Power Supply failure  

Re: Power Supply failure

Another one bites the dust. 😡

110V US based; it failed after a couple weeks of operation.

I had the ticking in sync with the bed heater LED.

I guess I’ll get with the chat folks tomorrow.

Re: Power Supply failure

It seems that all the failures are heat history related. Has anyone hung a small fan off the upper part of of the outside cooling slots to help exhaust warm air? You could wire 2 12V in series. They usually draw less than .1 amp and could significantly cool the board without blowing air across the plater. I have a few fans laying around and I might give it a try.

Re: Power Supply failure

I have hung a 40mm fan temporarily on the PSU to keep it cool, i have two silent 80mm 24v fans coming tomorrow (one for this mk3, one for the mk3 i get in 2ish weeks) and there are mounts available on thingiverse to hold them, only thing i will have to do is wire them to each printers PSU.

When printing PLA or PetG the mk3's PSU gets incredibly hot, and I think that is what causes a good amount of failures.

My mk3's PSU died about 120 hours in with no cooling, and since it was replaced under warranty, ever since the 40mm fan, I have printed well over 500 hours with no PSU failure.

Re: Power Supply failure

When printing PLA or PetG the mk3's PSU gets incredibly hot, and I think that is what causes a good amount of failures..

I think this is definitely a 120v thing. I know we have had one or two 240v failures, but the vast majority seem to be 120v.

My PSU even after an 18 hour print time (PLA) bed at 55c and nozzle at 210c, my psu is hardly even warm.
I haven’t printed anything other than PLA yet though. I’m on 240v mains here. I have over 37 days print time on the mk3 and so far it’s been no issue as far as the psu is concerned. I have had a filament sensor fail, but that was quickly replaced under warranty.

As prusa is using the same multi voltage psu for world wide use, it has to draw double the power on the AC side to supply the same current on the DC side. I believe this is what is causing the majority of the reported failures. Something in the psu just is not super happy at 120v power levels. Next time I do a print I will get the watt meter and test the ac side current on 240v. Maybe something for a 120v user to try also, so we can compare. I suspect that on 240v the current will be significantly less than on 120v. (In theory it should be halved) but we all know about theory.

I have the meanwell 350w in reserve as a backup just in case however.

I think the idea of a cooling fan idea is a good one, if not an elegant solution.
It might make the prusa psu last longer / indefinitely.

Hopefully prusa will weigh in and give everyone an idea of the root cause sooner rather than later.

Jason.

Re: Power Supply failure

... it has to draw double the power on the AC side to supply the same current on the DC side...

I'm pretty sure you meant double the current on the AC side for the same current on the DC side. Power consumption/usage should be the same (120/240V).

Re: Power Supply failure

I had an issue with the PSU when I woke up from sleep and the print had stopped. The printer wouldn't turn back on and still wouldn't even when I replaced the fuses. It turns out thrlere was a bad connection in the power supply and it was arching causing the PSU to go poof. I got another for free perhaps you could get the same.

Re: Power Supply failure

When printing PLA or PetG the mk3's PSU gets incredibly hot, and I think that is what causes a good amount of failures..

I think this is definitely a 120v thing. I know we have had one or two 240v failures, but the vast majority seem to be 120v.

My PSU even after an 18 hour print time (PLA) bed at 55c and nozzle at 210c, my psu is hardly even warm.
I haven’t printed anything other than PLA yet though. I’m on 240v mains here. I have over 37 days print time on the mk3 and so far it’s been no issue as far as the psu is concerned. I have had a filament sensor fail, but that was quickly replaced under warranty.

As prusa is using the same multi voltage psu for world wide use, it has to draw double the power on the AC side to supply the same current on the DC side. I believe this is what is causing the majority of the reported failures. Something in the psu just is not super happy at 120v power levels. Next time I do a print I will get the watt meter and test the ac side current on 240v. Maybe something for a 120v user to try also, so we can compare. I suspect that on 240v the current will be significantly less than on 120v. (In theory it should be halved) but we all know about theory.

I have the meanwell 350w in reserve as a backup just in case however.

I think the idea of a cooling fan idea is a good one, if not an elegant solution.
It might make the prusa psu last longer / indefinitely.

Hopefully prusa will weigh in and give everyone an idea of the root cause sooner rather than later.

Jason.

I have been reading this thread and I don't like what I am reading. I have ordered a meanwell 25 amp 24volt power supply figure if I more than double the current of the original power supply then it doesn't have to work at hard and shouldn't have any power issues. going to remote mount it away from the printer so if it does happen to get hot I can put a nice big fan on it 😀

Re: Power Supply failure

I have hung a 40mm fan temporarily on the PSU to keep it cool, i have two silent 80mm 24v fans coming tomorrow (one for this mk3, one for the mk3 i get in 2ish weeks) and there are mounts available on thingiverse to hold them, only thing i will have to do is wire them to each printers PSU.

...

Would you mind terribly posting a link to that PSU Mount for the 80mm fan? ..I purchased an 80mm fan for just such an occasion... and I haven't run across the PSU mount with a 80mm fan on Thingiverse

Thanks!

