Is the tiny E3D Revo Roto extruder really enough?
Vložit
- čas přidán 7. 06. 2024
- Thanks to PVCase for sponsoring this video! pvcase.com/ThomasSanladerer
Product links
E3D Revo Roto & Hemera:
from E3D directly go.toms3d.org/RevoRoto
from Matterhackers go.toms3d.org/E3DMatter
Bondtech LGX go.toms3d.org/LGXSlice
MicroSwiss NG go.toms3d.org/MicroMatter
Product links are affiliate links - I may earn a commission on qualifying purchases (at no extra cost to you)
🎥 All my video gear toms3d.org/my-gear
I use Epidemic Sound, sign up for a 30-day free trial here share.epidemicsound.com/MadeWi...
🎧 Check out the Meltzone Podcast (with CNC Kitchen)! / @themeltzone
👐 Enjoying the videos? Support my work on Patreon! / toms3dp
Timestamps:
0:00 Intro
1:08 Sponsor
1:44 Unboxing
2:50 Tech overview
10:37 Test setup
11:47 Initial results
13:50 Final results discussion
15:34 Caveats: 0.6mm HF nozzle
18:43 Conclusion - Věda a technologie
Update on the stepper driver theory: E3D reached out we're working on figuring out what the core issue is. I could really use some input from you - if you have 30 seconds, I'd appreciate if you'd take the time to answer four short questions on the community post here: czcams.com/users/postUgkx4YF7a5NP3arw3qAn-WiZ5PFIWksYeIVB
Kinda lame you kept the bad data in, we are here for a high end part and to know the best it can do, not to have your missing details about stealth chop be in the way of the data. Thanks for the video but its kinda pointless in this form sadly.
The stall of the roto is caused by the stepper driver: stealthchop only supports 300rpm. Use spreadcycle for up to 3000. (trinamic drivers)
Was going to comment the same,
ya I think there are some missing information about this test.
Yep, sound like a "User error: hit any user to continue" type of error.
This needs an update quickly
I also got a nema 14 as an extruder and i am able to push 50-60mm^3/s with a 0.4mm nozzle. Mabey a bit more current to. I am running 0.85a at 24v (watercooled) but 0.6a should be possible with aircooling
@@A_Guy_Building_Stuff50-60 mm³/s :ooo i only was able to get 38 on dragon uhf with cht and vulcano exension with sherpa mini.
what setup are you running?
Tom, your test rig might be at fault for that 0.6mm nozzle result. Check if the tmc2208 driver is in stealthchop mode. That mode caps around 300rpm and motor just stalls. You should set the driver to spread cycle mode for higher speeds. That tiny nema 14 motor needs to spin quite fast to get job done.
I keep the stealthchop threshold on 1. Gets rid of the stepper whine at idle and immediately turns off when the steppers move because the threshold is so low.
klipper recommends not doing that, i cant remember exactly why but you can probably find it online@@JrTanc
My guess on why the roto stalled is the stepper driver. It looks like you plugged in some fysetc tmc2208/tmc2209, those run in stealthchop by default, which puts a pretty hard rotational speed limit on the stepper. Doesn't help that the roto uses a 0,9° stepper in that case. Try some a4988 or swap to spreadcycle if you haven't done that. That removes the speed limit. My theory is backed up by the fact that the other extruders have roughly 400 steps per mm at 16 microsteps, the roto would be much closer to 1200 when taking a orbiter or sherpa mini as reference. Otherwise it doesn't make sense to stall that early as other small extruders can go much further than the roto, fot example it doesn't matter really if you are using a sherpa mini or a bmg, both have enough torque to loose traction
Thom, you need to issue a correction and retest the Roto. For any high speed stepper, such as the pancake stepper used by the Roto, you need to disable stealth chop on your stepper drivers. The time between steps will be too short, which due to a bug or design flaw, triggers a false over current fault in the stepper driver.
I encountered the same issue on my Orbiter V2 on my Ender 3. At high speed printing, or with linear advance enabled, the motor would stall at random. Disabling stealth chop fixed this issue.
He really should pull the video. It’s so wrong and hurts e3d
A month later, and the video is still up. Unreal.
for reference (according to my math), 6mm/s with a 1.75mm diameter filament is roughly 14.5mm^3/s. That makes about 180mm/s (theoretical) on a 0.4mm extrusion width, and 0.2mm layer height
which should be around 570mm/s at 0.4mm width and 0.2mm height if I didn't miscalculate.
