Heaters killing Rambo while heating up to temp

[summyman]

I recently completed the Rostock Max build and have tried to perform calibration procedures. Everything is fine until I attempt to measure the distance from the hot end to the heated bed. I turn the hot end on and it heats up to 190 no problem. When I turn the heated bed on after the hot end is at temp it instantly kills the power to the Rambo. I restart the Rambo (just by turning it off and back on) and heat up the bed first and then when I power up the hot end, the Rambo is instantly killed again. When both heaters were powered up together they both got to approximately 90% before the Rambo board shut down. If any one knows what might be the problem or how to trouble shoot it I would much appreciate the advice.

Summyman

Rostock Max
Repetier 8
Rep-Host 9C

[ApacheXMD]

Is the power supply cutting out, or are you sure it's the RAMBO that's shutting down?

Sounds like it may be a power supply issue; sounds like it's cutting out due to the load.
I never liked using hacked up ATX power supplies when a dedicated 12v power supply is so cheaply available (ebay, amazon).

[summyman]

I'm not sure if it is the Rambo or if it is the power supply. I just see the result of the Rambo losing power and was surprised to see it restart each time. The ATX power supply does appear to be of a lower quality unit.

[Polygonhell]

Does the fan on the PSU stop, that will tell you if it's the PSU cutting out.

For this particular usage cheap ATX PSU's can actually be better than expensive ones, some more expensive units have safety features that stop them delivering 12V underload without a significant load on the 5v line. I personally don't like using ATX PSU's for this purpose, but the old ATX PSU's that SeeMECNC used were actually very good. Several people seem to be having issues with the new ones.

You should test the resistance of the bed disconnected from the RAMBO board to verify you don't have a short, but it's probably a PSU issue, there seem to have been several of them lately.

[626Pilot]

I never liked using hacked up ATX power supplies when a dedicated 12v power supply is so cheaply available (ebay, amazon).

Agree. But if you want to see something worse - I've seen commercial, pre-assembled 3D printers that have two laptop-style power supplies, because someone was scared of doing it right.

I think the PSU is probably the culprit here. Probably it needs to be exchanged.

[geneb]

It doesn't have a THING about "someone afraid to do it right".

Do you honestly think letting a vastly inexperienced builder near a power supply where they've got to hard wire the mains plug to it in order to use it? That's begging for a lawsuit because Joe Derp burned his house down or electrocuted himself because of incorrectly wiring the power supply.

The power supplies that you find on eBay, etc. are typically open-frame or nearly so and NOT UL certified.

g.

[626Pilot]

It doesn't have a THING about "someone afraid to do it right".

http://www.kronosrobotics.com/3d/makerg ... ndex.shtml

Would YOU design a printer like that? I wouldn't. I might use that kind of setup if I was just trying to throw something together using whatever I had laying around. They use USB power for the controller board, a 19V power supply for the steppers and hotend, and a 12V for the bed. (12V to the bed but 19V to the hotend? Seems backwards to me.) For all that trouble, why not use 12V steppers like everyone else? For close to $2,000 fully assembled, I would expect a more elegant solution, or at least one that isn't awkward.

As for lawsuits... LOL. Some guy's Rostock caught on fire, and if he hadn't put a smoke detector nearby it could've been really bad. John is still selling Rostock kits anyway.

[ApacheXMD]

It doesn't have a THING about "someone afraid to do it right".

Do you honestly think letting a vastly inexperienced builder near a power supply where they've got to hard wire the mains plug to it in order to use it? That's begging for a lawsuit because Joe Derp burned his house down or electrocuted himself because of incorrectly wiring the power supply.

The power supplies that you find on eBay, etc. are typically open-frame or nearly so and NOT UL certified.

g.

I may have a different perspective than others when it comes to electrical safety (I'm and electrician by trade), but I'm not sure that wiring a hot/neutral/ground for a mains connection is any more difficult or more dangerous than the rest of the build of a 3D printer.

To me, trying to push 10amps via 12volts through a bundle of 18awg wire is asking for more trouble than a 120v mains plug.

