Some of us in the San Francisco Bay area managed to do a bulk buy of used UP Minis being retired from an education group. (Sorry, none left ) Naturally we're all trying to restore them to the best possible operational order. And I'm apparently the lead engineer on the effort.
So far, I've spent a couple dozen hours doing test prints and various diagnostics (using the latest version of UP Studio, Version 2.2.0 (540) from this site), with generally positive results. Used a couple machines, several print heads, always with good quality PLA, with standard temp (210ºC) and up to 30ºC higher. Have encountered a number of issues, and searched for fixes here in this forum, but generally found anecdotal reports without clear conclusions.
In the hope of coalescing the relevant info, here's a list of questions and issues. Please add your comments and suggestions in response, and I'll try to assemble a complete report:
- The most critical issue is the appearance of extrusion stepper skipping. Often occurs about 30 min. into a print, but can also appear on an irregular basis. Appears to be thermally related. Back-pressure in the hotend exceeds the stepper motor torque, the motor skips a step, and you get a clunking sound. Sometimes a single skip. In bad cases, skips continually and never recovers (in which case, extrusion stops and the UP Mini just prints air).
- There appear to be thermal control issues with the Up Mini. I put thermocouples on the nozzle and the black finned cooling block to try to diagnose them. The nozzle does tend to run about 10ºC cool and fluctuates about the same amount. But I don't think that's the cause of the stepper skipping steps (causing a clunking sound and ruining the print).
- At one point, I suspected black, finned cooling block in the hotend was getting too hot, causing the fiber to swell too high in the hotend throat and raising the back force. Doesn't seem to be a problem. With the control baffle rotated fully open, the block stays at a steady 39-40ºC).
- My current premise is that there's a software error in the UP 3D slicing software that isn't correctly matching stepper speed with the current horizontal speed (which controls diameter of the thread layer down). Is the feed going too fast at the moment for the viscosity of the melted plastic (related directly to nozzle temperature)? One thing that leads to this suspicions the dependence of the trouble to part geometry. For example, a snowman made of 3 spheres tended to always fail at the narrow spot at the neck.
- Some of the perforated platens (print surfaces) were missing or damaged (bubbles under the black layer where it delaminated). Any suggestions inexpensive replacements in the US? (Prices that I found seemed bit high.) There were several vendors at the Bay Area Maker Faire this last May who were offering new coated flex plates. Don't remember their prices. Anybody try the thin adhesive-backed stuff?
- Always printed with a raft, and usually didn't have an adhesion problem even without heating the bed, with two exceptions. In the first case, it appeared to be skin oils, etc., on the surface. Wiping the platen with isopropyl alcohol fixed this.
- In the second case, the nozzle was too close (touching?) the platen on the first layer and was effectively blocked, causing immediate continuous extrusion stepper skipping! (Hmm... Could an irregular, too small vertical layer step be causing similar stepper skipping later in the print?) This "touching the platen" seemed to be an isolated software glitch. Had just done the 9-point platen calibration.
- The UI for platen heating has several problems. The temp reading in the top line of the UP Studio window seemed to just show some erroneous random number, e.g., 60.42, for long periods. Not intuitively obvious what setting the bed heater time, e.g, to 15 min, means. (The bed will heat for that long before the print starts? Starts coincident with the print? No relationship to any print?)
- The units we got came without any fiber guide tube. Are they necessary? Presume a standard PTFE tube, 2mm ID, 4mm OD, will work. In any case the fiber routing seems strange -- the hinged translucent plastic top gets in the way of a smooth curve from the chassis back edge guide to the print head. And it appears that some guidance (tube?) between the spool and back edge guide would be useful to avoid tangles at the spool.
- Printed some 20mm calibration cubes (With X, Y or Z embossed on 3 sides). Sizes were accurate in all three dimensions, sides were flat. Except for one of the printers (mine ) where one of the sides had a protruding wavy bump between the edge and embossed letter. (Haven't checked belt tension, smooth linear actuator movement, etc.) Anybody else see this?
Thanks for your attention,