Rostock - Build Details

How is it going?
Sometimes good and sometimes bad,
can't really say now

As part of my custom build of a new delta printer from a ground up, I thought I'd thoroughly review my build experience with the Seemecnc Rostock MAX (V1) and use that as a baseline. A couple of the build issues I'll mention have been addressed in the latest kit update (V2). My views may be a bit biased, since I haven't worked with any other kits, and numerous delta printers have entered the market since my purchase late last year.

The Good

The Rostock MAX kit itself was actually not very well reviewed when I purchased it, but I found the overall build to be surprisingly straightforward, and maybe I was lucky, but I managed to get it printing decently right from the start. Granted, I took my time building it (~4 days, around 30-36 hours total), and I had access to a shop full of tools and spare parts, since I assembled the printer in Yale's CEID space.

A couple of good things about the kit that I appreciated:

Extra parts: Seemecnc was nice enough to include extras of all the fasteners and small components. That may seem like a small thing, but it definitely helped to not have to worry about losing or misplacing any small pieces during the build
Detailed Assembly Guide: The assembly guide, even though it was technically still in flux (and sometimes a bit confusing), was adequate enough to get me from start to finish.
Open-Source Files: It was really nice to have the DXF files on hand for all the major panels used in the kit so that I had the option to replace or extend certain components (which I would eventually do).
Melamine-Laminated Wood: The wood used in the Rostock kits, in my opinion, are far superior to the plywood (which tends to warp more) and acrylic (which may crack) used in other kits. I found it especially easy to drill extra holes later on as I was making more modifications. It's a shame that you can't get this wood easily on the East Coast.
Complete Software/Hardware Integration: I really didn't have to think very much when trying to get the software and firmware to play nicely. Seemecnc had a verified bundle of firmware and host software ready to go (both Repetier-based). All I had to do was plug and play, and everything played nicely with the default hardware components.
EzStruder: I didn't realize how good I had it with the EzStruder until I started looking at alternative extruder designs. This thing just works (at least for 1.75mm filament), despite its minimal design. Moving away from the EzStruder to a more custom (and likely printed) extruder design is something I'm not looking forward to, as I'm not confident I can put together something with as much reliability.

The Bad

The majority of my issues with the kit will probably sound very nit-picky and specific, but they're issues that I'm keeping in mind while formulating my new delta design.

Melamine-Laminated Wood: It was really frustrating to not be able to find an online source for this material that would ship to the East Coast. Even with the DXF files, I didn't have the reassurance that my material would cut with the same tolerances as the Rostock MAX material (and in most cases, it didn't).
Seemecnc-specific Components: Aside from the wood, I also had difficulty finding another source for the belt bearings and cheapskate carriage components aside from Seemecnc. I'm appreciative that Seemecnc is a reliable source for these components, but I didn't like the idea of being dependent on a single vendor for key components. I'd rather pay a little extra to have the flexibility of choosing the part supplier.
Lack of Electrical Connectors: I didn't really understand why Seemecnc didn't just include the Deans connectors that they recommended. It would've been a minimal additional cost but would have eliminated a major headache for those customers who weren't aware of the lack of connectors and didn't have immediate access to an electronics store. Assembly flexibility is nice and all, but I think kits should at least provide 100% of the components for their suggested approach.
Hotend Assembly: I had a great deal of difficulty fitting the power resistor into the hotend as part of the V1 build. It required me to wrap the resistors in aluminum foil to get a tight fit with the hotend. Was it really too much trouble to machine the hotend appropriately for the selected power resistors? Wrap the foil too loosely and you don't know if you're making sufficient contact, but wrap it too tightly, and the foil scrunches up and unravels as you try and insert the resistors. At the very least, a short, official video for this step would have been very helpful.
Effector and Arm Sanding: I always get a little bit unnerved when the instructions tell me to sand and do additional post-processing on a part. To date, I still don't know whether or not I sanded the end effector and arm joints appropriately, and it seems that this might still be a major stumbling block for other customers as well to date. It's not surprising that effector-arm upgrades are some of the most popular in the Seemecnc forums. Perhaps it would've been better to include Traxxas-style joints as a default in the kit?
Belt Tensioning: This was another part of the assembly that I eventually updated, as I didn't like how belt tension depended on just the friction from a bolt at the top of the towers. I also never knew whether or not my belts were tensioned appropriately, but this seems to be an issue with all other printers as well. Perhaps a calibration print piece can be designed specifically to test for loose belts? I would guess such a print would test the system's ability to handle fast motion transitions.
Bed Leveling: Oh man, this remains a headache to this day. In comparison with other delta kits, I think the Rostock MAX has a lot more components, which (to me) means greater odds of misalignment during the assembly. Dynamic bed leveling with a zero-probe in software might really be the optimal solution, especially for a delta printer, which doesn't have as much difficult adjusting for z-height during prints as a cartesian printer might. For the V1, I would have preferred that the tower-zero-ing was done in software/firmware (which I believe Marlin allows) instead of mechanically with the bolt travelling on the carriage. And this may be asking for too much, but it would have been awesome to have a full list of suggested adjustments for observed print issues. Even if they don't resolve the problem, at least it's an aggregated list of steps to go through for customers trying to debug their kits.