Snapmaker 2.0 A350 Power Supply Requirements

I have some issues with my Snapmaker 2.0 A350 lately with the unit stopping during 3D printing. I have had many long print sessions without incident, the longest being 150 hours for a scale model an engine block so these are new occurrences. The touchscreen and enclosure lights would just shut down but the power supply was still ‘On’. I would then switch the power supply off and then back on and the unit would usually go into power outage recovery mode. At first I thought it was a bad connector because I moved the power supply and when moving the connector on the power supply back and forth I would get a flicker in the enclosure lights. After removing and reseating the connectors I would get 100 hours of printing without interruption and then this morning it stopped on a new print without power outage recovery. I restarted the print since it was only about an hour into the job and it powered off again almost immediately. I moved the power supply so that I could put easily see the power monitor I had placed before the power supply.

I had placed the monitor in the power supply previously and saw that the unit usually operated around 90 watts once printing but now I notice that it will stay around 90 watts and then jump up to 310 watts for about 4 seconds or so every 30 seconds (those times are not accurate just ballparking; have not watched with a timer to get accurate counts; didn’t think that was the important bit here). I am not sure if it was doing that before but I can see it being necessary for increased power for the heated bed. My concern is that from the Quick Start Guide page 7 it lists the rated power as 320 watts.

So I have several questions that I am hoping to get some feedback on:
a) What are the average power requirements you are seeing on your machine during 3D printing? I want to make sure I am seeing somewhat typical draws before I fully condemn the power supply. For all I know my heated bed could be drawing more current than needed and causing the issue.

b) I know that there have been several other mentions of the power supply being underpowered, how did anyone resolve this? Is there information on the plug so that I could purchase my own heftier power supply? I know that I could probably figure it out but I would rather go down a path someone else has blazed.

c) Is SnapMaker going to provide a beefier power supply for purchase? I would rather buy an official approved one rather than cobble together one on my own. I have several here that could do the job but I like the fact that my SM2 is not a Frankenstein yet unlike my other 3D printers with all the upgrades and repairs.

On the next failure I will take a look at the logs, just found out how to get to them. I suspect there will be something since the extruder head moves to the side before the unit goes off; so I suspect that it is a controlled shutdown.

There is a new firmware version September 1st that I have not applied yet that will resolved an issue I saw with the print stopping at 99% which I am sure is a different issue. I am hoping that might give some resolution to this also.

Thanks for any feedback and help you can provide.


Interesting. I don’t have anything to add, but I’d like to see the logs if you find a way to share them.

On my other printers the bed has a dedicated power supply. It shouldn’t be necessary here, but it also doesn’t seem that hard. Hmm…

My understanding is the new firmware is worse, they attempted to fix the 99 problem but others have reported it still happening along with a slew of other new issues. Just so your warned.

You can use a larger Mean Well PSU say 400 watts or more at 24 volts but it will not fit inside the standard enclosure. If you make or print your own enclosure then it is just a matter of loosening all the PSU screws and swapping the custom power circuit board, with all of it’s LEDs, capacitors and proprietary power connector intact, and attaching it to the larger PSU. Would be nice to add an Emergency cutoff switch as well.

Here are some pictures from the inside and the power usage of my A350.

Power Usage, Printing-Module:
Startup 50W
Idle 46,3W
Idle with Enclosure-Light on 53W
Idle with Enclosure-Fan on 52W
Idle with Enclosure-Fan+Light on 58W
After homing, in Control menu of the touchscreen 50,3W
Heated Bed on 261W
Extruder on 81W
Heated Bed + Extruder on, startup peak about 290W, then about 272W
Heated Bed + Extruder on + Homing 280W
All on, it doesn´t go higher than 290W

Power Lost Capacitors:

Power Lost board + LED Connector for Breathing Light

Connector Board from the top

Connector Board from the bottom

Powersupply, Meanwell RSP-320-24

Snapmaker Powersupply, Lable

1 Like

Would something like this be acceptable as an emergency cutoff?

That would work but not really an instant stop due to the large capacitor bank on the 24 volt DC side. If you kill the mains with a switch like that, it will still be several seconds before the machine stops. I’m not sure what cutting the DC power will do to the Touch screen running the Android OS.

Thank you for the great information. I have not had another failure since I posted and I have been on the same print with the same 90 watts to 310 watts cycling.

It seems that I am running about 10 to 20 watts more than you which is probably within variances between machines. That still seems like we are running really close to the limit of the power supply. It looks like the enclosure fan and light add another 10 watt that are already accounted for in our measurements but would give me pause if any other other add-ons introduced.

The pictures you gave on the inside of the power supply are great. I have not opened mine up yet, but that is next after the current job is done printing. It looks like it will not be hard to swap out the PS for a slightly more powerful one. Researching that now…

I like how the connections are through a circuit board that screws into the ps. That will make it easier to upgrade this. I am not adverse to making or adapting the enclosure to fit a slightly larger power supply both physically and for cooling.

I have 21 hours left in my current print and will be taking the power supply apart for my own research and getting the logs then before I start my next print.

I also want to take more accurate power measurements of all the functionality (Laser and CNC included) for reference purposes. I have and plan to use the hell out of this machine :wink:

So here is a ps that looks like the one that from the picture:

And here is one that looks the same physical size but with about 30 more watts of power:

If they are the same physical size:
Pros: Easy Swap Out Cons: Not a lot of power added.

I would rather get more power to have even more headway for when more heat is needed for the ABS on the heated bed and high temps on the extruder.


