Some of you may remember my review of the Leapfrog Creatr. I wrote it after I’d had the machine for a few weeks and at the time I was quite impressed with my purchase. To a large extent, I still am, however I have had a fair share of issues and breakdowns. Some of these issues were down to the fairly hasty fashion in which the machine was assembled and others have been connected to design flaws in the machine itself. As I mentioned in my review, I have a fair amount of experience with electromechanical devices, being trained as a broadcast engineer in the tape era. Since then I have had to keep up with the times and subsequently, a good deal of my work now is connected with computers, networks and software. 3D printers are a good match for my skillset.
When I ordered my machine from the Netherlands, I was not under any illusions about the possibility of warranty support on the machine. I was on my own. After building my CNC router from scratch, the prospect didn’t faze me but other clients have proved to be much less forgiving of the machines idiosyncrasies and faults. I have seen some really angry threads on the forum attached to their official site and more than once the buyer has referred to the machine as a doorstop or paperweight. The term “boat anchor” hasn’t taken off in Europe! Back home, I was jumping the hurdles as they came along but other users were not expecting so many issues and were not holding back their disappointment. As much as I could, I joined in the discussions and offered technical advice where I felt I had some understanding and fortunately, there were other members, much more knowledgeable than myself, adding their considerable input. This extended right up to a beautifully sorted version of the Marlin firmware for the Arduino that ironed out a bunch of issues.
Okay, so what was good and what was bad? Let’s kick off with the bad stuff, it’s everyone’s favourite eh!
One of the things that I noticed early on with the machine was the wiring to the print head was basically a big bunch of wires with a helical wrap on it, that dragged back and forth across the bed with the movement of the head. Depending on what you were printing, it could also catch on the print and pull it off the bed. The fact is that keeping the print stuck to the bed is one of the most tricky aspects of FDM printing in plastic. Having the machine fighting you on this is massively frustrating. I must admit that I looked upon this as the main issue until I had my first component failure on the print head. The culprit was a thermistor and in the first instance, the wiring had snagged on something and caused the in-line connector to part. Easy fix, plug the connector in and move on. However, this was the tip of the iceberg. The next time the thermistor quit, it stayed dead and I could only call it a component failure. I sent an e-mail off to my contact at Leapfrog and he very quickly had a replacement in the mail to me. I waited and waited for it to show but fearing it had been sent via camel train, I bought half a dozen off a US site and specified transonic postage. I had the US thermistors well before the Leapfrog one arrived and proceeded to fit them. After many years performing board level repairs on equipment, I always order more of a needed component than I actually require. This is to cover off the rare occasion where you might destroy a component in the process of fitting it.
After removing the thermistor and seeing that it was held in place with a grub screw, tensioned onto its fragile glass envelope, more spares was a good idea. Sure enough, I mashed two of the replacements in the process of fitting them. I wanted a snug fit that would allow the device to read accurately and it was truly a microscopic fraction of a turn on the Allen key between good and gone. With the new unit in place, the machine returned to life and stopped throwing “minitemp” error codes. At this point in time I hadn’t dealt with the strain relief issue.
The next drama was also at the print head, this time one of the heaters had stopped working. The heaters on the leapfrog are neat little cans, not high power resistors stuck on with refractory cement. Unfortunately, this makes them a bespoke part for the most part and mine had succumbed to the constant movement of the cable loom. The steel lead had sheared off right at the can due to fatigue, so no chance of doing a bodge to reattach it. Even so, solder would be useless that close to the heater, it would’ve dried out (crystallised) in no time and failed. New heater required. I managed to get one on the way from Holland and in the meantime the machine sat idle. I know now, that I could’ve edited the firmware and run the machine on a single extruder but at this point, I hadn’t dabbled in the software guts of the Leapfrog. Eventually, the new heater arrived and was fitted. This time, I made a serious attempt at isolating the movement from the wiring loom. Instead of designing some sort of neat 3D printed doover to solve the problem, I used a couple of zip-ties to pin the loom down where it met the print head. I must have been successful as I’ve had no more cable fatigue issues. In addition to this, I replaced the spiral wrap on the loom to a split, convoluted cable sheath (like car manufacturers use) which has almost no propensity to get caught on the internal fixings in the machine or the print.