@@radry100 570mm/s? I have calculated it to 181,25mm/s ... 😐
@@radry100 Should be around 180mm/s max speed. 14.4/(0.4*0.2)
@@radry100 According to Wolfram Alpha, 14.5mm^3/s / (0.4mm*0.2mm) = 181mm/s
@@radry100 under 200mm/s for sure
pleas use mm^3/s insted of mm/s. it's way easier to understand and compare to other setups
It's done on purpose as this hotend isn't really outputting.
Flowrate is more relevant to the heater. Motor power is more related to mm/s
Firmwares and slicers use both mm/s and mm³/s, but it's a simple conversion factor of x2.4 from mm/s to mm/s³.
I couldn’t disagree more. mm^3/s doesn’t make sense to me unfortunately. Where as mm/s does.
Even though volumetric flow rate (mm^3/s) relates to the hotend and feed rate (mm/s) relates to the extruder, it would’ve been helpful to have the context of mm^3/s.
Since the purpose of the extruder is to keep up with the hotend’s capabilities, it makes answering the question “does this extruder meet the requirements of your hotend?” much more clear.
We know the expected flow rate numbers for Revo, so it’s much easier to see where the drop-off/tradeoff of different extruders are. Using mm/s feels like introducing 2 unknown variables, since there isn’t much context provided for how that translates to printing.
It feels like measuring a car’s speed in Mach number, it’s not technically wrong and you can convert, but using mph is way more clear because most people have context for what those numbers mean.
Tbh, I love it. I've always thought that the extruder motors for most printers were massively oversized (though a common size). Your testing proved that point; the revo-roto's the only extruder that isn't massively overspec'd for a .4 nozzle. So for printers that don't plan on going over .4mm nozzle, this is probably the simplest, smallest extruder out there that might allow them to reduce the size of other drive components.
And it probably stalls because stepper motors have a hard speed limit; over a certain RPM they can't switch steps fast enough, and instead just oscillate between steps. Microstepping helps, but there is a limit.
What do you mean "the only extruder that isn't massively overspec'd"? What about CNC sherpa, VZhextrudort or one of the numerous other extruders that use motors of a similar form factor.
Thanks for all the time you take testing and comparing!
Really pleasing color choices on the views containing the luminescent sign. Thanks for the test results. great vid.
This is exactly the type of testing and demonstration I expect from manufacturers. E3d should 100% contract you to show these testings of their products. There is no reason to hide anything. Endorsements don't need to be done, you are a quality product tester whose services should be recognized.
What a effort! Great work to inform us thanks a lot!
I’ve had issues with stealthchop/interpolation causing motor stall on my 2208s (weirdly enough, not the 2209s) with 0.9deg motors.
I think I have heard that mentioned before. Would be worth testing. Oh, Especially at that RPM
It's most likely a design flaw in the 2208 drivers. If the minimum time between steps is too short when stealth chop is enabled, this can trigger a false over current fault.
@@earthwormjim Precisely. It’s definitely not a fan of pressure advance or rapid and repetitive accel/decel. The smaller 222X variant will *not* work with pressure advance in standalone mode.
I really enjoy your videos AND your choice of music!
Impressive test, well done
Great video as always 👍😀
Thanks for the great videos!
Great video. I really appreciate the scientific approach to testing out the hardware.
Really interesting extruder! I do agree that the glossy diecast surface of the aluminum is not really much to my liking either. Some feedback about the video:
While measuring the actual feedrate of the extruder in mm/s is valid and justified because that is literally the figure used in G-code, most of us never need to know it in practice. I think mm3/s or actual achievable printing speed in mm/s would be much more useful info.
It looks like E3D took inspiration from Orbiter extruder and then made it as small as they physically can.
This product has its use cases, where they will be excellent choice, but users must accept its limits. This combo will be great upgrade for bedslingers with flimsy Z axis (yes, Prusa Mini or Ender 3 V2). On fast CoreXY printers, it can be interesting mainly for its easy nozzle changes, but combo of Orbiter extruder and some high flow hotend (Mosquito Magnum) can do it's job better.
i have a nf-crazy with a orbiter and cht nozzle. the extruder motor stalls at around 42mm³.
Loved this.
Thank you for the scientific approach.
I am looking forward to putting this with what duet3d was demo-ing on my e3d toolchanger
Great test!