@Edwin said, the controller is the limiting factor of the machine.
You could find this information in the excessive noise of the power supply thread, I guess. Or it was the thread about heating bed more then 80degree C, not sure anymore…

1 Like

What I’ve done in the past on other machines is use a relay to control the heated bed via a different power supply.
Main power supply powers everything except the heated bed, and the heated bed ON command is redirected to a relay coil.
The relay closes a contact that turns on the heated bed from a different power supply.

Please enjoy MSPAINT-CAD

In my previous case, the bed heater is a 120V mains powered silicone heater. I’m considering ordering something like this for the SM2:
But the inability to run a M303 PID autotune makes it more difficult to tune - you’d have to measure the Ku and Tu yourself.


On teh Board is a NCE P0140G N-Channel MOSFET which controlls the Heated bed. Not sure if a “simple” Relay is enough. THe MOSFET itself is rated for 40A but i somewhat doubt the board will take that without problems :smiley: Also i didnt see a Spec for the Connectors, so pushing that much curretn through them might get toasty (if you keep the Heatedbed as is and not just supply it differently).

That would be 960W, and would be incredible if the power supply could do that hahaha

I just measured the bed resistance, it’s ~2.5Ω, so ~230W, at ~10A at 20C.
Compensating those values up to 60C, it should be ~2.9Ω, so ~200W, ~8.5A

Also the temp sensor appears to be a nominal NTC 100K.
It reads 124kΩ at 20C. Pretty accurate since the datasheet says it should be 124.7KΩ.

I have bought relay boards like these before, with good success:

Edit: Bought the bed heater and relay lol. Will let you know how it goes :stuck_out_tongue:


So my last print job just finished without incident, not even the 99% completion bug. I decided to open the power supply and I confirmed that the enclosed unit was identical to
So I decided to order this one that has 30 more watts of rated power:
It will be here Monday and I will report if there are any issues but I don’t expect any as it is a an exact physical replacement from what I can see. It is also only $28 shipped. I know it is only 30 more watts but that may be difference between a failure once in a while and a failure never which will be worth it.

One thing I don’t like about the power supply design is that the fan for the enclosed unit is covered and seems to try and flow the air from the front of the unit into the PS and then out the back. I wonder how effective it is and also the fan is loud so I wonder if the way that the unit is designed is amplifying the noise. I would think about a new case that holds a larger silent computer fan to blow air into the unit to reduce noise and improve the effectiveness of the cooling.

I also exported the logs and could not see any indication of an issue. I found the section of the log where the incident occurred and did any see any magic message:

1601556847486,2020.10.01 12:54:07.486,INFO,SC,line 49000: G1 X212.843 Y182.463 E615.14125
1601556882514,2020.10.01 12:54:42.514,INFO,SC,line 50000: G1 X111.717 Y161.843 E619.78703
1601556912568,2020.10.01 12:55:12.568,INFO,SC,line 51000: G1 X148.917 Y233.894 E624.46989
1601556962780,2020.10.01 12:56:02.780,INFO,SC,line 52000: G0 F3600 X129.358 Y221.932
1601557060396,2020.10.01 12:57:40.396,INFO,SC,line 53000: G1 X172.221 Y264.845 E659.17315
1601557148395,2020.10.01 12:59:08.395,INFO,SC,line 54000: G1 X189.357 Y231.151 E706.78116
36934,1970.01.01 00:00:36.934,INFO,SC,Application started!
37022,1970.01.01 00:00:37.022,DEBUG,SC,Reset overrides.
37558,1970.01.01 00:00:37.558,INFO,SC,Connecting serial port...
37598,1970.01.01 00:00:37.598,INFO,SC,onConnection, connected = true
37600,1970.01.01 00:00:37.600,INFO,SC,Enable heartbeat.
37603,1970.01.01 00:00:37.603,DEBUG,SC,HTTP server started.
37625,1970.01.01 00:00:37.625,DEBUG,SC,Machine model 3, size x 345 y 357 z 334
38620,1970.01.01 00:00:38.620,DEBUG,SC,Head type 1 found.
38801,1970.01.01 00:00:38.801,DEBUG,SC,Route: Create HomeFragment
38804,1970.01.01 00:00:38.804,DEBUG,SC,Route: Create LauncherFragment
1601557346038,2020.10.01 13:02:26.038,DEBUG,SC,Wi-Fi connected.
1601557346040,2020.10.01 13:02:26.040,DEBUG,SC,Start checking new version…

I can include the entire log if requested but it showed nothing interesting in that section. I am looking at the rest of the logs for how it behaves in a power outage recovery mode versus what happened here…



Mine was doing a similar thing. Edwin saw the youtube vid and is convinced that it is the power supply.
Good to know that we can purchase an aftermarket ps if needed. My question is the plug has the main 24v out plus 2 other little connectors. What do they do? and does the mean well have them also. What about the actual plug style? Is that proprietary?



I saw exactly the same values at my device.

I am having the same issue and was wondering if this new PSU solved the issue for you?

It seemed to but then I was still getting the issue when I powered off the unit for a while and then started using it again and then seemingly nothing for a while. It lately got to the point where I would get the power interruptions consistently so I opened up the PSU and started to test without the PSU enclosure and I am not having any issues right now; so my current theory is not enough air movement to keep some of the components below a temperature threshold. The unit is just sitting by itself without the metal shell behind the printer and I have not had an issue yet. I have not gotten around to figuring out what to do to enclose it for safety and still provide proper ventilation.