As far as mechanical issues are concerned I have only experienced a couple. The bearings that the print head runs on, that make up the Y-Axis of the machine are only a press fit into the vertical sides of the structure. The front right hand side bearing came out of its hole and was sliding freely up and down the shaft. I remedied this with some Loctite, however, a better solution would be to have the holes counter bored from the inside face so that there was a land on the outside face. The bearing could be snugged in and then a set screw could be tapped into the underside of the horizontal plate above it to keep it in place. The other issue was the threads on the mounting plate that clamps the front drive plate to the print head carriage. I have had to access this part of the assembly so often that the thread in one of the holes on the plate has half stripped out.
The last set of issues that I have are primarily ones of design and materials. With a 3D printer, the extruders and print head are going to be the main areas where issues will occur. Subsequently, access to the extruders and drive mechanism is very important. Unfortunately, on the Creatr this has not been given a high priority in the design process. The drive motor for the filament and the extruder nozzles are both buried behind the cooling fans and a diffuser plate. To access the top of the extruder tube, where blockages occur, involves completely dismantling the print head bar the main horizontal plate and the plate where the extruder tubes mount.
The filament drive is behind the cooling fans and this can cause issues when feeding in a new filament. The incoming filament can miss the extruder tube and be uselessly coiling up in the area between the tube and the feed rollers. I find the only way to combat this is to have an LED torch handy so I can see what’s going on.
The extruder nozzles are another problem as they only protrude a small distance below the carriage. Melted and solidified plastic mess can accumulate under here and damage prints by colliding as the mechanism moves around. I have a pair of needle nosed piers handy to remove any accumulation or stray extruded plastic. Any work on the tubes or nozzles requires the carriage to be disassembled which can be time consuming and due to the tab and slot construction, care must be taken in the reassembly process that everything lines up.
The Z-axis on my machine runs on 10mm all-thread or threaded rod as we call it here in Australia. As with many printers, no form of linear guide is employed to ensure accuracy, the design relies on the rod to be straight. On my machine, I’d have to say the rods are decently straight and the effect known as “Z-axis wobble” isn’t too noticeable except on larger parts. This hasn’t been the case with all of the machines delivered to customers so the factory switched to some Chinese made ballscrews which have much tighter tolerances. I’m yet to retrofit a set of these but I believe it would be an improvement from the point of view of accuracy and wear.
Improvements to the machine have been made since I purchased mine. Apart from the ballscrews in the Z axis, the factory now fits an energy chain that takes the wiring loom from the chassis to the print head to eliminate the fatigue issue and there is also a new controller board for the machine with the stepper drive components integrated. This would be a big improvement on all the discreet wiring in my machine. Also, at the suggestion of an early adopter in Europe, the factory increased the depth of the cut out in the top plate to allow the filament feeders to move around unhindered. Previously, when the print head moved to the rear of the machine past about 3/5ths the filament tubes would be squashed under the top plate.
Aside from the issues I’ve had with the print head, I have had practically no issues with the remainder of the machine. Anyone who buys a machine like this should be aware that they are an early adopter. In addition to this, the printing process has many complex interactions that can ruin a print. I always expected to engage in some tinkering with the machine.
To the manufacturers credit, they have addressed some of the shortcomings of the machine aimed at increasing reliability. They have also incorporated suggestions from owner community regarding improvements. From my perspective, I have found the factory responsive to enquiries and requests for assistance however I wouldn’t say every client has had the same experience based on some of the discussions on the web forum. I remain very impressed with the chassis and finishing of the machine, not many machines on the market look as good as the Creatr in my opinion. To the machines credit, it has operated for long periods of time without incident and it’s especially satisfying when you can turn out a large, warp-free part.
By Ben Roberts