Possible Overheating causing Hot End Failures?

[JohnStack]

Howdy, I have a R1 with the Rambo 1.2 board. After a variable amount of time, I get a hot end failure. At first, I thought it was the hot end; however, after replacing three of them, clearly this is not the case.

1. I replaced my stepper with a new Kysan #1124090. I have solid connections to the Rambo.
2. I changed the extruder position to a top mount (as they are currently sold).
3. During this round, I have increased running bed temps to 87C to get my heated bed surface temp up to 70C. (I have some wonky spots on my bed so I got some 3/32 aluminum and cut it into a circle. I placed it between the bed and the glass.
4. I have a peek fan running on the hot end.
5. This happens running prints from both the SD card as well as from Repetier.
6. Before this happens, I'm getting beautiful prints. I have two sets of feet now from [img]http://thingiverse-production.s3.amazon ... atured.jpg[/img]

I don't have any other extras on this particular machine that adds load.

I've been able to print no more than 1.5 hours before I get the telltale nubs and moments later, nothing coming out of the hot end.

Each time, the Ezstruder chews up the filament - indicating that the hot end is stuck.

I have not changed filament to see if it is the cause but currently, it's SeeMeCNC ABS - Orange.

So:

What in the heck? I think I've been systematic. I have not placed a cooling fan on the board but can - but I have to wonder, is it the increased load from the bed causing the hot end to fail?

On a separate note, it sure would be GREAT to be able to log temps out to the SD card...I'm not one to sit and watch the printer.

[Eaglezsoar]

Sounds like a power supply that starts to fail after it gets hot, reducing voltage to the hotend and causing your problems.

[JohnStack]

I'm willing to bet that is the case. I needed some confirmation....

Ok, so I have 12V, 30A external power supply sitting on my desk for the heated bed.
http://www.amazon.com/Universal-Regulat ... +switching

What do I need to get this hooked up? I've looked at several threads involving "relay", etc - but no guidance to wire them or to config them - with a twelve-volt electronic relay. I would prefer not to use a mechanical one.

I'll do a write up worthy of a sticky if someone can tell me what to do and what to purchase!

(Personal note: I've really tried very hard to only do bolt-on mods to my Rostock. I guess this now qualifies. )

From what I understand, you want to trigger or turn on a 12V line coming from the separate power supply by sending an external relay some voltage. The thermistor acts normally, always telling the temp circuit what the temp is so that it can turn on and off.

I assume the principle benefit is that when you are using a relay and a separate power supply, there is not the drain on the Rambo - so reduced heat.

[JohnStack]

Heaven helps those who helps themselves...

Wiring:
http://forum.seemecnc.com/viewtopic.php ... =10#p13794

Opto22 Solid State Relay:
http://www.opto22.com/documents/0859_So ... _sheet.pdf

And then my only question is:

How do I protect the circuit?

If I were to put in fuses, what size and where?

[Eaglezsoar]

Put the other 12V power supply in the Red circled pins.
The Bed will run off of the new Power supply the rest of the printer will run off
the old supply. You don't need relays or fuses, the card is fused.
If this fixes the problems, then I would get a larger power supply that can supply
more amps and replace the old one.

[geneb]

Unless you're seeing a temp drop after 1.5 hours, what makes you think it's the power supply?

g.

[JohnStack]

Eaglesoar, that's a great suggestion!

Geneb, for the reasons I stated above? Can you tell me what else would cause repeated hot end failures after the stuff I've tried?

[JohnStack]

Taking @Eaglesoar's direction:

13 mins to heat to 90 (with an aluminum spreader plate)
24 mins to heat to 105

I have the voltage up to about 14 - which I probably should not do. I assume that heat time as 12v would be a little longer but probably not that much. Next up, seeing if I can keep printing (which says it was an overheating problem.)

[Eaglezsoar]

Taking @Eaglesoar's direction:

13 mins to heat to 90 (with an aluminum spreader plate)
24 mins to heat to 105

I have the voltage up to about 14 - which I probably should not do. I assume that heat time as 12v would be a little longer but probably not that much. Next up, seeing if I can keep printing (which says it was an overheating problem.)

John, you could take the voltage up to 24V if you wanted to. I like to keep it down to about 16V or so because I don't like a lot of current going through the traces on the card. I think the card will
handle the 24V and some run it that way but I do not trust the traces not to burn up.

[JohnStack]

Hi, thanks. This maxes out at 14.8, it's only a 12V PS.

It will do me for now - I'll get the hot end back in running shape today and do a long print. I'm really hoping that's it - but then again, I have no idea why. The increased draw should _not_ cause overheating.

[Eaglezsoar]

Hi, thanks. This maxes out at 14.8, it's only a 12V PS.

It will do me for now - I'll get the hot end back in running shape today and do a long print. I'm really hoping that's it - but then again, I have no idea why. The increased draw should _not_ cause overheating.

