After calibration, first print is good, all other prints fail

[rhoded]

I recently upgraded my hot end, which required I take the printer almost completely apart and put it back together. Since then, after a calibration, I can print once fine. Second and other prints are not so good or worse. The calibration is being lost since the first layer quality is worse on subsequent prints.

I'm sure this isn't enough information for a perfect diagnosis, but what should I be looking for in this case? My gut tells me belt tension is the cause, but the belts are tight. Could they be too tight? I do get a weird loud noise from the Z-tower...

[geneb]

Did you install an HE280 hot end & accel. board kit?

Does your v2 use the ball/cup arms or the old style u-joints?

g.

[rhoded]

I did not install that HE280 hot end & accel. board kit, I did install an E3D Chimera. Have yet to use it for multiple filament prints (short on time at the moment + too many obligations) but first thing I'd like to try is the nylon and extended heat range printing that come standard with E3D heads.

My v2 does use the ball/cup arms and I've heard of people "sanding them down" but I'm not sure what that means. So far, the printer has been fairly good; I've had a handful of hot end clogs and I have had to calibrate every once in a while, but lately it is not behaving itself. I just printed three small (<45 minute) prints and each one was worse than the one before it. The last print started pealing (I'm printing on heated glass with PLA) and when I re-calibrated Z-height, it was off by -0.15mm, which is way more than usual.

I have started experimenting with the belt tension. At the start of my last three print session, I did loosen the Z-tower belt-tension because it was making a loud noise. It is now much quieter and getting three prints done consecutively feels like a success. I have also loosened the Y-tower belt-tension before this next round in case that is the issue.

I've been looking at the TrickLaser upgrades for the arms. Before, I had no reason to get them because the printer seemed to be fine; I just didn't like the look or feel of the plastic arms and cheapskates, especially that the end-stop screw on one of them is all the way in and on another, the screw is out quite a bit and actually isn't that tight, so it pivots a bit. I presume that the TrickLaser parts are much better since they are made out of aluminum and don't have the word "cheap" in their name...

[geneb]

To be honest, if someone is sanding down the ball cup arms (and I'm not referring to the small amounts of flash around the ejector pin marks), they're idiots. They're precision made, injection molded parts. Sanding either the ball or socket will basically ruin them.

Do you have the melamine carriages or the injection molded ones? If you've got the IJ ones, there's no reason to "upgrade" them - there's no need.

If you DO have the melamine carriages, I would suggest you order the IJ set from SeeMeCNC.

One thing that can help with the loose adjustment screws is to change to a longer screw - only needs to be 1/4" or so longer. If you've got one screw all the way in and one nearly all the way out, you've got some kind of mechanical issue with the printer. Make sure your towers are square to the base and that the heated bed is fully perpendicular to the base.

You can use the accelerometer probe board with any of the E3D hot ends and your life will be a LOT easier if you get it - it makes calibration a breeze.

g.

[rhoded]

Sorry for the delay. I'm glad that sanding isn't needed because I didn't do any of that (and kind of glad it isn't because that would make me second guess my purchase). I have the injection molded carriages, so also a plus according to what you say.

So your tip about making sure the towers are square to the base was the bingo moment. Mine were actually a bit loose, so I must have forgotten to tighten them up when putting it back together. However, even after fixing them, the printer slowly loses calibration. What kind of "mechanical issue" could I look for next? I will definitely take another look at the towers in case they still aren't right, but I triple-checked them last time.

As for the accelerometer probe board, I would LOVE to install one. However, the hardest part about upgrading to the Chimera hotend was fitting all those !^^!@%$!ing wires through the towers. There is no way I'll be able to fit any more in any of the towers. I might have to just wire them some other way, do you have a suggestion about that?

Thanks for your help so far!

[Xenocrates]

The pulleys may also be slightly loose, or the idlers loosing tension. There is also the very real possibility that your endstops are moving slightly, or the screw which triggers them is. As for attaching the accelerometer, you can place the wiring up the outside either around a dowel or bar, or inside a pipe/conduit/duct. I use some standard wiring duct on mine (A side effect of having nearly 100 linear feet of it laying around)