Notifications
Clear all

Power Supply failure  

Page 27 / 35
  RSS
J
 J
(@j)
Eminent Member
Re: Power Supply failure

Add another dead supply to the pile, from another 120VAC user.

We got two printers a month or two ago.
One just died mid-print. Now almost immediately after the heatbed is turned on, the LCD goes mostly dim and flickers, and it won't come back at all until the AC power switch is flipped off and on.
Resistor R4 looks ok yet, and there was no popping or smoke from the supply.

Checking the voltage readout under the "Support" part of the menu.... 21.9V.
The working printer is reporting 24.0V.

Stats say 720m of filament used, 21d:11h print time.

I've got an email in with Prusa through their main site, hopefully I hear back from that.

Some numbers, based on what I've read in this thread so far:

If this is a 240W power supply, and per the specsheet's nominal 85% efficiency rating, that's 282W into the input. People here are measuring around 300W current draw at the AC side. Efficiency curves do vary on a power supply according to consumption, but this supply's datasheet doesn't include that.
I also saw elsewhere, in a post from Prusa about building an enclosure for a 3d printer, that their supply needs to be relocated outside of it so that the heat in the enclosure doesn't kill the supply, so there's not a whole lot of headroom. All of this sounds to me like they're running the supply too close to the limits of its safe operating area. And some, let's say, "budget" power supplies don't work at their rated outputs. I once had a PC power supply blow out a power resistor at about 70% of full load.

I wish I knew what the specific fault is inside the power supply. Darn cost-cutting....

Publié : 21/08/2018 7:23 pm
ejberg
(@ejberg)
Estimable Member
Re: Power Supply failure


If this is a 240W power supply, and per the specsheet's nominal 85% efficiency rating, that's 282W into the input. People here are measuring around 300W current draw at the AC side.

Maybe the PSU is less efficient at lower input voltage. I've measured mine, and it always draws about 245W on the input side, when I start a preheat from cold, then sightly dropping as the hotend and heatbed gets warmer. This is with 228V 50Hz input.
I haven't heard any clicking sounds from the PSU (yet), but I am prepared for the worst. I have a Siemens 300W supply waiting on the shelf if/when the original goes poof.

Publié : 22/08/2018 11:01 am
J
 J
(@j)
Eminent Member
Re: Power Supply failure


Maybe the PSU is less efficient at lower input voltage. I've measured mine, and it always draws about 245W on the input side, when I start a preheat from cold, then sightly dropping as the hotend and heatbed gets warmer. This is with 228V 50Hz input.
I haven't heard any clicking sounds from the PSU (yet), but I am prepared for the worst. I have a Siemens 300W supply waiting on the shelf if/when the original goes poof.
True; without an efficiency curve in the datasheet or a live reading of the current draw on the DC side, it's hard to say exactly how hard the supply is being pushed. And the efficiency curve depends on the power supply. One datasheet I'm looking at for a different supply shows greater efficiency at 115VAC than 230VAC, while a different brand shows greater efficiency at higher voltages. I guess it depends on how the circuit has been optimized, and what topology they're using. Both of those supplies are more efficient at higher loads.

All three I3 MK3 printer power supplies I've been around have clicked when the bed is holding temperature, but once the motors also kick in during a print, the clicking goes away either entirely or completely. When the heaters and motors are inactive and the Einsy board is sipping power (very little load I'd presume, at least from the standpoint of a >200W power supply), and then the heaters start doing their PID thing while trying to maintain a temperature setpoint, the power supply is seeing rapid swings from no-load to high-load. At least with some steppers running, you are now between modest-load and high-load, which the power supply likes more, so no clicking. Best guess, anyway. (Works the same way with driving, right? Going from 0 to 60 to 0 to 60.... less fun than 45 to 60 to 45 to 60.)

Publié : 22/08/2018 3:40 pm
dimprov
(@dimprov)
Noble Member
Re: Power Supply failure

Here's Tom Sanlanderer's video on how to fix a power supply:

Publié : 22/08/2018 9:30 pm
Nikolai
(@nikolai)
Noble Member
Re: Power Supply failure


Here's Tom Sanlanderer's video on how to fix a power supply:

I don't like his approach: "Look which parts are looking damaged, replace those and hope it works again".
Without deeper understanding of electric circuits nobody should replace anything in the circuit. Especially if it's handling main voltage.

BTW: He is doing a huge mistake how he discharge the capacitors. If they are full and he is doing a short circuit, the capacitor can blow and/or he can weld his screw driver. You should always use a resistor for this.

