 10seconds. then when my heatbed is done heating up it goes up to 235°C, in my case.
The initial klippy.log shows a lot Transition to shutdown state: Heater extruder not heating at expected rate and later the same for the bed, what is not normal.
loose cable, fixed that.
Just to be sure: the heat bed is still the original one or did you change/mod anything hardware related on that printer?
original heatbed.
or a higher resistance between PSU and heating bed.
i checked the resistance and bend the wires in multible ways to check for fractures, no weird measurements.
Elegoo specifies the heat bed for a maximum of 85°C only but you set
whoops, didn’t really notice that, changed it.
At least the bed thermistor differs from the one you set in your initial klippy.log .
