I disagree with the article. Metering the hotend is all well and fun, but it does not collect the data that actually matters.
First off, makerbots extruders suck. They fail because of a poor design, and nobody cares to find out why and actually fix it. That is because it is a proprietary piece of crap, and any design improvements will be stolen and patented by makerbot. I’m not making that part up, as makerbots current extruder design is a stolen derivative of an open source design, which they then patented.
Next, the failure of an extruder can be caused by a variety of incorrect settings or abuse. The filament could be dirty or poorly manufactured. The idler tension too high or too low. The type of filament is incompatible with the extruder. The motor current is too high or too low. The motor gets too hot. The motor is underpowered. The cooling fan fails. The head crashes into the bed. The thermistor breaks or loosens. The heater fails. A wire gets loose or breaks. The nozzle loosens. The PTFE melts.
Adding a timer to the extruder will not give any meaningful data, as any one of these settings could be incorrect. If you want meaningful data, you need to collect as many variables as you can, and mitigate the rest.
Ultimaker has this in their firmware. I agree it’s not the be all and end all but could be useful to have. More like a odometer on a car, just nice to know how many miles the old girl has actually done
Makerbot’s smart extruder is the most unreliable outside of the QU-BD, and the smart extruder is also the most complicated one outside the industrial machines. Making them more complicated isn’t always useful.
I think it’s possible to make a lot of the jammed ones to work again, but it’s hard to make it worth investing money in their machine if it turns out the part is locked up with encryption.
I wish people would look up a product’s reputation before buying it. After all, it’s $3000 they’re spending on a machine.
The Ultimaker stats code was linked in the article comments, and it only writes the numbers to EEPROM once an hour to avoid wearing it out. Seems like lifetime stats would have pretty large rounding errors if the printer’s not powered continuously.