Increased heating was not your problem, the temperature of the hotend kept dropping preventing the flow of filiament to your parts. That's the initial problem as you stated it.

[JohnStack]

True. Only an observation that heating times reduced as well.

I'm hoping that the separate PS keeps voltages in check. Still confused by why this would be necessary - but we lack any instrumentation within Repetier to know.

[Eaglezsoar]

True. Only an observation that heating times reduced as well.

I'm hoping that the separate PS keeps voltages in check. Still confused by why this would be necessary - but we lack any instrumentation within Repetier to know.

One of the biggest reasons that sometimes that this is necessary is the power supplies included with the kits are just barely powerful enough to run everything and in many
cases they are not powerful enough. When are you going to conduct your experiments to see if this fixes the problems you have been seeing with your hotends?

[JohnStack]

Tonight! All set to go.

[JohnStack]

THANKS to EagleSoar!!! Started at 6:30pm PST, still going at 6:15 PST.

This is a 12-hr+ print. A solid Yoda-gangsta. Why burn half a spool on Yoda? Cause he's done it for us.

[geneb]

Eaglesoar, that's a great suggestion!

Geneb, for the reasons I stated above? Can you tell me what else would cause repeated hot end failures after the stuff I've tried?

You haven't said anything that makes me think "Power Supply!".

You can click the "Debug" radio button above the repetier log window and you can see what temps are being reported continually. You'll get ALL the traffic between the host and RAMBo.

I'd almost suspect a wiring or cracked resistor issue before I'd go glaring at the power supply.


g.

[Eaglezsoar]

THANKS to EagleSoar!!! Started at 6:30pm PST, still going at 6:15 PST.

This is a 12-hr+ print. A solid Yoda-gangsta. Why burn half a spool on Yoda? Cause he's done it for us.

Please let me know if everything finishes okay. And is this the first time a print ever completed after so many hours?

[JohnStack]

Thanks, I appreciate it. Really. Glad to be back to printing and not worrying about whether or not it will...

Before I went to the Rambo 1.2, I was consistently getting up to about six hours (not having tried for anything more).

What I think exacerbated the problem:

Bumping up the temps to use the aluminum spreader

The aluminum spreader did even temps out on the glass but didn't completely resolve me having cooler spots. I still get edge curl in those areas with larger flat pieces (even with mickey mouse ears.) Steve told me to try the new v3 Onyx - so I'll order it next week.

My bar was to get a 12-hour print that I liked in terms of quality and size. I have gradually whittled away at QA/performance. With that done, I can move on to dual extruders and see what that can of worms holds for me!

[Eaglezsoar]

Thanks, I appreciate it. Really. Glad to be back to printing and not worrying about whether or not it will...

Before I went to the Rambo 1.2, I was consistently getting up to about six hours (not having tried for anything more).

What I think exacerbated the problem:

Bumping up the temps to use the aluminum spreader

The aluminum spreader did even temps out on the glass but didn't completely resolve me having cooler spots. I still get edge curl in those areas with larger flat pieces (even with mickey mouse ears.) Steve told me to try the new v3 Onyx - so I'll order it next week.

My bar was to get a 12-hour print that I liked in terms of quality and size. I have gradually whittled away at QA/performance. With that done, I can move on to dual extruders and see what that can of worms holds for me!

I am glad it worked for you, John. Now you can move on and not worry. I like the dual extruder BubbaSnow came up with, in fact I am going to print his parts.

[JohnStack]

Me too!!! Now, my only question will be....will the original PS handle both hot ends! LOL

[Eaglezsoar]

Me too!!! Now, my only question will be....will the original PS handle both hot ends! LOL

As long as you keep the configuration as it is now, I think that the dual extruders will be fine.
When I get a chance I want to find a power supply that I think would work for you and you will not have to use two.
Give me a week or so, I have a lot of other things I am doing or is taking a lot of time.

[Eaglezsoar]

John, the Yodsta model that I downloaded has errors that Netfabb basic won't fix. Did you print it with the errors in place or
did you find a way to repair the errors?

[JohnStack]

I repaired the errors in Netfab Basic, then ran Kisslicer. It's at the shoulders now. It's a solid print so I'm getting some cracks but overall, it looks good. I would definitely not do a solid version of it!!! Personally, I think it would be fine with 25% fill every other layer but I haven't tried it.

Personally, if I were to print it again, I would do 5 solid layers on the bottom because the right foot is raised up at the heel and the left foot has a canted edge on the outside. There is very little "footprint" to this model.

My dropcam is showing the print now. The printer is oriented to get a full view of the print (because I'm out of workbench space with the two printers I have running)

[JohnStack]

Solid form, Yoda will not take.

Got all the way to the eyes and the same failure occurred. Time to go to Fryes and buy a new power supply.

19.5 hours into it.

[mhackney]

John, just to be clear, when this happens your hot end has shoot down or cooled off preventing extrusion?