MK4 not that good? Slow Wifi, load cell sensor issues  

RE: MK4 not that good? Slow Wifi, load cell sensor issues

I bought one of first 100 MK4 and it is more polished than MK3, when it was released. Speed of transfers get better with last FW and if you are using Prusa Connect, it is OK, bcs you just upload it to the Prusa Connect and then transfers data on the background, which is nice work around and you do not need another PC like RaspberryPi.

Also have no issue with loadcell sensor. It can recognize the residue of filament automatically and just repeat the measurement. I just press print and does not care much about the printer and it works all the time. Recently I switched to 0,6 mm nozzle to speed up one big project, also with no issue with stock profiles from PrusaSlicer.

At this moment I thinking about selling my last MK3S+ and buy another MK4 right away.

Posted by: @avd

I have a ordered a MK4 kit, but if I'm looking now to a lot of review it seems the MK4 is a half finished product with quite some issues. I'm mainly reading that the WIFI and ethernet is extremely slow. Is that something firmware related or is it a hardware limitation. Can it be fixed in the future? I'm now using a MK3s+ with a PI zero 2 which has good transfer rates, I may hope that it isn't slower then that. 

I'm also reading the loadcell sensor leaves residu on the printbed before printing, as the nozzle doesn't cool down enough and the filament keeps  flowing. Is that still an issue with the latest firmware? I now still have the option to cancel the order so I'm hoping for some real life experiences. 

RE: MK4 not that good? Slow Wifi, load cell sensor issues

I have several Mk3S and Minis, and an XL. Just added my first Mk4 and I'm happy with it. Yes, the biggest issue is clearly the speed of data transfer. Ethernet is a bit faster than wifi but it's still atrocious, and from the analyses I've seen it's a combination of hardware and software, so I'm not sure it's an easy fix. 

Having said that, it's not as bad as it sounds, at least in my hands. Some big files may take a while (say a minute) but in the grand scheme of things (time to complete a print), it's acceptable. Sort of the time it takes to do 7x7 bed leveling on the Mk3S, which seems to take forever but it's a rounding error on a long print. To be clear, I think it's a big black eye for Prusa but not a deal breaker.

Loadcell: Very nice. Not a panacea, I've seen some issues. I found it important to do a good clean of the sheet with IPA between prints, and keep the nozzle really clean. But it's certainly an improvement over the old manual Live Z calibration. I have seen the filament deposits once in dozens of prints, but it seems filament dependent.

One feature I didn't appreciate until I had an Mk4 in my hands: The new hotend assembly feels nicer. E.g., you can just fold the fan to the side to get easy access to the nozzle (better than what I have on the XL even though it's the same nozzle). I should point out that the new nozzle is not nearly as easy to swap as I had hoped for. I'm spoiled by the Dragon hotends on my Mk3S and Mini printers. It's true single-handed nozzle swaps as the heater block is arrested in place. With the Mk4 (and the XL) the recommended procedure is to unplug the connectors, loosen the nozzle assembly (with two thumbscrews on the Mk4 and an Allen wrench on the XL) to drop it down, then unscrew the nozzle from the heater block and insert the new one. On my XL this led to a broken thermistor cable after a dozen or so swaps from pushing the cables back behind the fan. I can probably only blame myself but it's something to be aware of. I ended up swapping nozzles now without unplugging the cables. I just drop the hotend a bit, then unscrew the nozzle in situ while holding the heater block with the universal wrench. Even easier on the Mk4 because of the easy access to the hotend.

TL;DR: So far, the Mk4 feels like the upgrade to the Mk3S people have hoped for. How they thought the wifi transfer rates were acceptable before releasing it is beyond me. But then again, in my day-to-day use it's not as much of a factor as some posts have made it out to be.