Power Supply Unit (PSU) overload?

I have a PSU which I have been using with my printer for about a month: [www.staples.com]

So far, I had three instances of this PSU going out.

1. The firs time, I don't know what triggered it. I was not printing anything. I think I was moving the X carriage up and down - to zero in on Z axis, and all of sudden I realized nothing was moving. I noticed that the PSU fan wasn't moving. I tried connections here and there, didn't find any problems, but it did not turn back on when I switched on and off several time.

2. The second time I may have shortened it or something. I was trying to connect a small cooling fan, which had three wires, black, red, and yellow, and I accidentally connected the fan's red and black wires to the red and yellow wires of PSU. When I made that connection, again the PSU fan stopped, rather graciously (instead of blowing up), and no more power.

3. The last instance took place last night. I was printing some object. Just before that I printed a couple of small objects without any issues as well. It was about half way through the printing.. I think it was printing for at least 30 minutes or more - even though it was a small object I was printing at 0.2 mm layer height, so it was going pretty slow - everything just stopped. When I looked at it (I was right there), I could see the fan in PSU was spinning down to complete stop.


Each of those three instances, I could not turn the PSU back up right away, but after 5-10 minutes I was able to turn it back on, and everything seemed to be working fine (until the next power off).

Can this be a sign of overload? I don't know what the total wattage consumption of my printer is, but I do have heated plate and also a cooling fan (although small). I believe these instances took place after I installed the fan, so I not only had an extra load with the fan, but also the hot end heater must've put in some extra work to maintain the temperature while to fan keeps blowing in air to cool it down.

If it is a case of overload, then I guess I can simply get a bigger PSU, or add another PSU (and use it for heated plate exclusively).

Or, can this be simplay a sign of faulty unit? I actually bought this PUS from Amazon (although the link above is for Staples) about a couple of months ago, so I think I could still get a replacement.

Sounds a bit like it.

That PSU has dual 12V rails internally (according to the page you linked). Each rail will have a max current rating, and if all your electronics are on the same rail, then it this might be the issue.

You should try using one rail specifically to run your hot end/motors/fan, and the other to run the heated bed. I don't have the same PSU here, so I can't tell you which one is which.

Note: The fact that it works again after being off for a short while usually means it's tripped an over-current limit. It switches off to avoid dumping a lot of current into a possible short circuit, and the only way to get it back is to disconnect the power and leave it till certain internal currents have discharged (time will vary depending on the PSU, and the nature of the over-current issue).

Edited 1 time(s). Last edit at 08/17/2013 01:13AM by Cefiar.

Cefiar Wrote:
-------------------------------------------------------
> Sounds a bit like it.
>
> That PSU has dual 12V rails internally (according
> to the page you linked). Each rail will have a max
> current rating, and if all your electronics are on
> the same rail, then it this might be the issue.
>
> You should try using one rail specifically to run
> your hot end/motors/fan, and the other to run the
> heated bed. I don't have the same PSU here, so I
> can't tell you which one is which.
>
> Note: The fact that it works again after being off
> for a short while usually means it's tripped an
> over-current limit. It switches off to avoid
> dumping a lot of current into a possible short
> circuit, and the only way to get it back is to
> disconnect the power and leave it till certain
> internal currents have discharged (time will vary
> depending on the PSU, and the nature of the
> over-current issue).

Ah.. makes sense.

I haven't looked at the circuit diagram of my RAMPS (because I probably won't understand it even if I did LOL), but there are two connectors each of which I connect each of two 12V output from the PSU. So, I would assume the designer of RAMPS made sure not all electronics are using one rail...

Having said that, I think I probably have connected my fan to the rail which the heated plate is connected. The fan runs all the time, as soon as I turn on PSU. Then when I turn on the heated plate, then I can hear the fan slowing down.

Even though each of the rails have separate max current rating, I wonder connecting too many electronics/devices will make them compete for the power and affect each other, even among the ones which are connected to different rails...

As I have stated, when I added a fan to my print, I believe I ended up adding more load to my PSU than what this fan draws, because the fan would have made the hot end to use more electricity for heating.