I know for sure that the Viotek ATX power supply that came with my RostockMAX is definately NOT UL listed. And even if it was, the instant you cut/splice the wires, your UL listing went out the window.

My view is that ATX power supplies were chosen purely for their low price. And I think from what we've seen with the trouble that some folks are having, low priced, questionably rated ATX power supplies are probably not the best choice for our printers. They seem inconsistent in their current pushing capabilities and often, actual voltage depends on a load on the 5v rail. And based on my own past experiences, and from posts on this forum, they also sometimes seem to cutout prematurely without an apparent "overload".

Like so many things in life, Good, Reliable, and Cheap are usually mutually exclusive. Like the old saying goes: pick two.

I went with a no-name switching power supply, not so much because they are more reliable, but because I feel they supply the power required without all the headache of dealing with the huge bundle of wires.

An advantage to the cheapy ebay/amazon supplies is that one is free to use large gauge wire for the heated bed.
And you can run your own smaller wire to your lighter loads.
And output is 12volts with everything from no load to max load.
And on my power supply there is a trim pot that lets you boost the voltage a bit.
And it's physically smaller.

What would be nice is if SeemeCNC made up a wiring harness that plugged right into the connectors of an ATX supply. The harness could combine multiple 12volt wires into a single larger gauge pair of wires for the heated bed. It could have labeled leads on the other wire ends telling you exactly where the wires would go. Wiring harnesses for car audio applications have nicely labeled wires for easy installation, and are very reasonably priced too. It probably wouldn't be too hard for Seemecnc to injection mold a harness to include with their kits. But that is still contingent on them choosing an ATX supply that will actually output the current advertised, and stay on consistently.

This is all just my opinion, and everyone is free to use whatever he wants.

[Eaglezsoar]

Apache, I agree with your point of view but I also agree with Geneb. You absolutely cannot design anything for sale that requires the user to wire the AC mains. The chance of a
lawsuit is absolutely huge. If the user chooses to "upgrade" to the type of supply that requires an open type of AC mains that is their right. I do agree that it would not be that
difficult to design a plugin type of harness from a higher output supply. It would be a huge improvement over what is sent presently with Rostock kits. I have read many times
on this forum that the power supply that they ship with the kits just doesn't have enough output to handle the requirements of the electronics and the harness that plugs into the
24 pin output of these supplies is commonly available. I understand the need to keep the costs down so that the kit can be sold for a reasonable cost but the improvement of the
supply and the way it connects is truly needed.

[JohnStack]

They seem inconsistent in their current pushing capabilities and often, actual voltage depends on a load on the 5v rail. And based on my own past experiences, and from posts on this forum, they also sometimes seem to cutout prematurely without an apparent "overload".


I went with a no-name switching power supply, not so much because they are more reliable, but because I feel they supply the power required without all the headache of dealing with the huge bundle of wires.

What would be nice is if SeemeCNC made up a wiring harness that plugged right into the connectors of an ATX supply. The harness could combine multiple 12volt wires into a single larger gauge pair of wires for the heated bed. It could have labeled leads on the other wire ends telling you exactly where the wires would go. Wiring harnesses for car audio applications have nicely labeled wires for easy installation, and are very reasonably priced too. It probably wouldn't be too hard for Seemecnc to injection mold a harness to include with their kits. But that is still contingent on them choosing an ATX supply that will actually output the current advertised, and stay on consistently.

Thanks for putting this in here (and giving me a few more decisions to make.) Safety is job #1.

As you know, I'm designing a base and top. It's "open concept". In other words, I had planned on cutting in a space on one of the wings for an ATX power supply. Now, after reading your post, I'm considering leaving the builder to their own devices w/rt power. (Not having a place to put the PS on the base) Everyone is used to external power supplies on all of their other devices anyway and this is definitely _not_ a computer.

Question: If I were to change to a switching PS, what gauge wires did you use from the switching power supply to the connector that hooks up to the Rambo?

[Eaglezsoar]

12 gauge for 20 amps
10 gauge for 30 amps

[JohnStack]

Thanks! The difference in gauges sealed the deal. I'm not going to do mounts for a PS. If I ship these commercially, customers are adopting my design and some thoughts by the community on what works best. Most of the pros including yourself have switched over to external PSs.

