Got mad at my hot end jamming so I chopped it in half 
Been wanting to do this for ages.
Curious if anyone else will spot the hardware problem that was causing the jams 
I know the nozzle does have a hole in the end, and we’re not just seeing it in this particular cross section. My eyes tell me there is something of interest just before the heat break ends and the nozzle begins, there’s some type of round formation in the filament, can’t tell if you’ve got a ball bearing, a machining imperfection, or what.
Noob guess (machinist noob to additive). On retraction pressure drop is drawing air into the channel at the junction between the nozzle and the throat (brass end and steel threaded tube)? It looks like there is an air bubble right at that junction, and from my limited knowledge of the functions performed in this part I am suspecting air bleeding in on retractions.
What you’re seeing here is the last skim pass before the heat break jumped out of the hot block and got mangled. (Superglue wasn’t enough. Will epoxy it together next time.) I also think the insulation on the hot block caused it to cut a bit tilted. I think the slice is maybe 1.5mm above centerline at the nozzle tip and 0.5mm above centerline at the inlet of the heat break.
There IS a long air bubble across the entire melt zone from the final retraction that caused the jam. That’s normal though. Always happens unless you do super-tiny retractions. Doesn’t cause any problems on its own.
A detail you guys probably won’t guess… this is supposed to be a Makerbot mk7 style thermal barrier tube with an internal diameter step.
Waaaaaayyyyyy too much metal on that heat break. The wall thickness is insane.
I think that hotend has seen better days
Well,… @Ryan_Carlyle ​​ I think you got the jam out…
@I_m_on_time nope, this hot end printed reliably for hours as long as the frequency of retractions was low. What you’re seeing here through the milling debris is a classic retraction-induced cold zone jam, where retracting hot filament into the cold zone caused the filament to mush up against the walls and freeze into a solid plug. Very common with all-metal hot ends unless they manage some specific details right.
@ThantiK you know, I hadn’t noticed that. It’s a little thinner in real life than the picture looks, but it IS thicker than it needs to be. Still, that thickness is pretty normal for a mk7 style tube. Those work fine as long as the cold zone is cooled aggressively. Replicator 1/2 (not 2x) cold zones have very good cooling and work exceptionally well if you assemble them right.
I’ll post pics tomorrow of what the big issue was… I don’t think anyone here has enough Makerbot hot end expertise to spot the thing that’s out of spec 
Unimportant detail if anyone is curious, the filament was a fresh, reasonably dry spool of ColorFabb PLA/PHA. Print looked great before the jam.
How much did you retract to pull it into the cold zone?
I agree with @ThantiK that the heat break looks big! Would also be nice to see it milled along the other axis for clarity of what it looks like.
@Ryan_Carlyle looks to me that the stepped area is way too long. I think it should stop short of the the necking on the thermal tube, keeping it on the hot side. Regardless awesome works and thanks for sharing!
Looks like you need a new E3D. http://store.timeshell.ca. (Shameless plug).
@Steve_Johnstone basically got it. With this thermal barrier design, there’s a step in diameter that is supposed to be right at the bottom of the cold zone, directly above the heat break neck. The purpose is to have a small diameter for a “snug” fit in the cold zone for maximum filament cooling to prevent heat creep, but larger clearances in the transition and melt zone to reduce viscous friction. According to some Stratasys patent literature on related designs (not this one specifically, this is open) – the step acts as a “meniscus barrier” to keep molten filament from flowing up into the cold zone and causing a jam. It normally works really well. It’s in all the Replicator 2s for example and those things are workhorses. (I personally think this design is more compact, robust, and reliable than an E3Dv6, and it has also been around a lot longer.)
Problem is, this specific tube was mis-drilled. The step is up way too high, so the large diameter section extends into the cold zone where the molten filament will freeze into a solid plug. The meniscus barrier is in the wrong place so repeat retractions aren’t controlled properly. And (in a detail you can’t see in the photos) the second drill-out of the larger diameter across the melt zone and transition zone was REALLY rough, like there was something wrong with the drill, so any plugs that did form were super-mega-stuck. I’ll see if I can post some microscope shots when I’m on my computer later today.
For the people who asked, retraction was 1.3mm at 25mm/s, direct drive extruder. Default PLA setting for this hardware. The problem was REPEAT retraction. I could print for hours without issues but then jumping between any kind of small features at the end of the print would jam it instantly.