I would love to see you using your gear to test the Voron Clockwork 2 and Galileo 2!
I wonder if you could bump up the extrusion force by increasing the run current to the 0.85a recommended for the Orbiter V2. Granted, the Orbiter doesn't have an integrated hotend heatsink which means the motor is probably staying cooler than the same motor running in a Roto form factor, but something like the "turbiter" (an optional fan that sits on the outer face of the Orbiter NEMA 14 to provide cooling when printing in a heated chamber) might help with this. I can't find any information on what specific stepper motor the Roto uses, but the Orbiter V2 at recommended current has pretty exceptional extrusion force for its size/motor type. The creator of the Orbiter V2 has a blogpost that explains why, despite 0.85a seeming like a high run current and making the motor hot to the touch, stepper motors can function well at much higher temps than we normally use in 3D printing, and the limiting factor in the Orbiter is actually the melting/deformation point of the delrin gears. Maybe (assuming they use different steppers) you'd have to swap out the Roto motor with one more similar to the Orbiter V2 motor for best results, though. I'd be really interested to see some tests that look into this!
Or get a moons nema14 and you can run 0.4-0.5A and get way more torque, the LDO motor bundled with the orbiter v2 is really not great, needs way more current for its' performance than an equivalent(or even smaller) moons motor; the moons motor runs substantially cooler as well.
That is a reasonable flow rate for a unit that size. I like it. I was hoping for a load cell when you said sensors.
Thanks, very helpful!
Impressive work.
How does it compare with a sherpa micro? Would using a cht insert with a stock nozzle increase performance? Or the chc pro.
Did you use Stealthchop or Spreadcycle operation mode? Did you use uSteps interpolation?
I’m pretty happy with my phaetus rapido with the voron clockwork 2 large gear mod… it pushes 45mm3/s with ease on a 0.6 nozzle… stupid fast printing at an almost perfect quality too… happy to see e3d trying new things…
The other thing worth testing for is how effective the printed cooler is when running as a direct drive.
One danger is that while it might cool the heat break effectively, its blowing warm air around the gears too... I wondered if the pla would soften in the gear area causing a jam.
Could you also compare with Orbiter, Sherpa and other Nema 14 extruders? Yes, they lack heatsink, but still interesting.
Im super interested in that tiny filament pressure sensor
i'd be curious how much relatively lighter can one make print heads of common 3d printers by putting on this one, and to what degree will that improve print quality at high speeds, and whether input shaping capable hardware balances out this effect somewhat.
Would this hot-end extruder be light enough (at 140 gr) to make a Prusa Mini a direct drive printer? That could help with flexible filaments, right?
Also would love to know this
This was my first thought. I’m sure some brave soul will try it soon enough. If it works on the mini it’ll be an instant buy for me (I already have Revo on my Mini and would love to go direct drive!)
Are there any mounting STLs provided for common printers? I can't find any.
This is so exciting I can't wait to build a Voron 0
This is really tempting me to upgrade my prusa mini to direct drive - the only glaring issue is that going by the source code, I don't believe the mini has firmware adjustable motor currents. That could be a big issue; I don't fancy replacing a surface mount resistor.
Considering the issues with a high flow hotend, e3d could do a better job with the bundles they offer on their site. The only revo hotside sets available are the high flow and high temp ones - going down to a regular brass revo nozzle and 40W heater would save a fair bit, but the buyer then has to manually select the parts from the spares section (which makes it easy to forget something important like the revo spring).
I would really like a comparison video of a prusa mini direct drive conversion with that thing. Especially on flexible filament !
Looks like a slightly more compact version of the BIQU H2 system, I like it, Im a big fan of the H2 and V2, Ill have to give this one a shot.
Would love to see this compared to some of the other extruders that use the Nema14 round motor, e.g LGX Light, Orbiter, Clockwork 2 etc.
Would be nice to know if this is just an issue with the Roto or Nema14 extruders in general.
I've used the clockwork 2 and orbiter at up to 25- 30 mm per second with a rapido hot end. I can print my info at like 350 mm/s speed. Also used the rapido 2, same results.
What Was the Voltage for the motor? With Higher voltage it shuld be able to keep up with more speed.
I would like to see mm^3 insted of mm/s
Is there a sweet spot between size (and weight) of the hotend and it's motor size? The Hermera might run into problems with high speed because of it's mass, I imagine.