I like Tom's videos but sometimes he is doing stuff very unprofessional.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram

Publié : 22/08/2018 11:05 pm
dimprov
(@dimprov)
Noble Member
Re: Power Supply failure



Here's Tom Sanlanderer's video on how to fix a power supply:

I don't like his approach: "Look which parts are looking damaged, replace those and hope it works again".
Without deeper understanding of electric circuits nobody should replace anything in the circuit. Especially if it's handling main voltage.

BTW: He is doing a huge mistake how he discharge the capacitors. If they are full and he is doing a short circuit, the capacitor can blow and/or he can weld his screw driver. You should always use a resistor for this.

I like Tom's videos but sometimes he is doing stuff very unprofessional.

What kind of resistor?

I've used his same approach in fixing some failed electronics before: find the failed part and replace it. It has worked well for me, and without any deep understanding of the circuit the part was embedded in. I mean, aside from the electrocution risk, what's wrong with it?

Publié : 22/08/2018 11:09 pm
Nikolai
(@nikolai)
Noble Member
Re: Power Supply failure


What kind of resistor?

https://www.wikihow.com/Discharge-a-Capacitor


... I mean, aside from the electrocution risk, what's wrong with it?

I hope it's ironic meaning. That's exactly what's wrong with it. Risk of getting killed or take a serious damage should be enough explanation. Once you blew up some electronic parts, you will be impressed how much damage it can cause.

I'm not saying his method doesn't work. Being unprofessional doesn't mean to have no success. He has also 50% success rate in his video.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram

Publié : 22/08/2018 11:53 pm
nathan0876
(@nathan0876)
Estimable Member
Re: Power Supply failure



What kind of resistor?

https://www.wikihow.com/Discharge-a-Capacitor


... I mean, aside from the electrocution risk, what's wrong with it?

I hope it's ironic meaning. That's exactly what's wrong with it. Risk of getting killed or take a serious damage should be enough explanation. Once you blew up some electronic parts, you will be impressed how much damage it can cause.

I'm not saying his method doesn't work. Being unprofessional doesn't mean to have no success. He has also 50% success rate in his video.

I agree, its a bad idea to short caps, sure the small ones will be fine but try that on one of the main caps in a ATX PC PSU ant let me know how that works out for you, or better yet try that on a car audio cap thats the size of a few pop cans, would likely cause an explosion. This is exactly the issue with these types of cut corners videos, people see it working on small scale then can seriously injure themselves when trying it on larger components.

Always best to use a resistor.

Publié : 23/08/2018 12:07 am
dimprov
(@dimprov)
Noble Member
Re: Power Supply failure

Safer, eh? I think I'd rather weld my screwdriver than burn my fingers:

Wouldn't you?

Publié : 23/08/2018 12:44 am
Nikolai
(@nikolai)
Noble Member
Re: Power Supply failure


Safer, eh? I think I'd rather weld my screwdriver than burn my fingers:
Wouldn't you?

I'm not really sure why you're trying to fight. I would rather recommend you to spend more time on reading. Under this very colorful image you will find:
5.
Touch a device to the capacitor leads for several seconds. Doing this will provide a path for the electricity to go and discharge the capacitor. You can use a 5- to 10-watt resistor, voltmeter, test light or ordinary light bulb. Using a meter or light will show the progress of the discharging, either with a digital display or a progressively dimming bulb.

I wouldn't burn my fingers because I know what I'm doing in this case. Pick the correct resistor and you will feel in a worst case slightly warm resistor. And just in case you really want to make it better instead of trolling, here is a howto

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram

Publié : 23/08/2018 2:18 am
dimprov
(@dimprov)
Noble Member
Re: Power Supply failure

Thanks for the warnings. They spurred me to watch some youtube videos, and I have more respect for the risks now.

This guy is long-winded, but he gives a pretty thorough explanation on how to pick a proper resistor value and how to make and use a proper discharge tool:

Publié : 24/08/2018 2:19 am
Protoncek
(@protoncek)
Reputable Member
Re: Power Supply failure

Ok, there's no need to do a "Ph.D. disertaition" on this one. value is far from being so important - i usually take 22ohm 5W for smaller voltages (say up to 40-60V) and 330 ohm 5W for 230V (only because i have plenty of those). Don't worry, it won't get too hot.
BTW.. 2kohm on iFixit video is way to big. It will take ages to drain a capacitor. Capacitors are quite capable of dealing with pretty high currents (of course short circuit is not among them)

Why there's risk of explosion if you do a short? Well, let's say that resistance of screwdriver is ... say 10 mohms, in which case start current would be around 32000A 😯 . I guess that pretty much all welding machines use waaay smaller current...