Re: Power Supply failure

Perhaps a fix that does not involve modifying or replacing the PSU could be to use an external voltage doubler? In other words, the circuit used in the CR-10 PSU to allow it to be used on 120V? (Apparently, the CR-10 also tends to have PSU issues because it's undersized as well.)

Re: Power Supply failure

Just to keep the numbers moving I have had one PSU power supply go poof also on the MK3. It was a prebuilt unit and I was 3 hours into a PET build. I heard it pop and saw sparks, tried replacing the fuse and boom instantly the same stuff. Prusa was on the ball and replaced no problem but the new one clicks even with PLA. I am afraid to print PET now so just ordered a NES-350-24. Seems people feel this one is better and on a printer like this the $20 seems worth it to me.

Re: Power Supply failure

... it has to draw double the power on the AC side to supply the same current on the DC side...

I'm pretty sure you meant double the current on the AC side for the same current on the DC side. Power consumption/usage should be the same (120/240V).

Yeah, sorry. I meant to say it has to draw double the current on the 120v AC side for the same output wattage on the DC side, as it would if it was on 240V.

Next time I have mine running, I will hook up the watt meter and get some numbers.
Hopefully we can compare someone on 120v to see what the difference is.

Jason.

Re: Power Supply failure

Maybe something for a 120v user to try also, so we can compare. I suspect that on 240v the current will be significantly less than on 120v. (In theory it should be halved) but we all know about theory.

From my own measurements and from what I vaguely remember other users posting, with the PSU maxed out around 240watts DC (during preheat), the AC side of the PSU measures around 285-290watts AC @ 120v. So in the neighborhood of 2.4amps AC @ 120v.

Re: Power Supply failure

I'm starting to assemble my printer (in the US) and am thinking about proactively using an HEP-320 power supply instead of the stock one.
I'm thinking to put the supply on the table (not attached to the printer) and skipping on the power panic for simplicity to start.
Is there any known downside to having the power supply module NOT attached to the frame in terms of frame stiffness?
I can imagine making some kind of stiffener from the upright to the horizontal rails.

Any suggestions?

Re: Power Supply failure

I'm starting to assemble my printer (in the US) and am thinking about proactively using an HEP-320 power supply instead of the stock one.
I'm thinking to put the supply on the table (not attached to the printer) and skipping on the power panic for simplicity to start.
Is there any known downside to having the power supply module NOT attached to the frame in terms of frame stiffness?
I can imagine making some kind of stiffener from the upright to the horizontal rails.

Any suggestions?

Already addressed:

https://www.thingiverse.com/thing:2772947

Re: Power Supply failure

Happened to me on Mar 27th during a print after 2 months of regular printing. It was running on 120V.

My request for a replacement was put in on the 28th and I'm still waiting for it to arrive.

For those of you that got a replacement, does Prusa send just the PSU or the PSU and printed parts/switch etc?

Re: Power Supply failure

I ordered the RPS-320-24 before reading this thread.

I am currently modelling the bracket for this and should have it ready to put onto Thingiverse for anyone who also decides to use the RPS-320-24 as well as a photo-log on how to make the swap.

Some notes about the RPS-320-24
- 320W vs 350W of the NES-350-24/LRS-350-24 (RSP-320-24 outputs 13A compared to the stock 10A power supply -- NES/LRS-350-24 output 14.5A)
- built in PFC Function
- no 110V/220V selector switch
- same footprint/case size as the LRS-350-24 -- however, has more air vents than the LRS-350-24, thus a new bracket is needed
- RPS-320-24/LRS-350-24 has a 3 year warranty compared to 2 year of NES-350-24
- RPS-320-24 is more expensive than LRS-350-24/NES-350-24 (prices from mouser US)
> RPS-320-24 = $53.59
> NES-350-24 = $48.13
> LRS-350-24 = $32.46

There are some other minor differences, but those are the main items.

More soon on the swap out

Re: Power Supply failure

I just started a 50+ hr print.
So I got the power meter and got some numbers.

On preheat for PLA it was drawing about 300W and 1.3A on the AC side at 240v
Once printing, it was about 120-150W and 0.6A.

I’ve only printed with PLA so far, so don’t know what to expect when I start playing with PETG.

Jason.

Re: Power Supply failure

I just started a 50+ hr print.
So I got the power meter and got some numbers.

On preheat for PLA it was drawing about 300W and 1.3A on the AC side at 240v
Once printing, it was about 120-150W and 0.6A.

I’ve only printed with PLA so far, so don’t know what to expect when I start playing with PETG.

Jason.

Really now... It's a 240W PSU, right?

Re: Power Supply failure

It's 240W on the output, power supplies aren't 100% efficient

Re: Power Supply failure

One difference between the Prusa PSU and a Meanwell is that the Meanwells are actively cooled. I proactively added cooling fans to my PSU. I used two 12 v 40 mm quiet fans wired in series, powered by the PSU and mounted on the outside near the top of the PSU to exhaust warm air. The fan holder simply plugs into slots in the PSU case. The fans are on the outside, so that the air flow should not disturb the print bed. The fans draw 0.035 amp, which is a negligible extra load on the PSU.

fans I used: https://www.amazon.com/gp/product/B000VRVHFO/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1