Now I am having a new problem, which I never had before. Near to the end of a big printing job (yeah, it really sucks!), my printer started shifting in Y direction. It created a quite nice looking (but useless and unwanted) even steps.

Ever since that, I keep having this issue with all my prints, sometimes it starts right away and other times it starts happening after 5-10 minutes.

I tried to adjust the potentiameter in Y motor Pololu. It seems to work better when I increase the current. However, I still haven't had a successful print - I am increasing the current by very little at a time, so I basically keep increasing the current ever so slightly and trying over and over, until I no longer have this shifting issue.

Anyway, I never had this shifting problem until now, and I haven't changed my Y motor Pololus setting for quite a while. So, it really makes me wonder if my Y motor Pololu is not getting enough share of the current, that now it cannot drive the motor properly..

I really need to insulate my hot end nozzle...

Yes, RAMPS has 2 inputs for voltage.

The 5A input runs the motors/hot end (D10)/switched fan (D9) outputs and your Arduino.

The 11A input runs the heated bed (D8).

The heated bed draws the most current, so if you're having issues when it's on (eg: fan slowing down), this means the voltage is dropping when it's in use. If you're using the same PSU rail for the motors, this will reduce the voltage being given to the motors, which could make them to miss steps occasionally.

Questions:
How are you getting the power out of the supply and to RAMPS (a pic might be best)?
Have you placed a load on the 5V line (eg: a 10 ohm resistor) of the PSU?
Where are you running the fan for the hot end from (D9, or constantly connected to the PSU)?

I have attached the connection diagram from the build instructions I got from MakerFarm.

Using the same terminology this diagrm uses, I currently do not have a Print Cooler running (fan to cool the printed objects). What I have is a Cooling Fan. So, I am not using D9.

I just noticed that the diagram shows Cooling Fans are connected to the lower 12V line, but I have my fan connected to the upper 12V line. At the time I didn't think I would matter, but now I think it may do - like we've been saying, I may have connected my fan to the rail used by the heated bed.

Having said that, I think it actually would be better to make the fans share the same line with the heated bed, however. I would assume the motors are connected to the other 12V rail (it only makes senses to distribute the load like that), so in a way by connecting the fans to heated bed's 12V line, I am not impacting the motor's voltage/current/movement... ?

But then again, these two 12V lines are coming out of the same PSU, so at some point load on one of the rails could/would still impact the electronics/devices on the other rail... You know, like when a light bulb dims slightly when a fridge comes on...

Good diagram.

The only thing is, while you've got 2x 12V inputs shown on the diagram, I can't tell what your PSU is putting out as this doesn't show where on your PSU those inputs go to.

Are you using 12V power from the Main ATX12V 2.x connector plug (24 pin - I will just call this Main ATX from now on), or from the ATX12V (4 pin) or EPS12V (8 pin) connectors, or the 6 pin PEG connectors?

Not all PSU's do this, but the Main ATX, the ATX12V/EPS12V connectors and the 6 pin (or 8 pin) PEG connectors are "usually" on different internal rails on the PSU (or there is some grouping, so ATX12V + PEG may be grouped together). This means they usually have different current limits (see the link below about "Rails").

I would try connecting the 5A input to one of the PEG or ATX12V/EPS12V outputs from the PSU, and the heated bed to the Main ATX connector outputs.

Also, you usually need a load on the 5V ATX supply output, so if you don't have one, I would connect a 10 Ohm power resistor (one of the white ceramic jobs, rated to at least 5W) between the 5V and Gnd on the ATX connector. A lot of PSU's expect current to be drawn on the 5V line, and if there is too much on the 12V without a certain amount on the 5V, the PSU will simply shut down.

BTW: For a pic of the connectors so you know which is which, see:

[www.hardwaresecrets.com] - Main ATX, ATX12V & EPS12V connectors
[www.hardwaresecrets.com] - PEG connector (and others)
[www.hardwaresecrets.com] - Talks about the concept of "Rails", particularly near the end of the page.
[www.hardwaresecrets.com] - Pinouts and wire color codes for the connectors.

Edited 2 time(s). Last edit at 08/17/2013 08:11PM by Cefiar.