Publié : 24/08/2018 7:12 am
Nikolai
(@nikolai)
Noble Member
Re: Power Supply failure


BTW.. 2kohm on iFixit video is way to big. It will take ages to drain a capacitor. Capacitors are quite capable of dealing with pretty high currents (of course short circuit is not among them)

Ages is something different. 🙂 If you have a pretty big one with 2200uF, it will take around 8 seconds:

Here is an online tool if you want to play with it:

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram

Publié : 24/08/2018 6:42 pm
Protoncek
(@protoncek)
Reputable Member
Re: Power Supply failure

Like i said: there's big difference between theory and practice. I drained big capacitors with high voltages with 330ohm. Never got burned in my fingers, never burned resistor nor had explosions...
And to be honest: 2200 uF on 220 volts is pretty rare.

Publié : 24/08/2018 8:42 pm
J
 J
(@j)
Eminent Member
Re: Power Supply failure

For anyone doing the replacement using Meanwell's LRS-350-24, and wants to dive in to replace it with something quieter, these are the fan specs of the one I got: (Only do this if you have some idea of what you're doing. Those bulk capacitors can pack a punch, which it looks like is being discussed above.

ARX DC Brushless fan.
FD1260-S3112C.
2-wire.
60x60x15mm.
12V, 0.13A.

Part# decoder: http://www.arx-group.com/ARX_%20Fan%20catalog.pdf

Sleeve bearing :x, and the 3 indicates "Low" fan speed.
So that looks like it's 15.27 CFM, 34dBA, 0.16"-H2O pressure.

Good news here, the supply is spec'd to run with the stock fan at 100% load at up to 50°C/122°F ambient. So, personally, I'm not too terribly worried about matching the fan spec exactly. I'm not planning to leave this thing running out in the summer sun, and the supply is not going to be putting out full power all the time. (If it was going to be run harder, I'd be more particular. These supplies are built to a price, and from experience, some of the components are pushed to the edge of their safe operating area.)

I'm going to give this one a try:
NMB's 06015VA-12L-AT-00
Slightly lower CFM, same static pressure, a few decibels quieter, power draw is comparable to the stock fan, and ball bearings. It does have an RPM/tach output wire. That's just going to get snipped short and capped with heat shrink if I think it needs it.

Publié : 29/08/2018 10:15 pm
mkmkmk
(@mkmkmk)
New Member
Re: Power Supply failure

My 3 day old printer just died and won't turn back on. What kind of quality control does Prusa have? Apparently I've made a $1000 mistake...

Publié : 02/09/2018 5:50 am
Phil
 Phil
(@phil-4)
Trusted Member
Re: Power Supply failure

Contact support. I’m sure they will help. Prusa is generally known for good support. Use the live chat on the shop website for quickest response.

Publié : 02/09/2018 6:53 am
Protoncek
(@protoncek)
Reputable Member
Re: Power Supply failure


My 3 day old printer just died and won't turn back on. What kind of quality control does Prusa have? Apparently I've made a $1000 mistake...

There's no need for this kind of thinking. I know a guy who bought 100k€ Mercedez and died on a way home. So, did he do a 100k€ mistake?

All things CAN die. You can test certain thing 100 times but it will die 101st time you plug it in. Or you can test it 10000 times but it will die 10001th time.

That's why there's a warranty. Call them, it's probably PSU unit.

Publié : 02/09/2018 9:45 am
mkmkmk
(@mkmkmk)
New Member
Re: Power Supply failure



My 3 day old printer just died and won't turn back on. What kind of quality control does Prusa have? Apparently I've made a $1000 mistake...

There's no need for this kind of thinking. I know a guy who bought 100k€ Mercedez and died on a way home. So, did he do a 100k€ mistake?

All things CAN die. You can test certain thing 100 times but it will die 101st time you plug it in. Or you can test it 10000 times but it will die 10001th time.

That's why there's a warranty. Call them, it's probably PSU unit.

This has been a known issue for months. This thread and others on the web have countless examples of people with this issue. Many resort to simply spending more money for a substitute. That's flawed thinking IMO.

Prusa has clearly done amazing things for the industry, but as a company, they should have stepped up months ago and found an alternative supply that does not have this kind of failure rate. The fact the _company_ Prusa (not the individual) has not taken corrective action speaks volumes of what they think of their customers.

Publié : 03/09/2018 4:00 am
Nikolai
(@nikolai)
Noble Member
Re: Power Supply failure


This has been a known issue for months. This thread and others on the web have countless examples of people with this issue. Many resort to simply spending more money for a substitute. That's flawed thinking IMO.

If you have followed this issue here and in other places, you might know that it's still unclear why this power supply is failing for some people. It's not about just spending more money. Power supply is under warranty, so it costs money for PR to replace those. Obviously the fail rate is not that high and designing/using better power supply will cost a lot more. But I can fully understand why people are frustrated if that happens. Everybody wants to print 24/7 and any downtime just hurts.

Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram

Publié : 03/09/2018 5:37 am
Page 27 / 35
Partager :