I wonder how it handles nozzle sizes below 0.4
Was just looking for a review of the smart orbiter v3 but could not find a qualified review. Would be great if you could consider to do this. Thank you to consider.
The choice of 0.9 degree stepper is an odd one, 0.9 comparison same for same have lower high end torque, with such high gearing, resolution isn't an issue, so why 0.9? Would be interesting to compare it with one of the popular bondtech geared extruder diy mods -- i.e sherpa mini and derivatives like the hextrudort. The size of the sherpa and the like is similarly small, but the gearing isn't as intense + use of a standard LDO 1.8 degree nema 14 round stepper would likely outperform this. You can see Vez pushing a lot more plastic than ~15mm^3 sec on the vzbot.
for higher torque on high rpm 7.5 deg motor from old printer or scanner can be used
If the stepper driver was limiting the motor in your review, I think we'd deserve an update video that more accurately represents the benefits and flaws of the product. Many people might not want to buy the product because they don't want or can't modify their stepper drivers, but that's a different message than "the motor can't handle that"
This would be perfect for a delta build if the volume on .6 could keep up... Would really like to see that resolved, maybe with the high flow or high temp hot end?
did you try turning the driver current up a bit? their recommendations seem quite low
any ETA on orbiter v3 hotend?
I have been able to get consistent extrusion of 100mm of ESUN PLA+ (at 225 degrees Celsius) filament at a feed rate of 645mm/min. 0.6 Revo HF nozzle, 60w heater core. Roto stepper running in SpreadCycle at 350Ma with default 0.9 chopper timing. (Marlin)
I wonder how it compares to an Orbiter? I've liked that particular extruder.
Always like how you just tell it.
Strange results, since the Sherpa mini uses the same motor, similar gearing (8.x:1) and current (0.35A) and it can reach well above 30 mm3/s
Yeah idk if the testing is flawed or e3ds implementation of the motor is flawed, but I've used orbiters and CW2 extruders at like 25mm/s easy
@@gotmilkbutt Flawed testing, stealth chop is still enabled in Thom's stepper driver. You can't use stealth chop at high speeds.
@@gotmilkbutt which, to make it clear, means about 50 mm³/s.
Great video as usual Tom, but please limit the use of autofocus in your video when you're zoomed in close to the items. I'm not particularly sensitive to these things usually but the constantly changing focus on the entire screen was very straining.
Im a bit disapointed that the e3d shop sells the extruder without the hot side parts which moves the price from reasonable to unappealing for me.
The Biqu (Bigtreetech) V2 Revo is a similar combo extruder and albeit slightly heavier (~200g vs 160g) and larger it does include E3Ds heater core and revo nozzle for noticably less.
How heavy is it with the nozzle and element fitted? Pretty essential factoid missing here.
What Voltage were you running the motors? I assume 24v , so I am wondering what it might be able to do with a higher voltage drive.
@@Papinak2 no . The problem with steppers is that the rise -time of the current pulses at high speeds becomes a substantial part of the pulse length , with out sufficient emf (volts) it limits the speed of the motor.This is why there are many newer high voltage drivers available.
I like compact, light print bead assemblies and, this almost delivers the goods after far to many '1kg' lump hammer heads of late. Just a shame they will now clearly have to add the concept of high flow extruders to the product lineup lexicon.
Mind, a slightly beefier motor than that one in the back, and alls good. Though, maybe it's time to start looking into using brushless DC motors (as used for RC cars) with some manner of closed loop feedback in extruders instead. They pack a lot more torque and quiet smooth operation into a smaller package. Then there is also 3 phase style stepper motors used in some places, they are pretty much identical to brushless DC motors, except optimised for low speed, high torque operation. I think somewhere in that mix is an Innovation waiting to happen :)
Hey Thomas. Are you sure you reached at least 20 mm/s or 48.1 mm^3/s using a Hemera with the 0.6mm HF nozzle on Revo? That seems way too high for that hotend. It can do a lot for sure, but that's UHF territory.
Does the motor overheat if you step the current up a bit? I know I think its a company called moon or moons makes a high temp 14nm pancake stepper just like that you can take up to around 0.8 and get a lot more torque out of it. I use one on a direct drive conversion for my super racer that works really well. Just kind of curious if its possible to get more out of this thing.
The motor isn't the limit, there's plenty of torque with the gear ratio. Thom has stealth chop enabled, which needs to be disabled to operate higher speed motors, such as the pancake stepper used by the Roto. The stepper driver is halting, if the time between steps is too short with stealth chop enabled, this causes a false over current fault.