To sum it up: Desk space vs. Safety plus a little more hassle for a maker.... Safety wins.

[626Pilot]

I went with a no-name switching power supply, not so much because they are more reliable, but because I feel they supply the power required without all the headache of dealing with the huge bundle of wires.

I notice the current the stock PSU puts out varies significantly with load. The LED ring I put together definitely shows the effect, which is annoying. If I use a better power supply, will it limit that effect, or am I better off going with a voltage regulator?

[lordbinky]

Both , what surprised me most about my stock power supply was how LIGHT it was. I want big beefy inductors and capacitors in my power supplies damnit. I hate working inside the base though so I'm sticking with the stock until something forces me not to.

I did try an old antec 480W PSU that had much better specs and listed 0A as the minimums for it's rails and was just so much heavier it gave me (false) confidence in it. I pulled it from my random parts box so whether it's good or not is completely up in the air, but it would shut down trying to heat the hotend or bed so I went back to the stock one.

[Tincho85]

Seems I'm having the same issue. At first I thought it was a usb problem, because I use a 16ft cable.
But now, before printing, I heat bed/hot end using the lcd controller, and print with the SD card.

The Rambo doesn't shut down, it just says "Killed" while it's heating. Then I can start the heating again with no problem.
The weird thing is that this doesn't happen every time.


Is the PSU? If so, what's the best option? a 12v switching one? how many amps?

Notes:
Using atm a generic 600w psu with 2 fans.
Bought a 24v 30a PSU + SSR 100a DC-DC for the heated bed only, it's not intalled yet.


Any comments are highly appreciated.
Martin.

[guanu]

The Rambo doesn't shut down, it just says "Killed" while it's heating. Then I can start the heating again with no problem.

the "killed" message appears when the inactivity timeout kicks in.. its set where if you were to turn on the heat and forget about the machine, it will shut heat off by itself as a protective measure. I'm not sure what the inactivity timeout is set at by default offhand, but I believe it may be in the eeprom if you want to check on that.

Guanu

[Tincho85]

the "killed" message appears when the inactivity timeout kicks in.. its set where if you were to turn on the heat and forget about the machine, it will shut heat off by itself as a protective measure. I'm not sure what the inactivity timeout is set at by default offhand, but I believe it may be in the eeprom if you want to check on that.

Guanu

Sin título.jpg (11.15 KiB) Viewed 12013 times

This value?
1800000 ms = 30 minutes

and what "Printer active " does?

[geneb]

That's the number of seconds the printer has been moving.

g.

[Nylocke]

I think he meant the highlighted value?

Yeah thats the right value. It will turn off the heaters after 30 minutes of inactivity (No machine movements/temp adjustments)

[Eric]

It doesn't have a THING about "someone afraid to do it right".

http://www.kronosrobotics.com/3d/makerg" onclick="window.open(this.href);return false; ... ndex.shtml

Would YOU design a printer like that? I wouldn't. I might use that kind of setup if I was just trying to throw something together using whatever I had laying around. They use USB power for the controller board, a 19V power supply for the steppers and hotend, and a 12V for the bed. (12V to the bed but 19V to the hotend? Seems backwards to me.) For all that trouble, why not use 12V steppers like everyone else? For close to $2,000 fully assembled, I would expect a more elegant solution, or at least one that isn't awkward.

No, but that doesn't make it wrong or inadequate, just awkward. Having a separate power supply for the biggest load on the machine (the bed) is actually a reasonable design decision, especially since that big resistive load isn't very fussy about the quality of its power. It also means the main power supply doesn't need to be as big or deal with huge load spikes from the bed.

As for steppers, you've leaped to the wrong conclusion. I doubt they're using anything special for steppers. The steppers in most of your machines probably have a rating in the 2-5 volt range, yet they are commonly used at 12V...or 19V...or 24V. That's because your motor drivers are current limiting devices, and an appropriate current setting is what really matters.

A more likely reason they picked those particular supplies probably has more to do with finding a cheap supply of them, if I had to guess.