Seems like a somewhat niche product for those that really need the small size but don't have very demanding performance requirements.
looks great to me - I just use .4 anyway... given the light weight, might be able to increase print speeds?
Yes, to some degree. But this combo will get into "low melting performance" bottleneck relatively soon.
If you are printing with 0.4mm nozzle and 0.2mm layers, it will keep up with speed roughly up to 250mm/s.
It will allow you to use higher acceleration and jerk values, while keeping the same print quality, because it is light and printer will need lower forces to move it around.
Why don't you compare it against a Vz-HextrudORT that uses the same nema 14 motor? 🤔
What about TPU? The LGX Lite has a lower tension position for TPU. The Revo Roto appears to only have one tension.
I don't know about the Lite version, but the regular LGX has a non-sprung idler lever, meaning you're not actually setting *tension* of the lever, but instead you're directly adjusting how far the filament gets pinched. Spring-loaded extruders should do better when it comes to handling a wide range of filament harnesses, but may have a lower level of ultimate grip.
Great review. Now if someone tests if this fits Creality K1 Max with the lid on.
looking at extruders like the sherpa mini / hextrudort / orbiter .. they all use the same small motor but are able to print high flow >30mm³/s
Do these mount up to Creality printers?
I think there's something odd, normally nema14 don't have a lot of torque, but can spin fast. Sherpa mini is a beast and even smaller, and folded ascender is even more powerful despite a 20:1 reduction
In addition to comparing size, maybe would be nice to compare weights (mass) ?
put up a comparison of one of those vs a v2 extruder.
It's not the size of the motor that's the issue, CW2 extruder uses a round nema14 motor with different specs and I can get 35mm/s3 out of my bozzle nozzle.
Nice video! what I got out of the video is that I use .6mm HF REVO so it is not going to work for me.
hast du echt den sidewinder zerschnippelt um den test extruder galgen zu bauen ? :D
Massive testing job...thank you!
Im usually always a fan of getting more sensors to make sure stuff work but ive never had hotend issues because of that specific issue and outside of heated chamber with PLA i cant thibk of any nirmal user needing it
Also yes, please test with flexible filaments, and let us know the limits. And can it print nylon, etc.?
It's only rated for 40C so even printing ABS enclosed is probably going to be pushing the limits.
@@user-ee2oq3ek3uYou're right, I just checked the specs, very good catch! It should have been mentioned in the video.
Damn I want that lgx hothead so bad
This would be good if you’re going to be down in nozzle size from 0.4 but not up. Great extender for things like DnD figures or small detailed functional parts, but it’s not for me. I really need the high flow capability of the Hemera.
Can this go on a voron
That finish looks like the acid etched parts we do out of our machines.
Didn't even finish the video but I already want to put that thing on my Voron 0.2
Me too until he got to the part about achievable feed/flow rates 😢
@@Brunoku are you printing at more than 200mm/s realistically? if not then it will be fine.
@@thegarageluthier Nope, but the combination of my line width/height/speed puts my flow above 20mm/s^3 pretty frequently for quick prototype prints
So if I am interpreting this combination of weight and flow capability correctly, this seems positioned for fast but fine layer heights. Would love to see tests getting this thing flying around a print but pushing only fine and super fine layer heights. .....which actually would be a "killer app" of this product......at least for me...... fine but fast layers.
10:04 wee little boi KEKW
Who's making the round stepper motor for them?, I have the LDO round stepper which is rated at 1 amp running at 700mA 24/7 on a Hevort/vz-bot type extruder with BMG gears printed in petg and the motor doesn't even get hot, looking at their documentation it says recommend current of 0.45A peak and max of 0.6A, seeems like they have used a lower current motor that's not up to much, why did they spec that motor for the roto extruder, maybe they could try the LDO or Moons motors that are rated higher?, don't know about the stealthchop issue that's mention below as I'm running a duet ethernet controller with TMC2660 which don't support stealthchop.......
That Plastic release on the filament tension level WILL break, I have NEVER had anything with that kind of Plastic Hinge/spring EVER last and that was at ROOM TEMP after enough heating and cooling cycle that thing WILL get brittle
If the motor starts skipping steps due to high frequency.. I wonder, if a motor with less steps per revolution would provide better results. With 10:1 ratio, it'd be precise enough even with 2-3° per step.
head to head with the smart orbiter?
You should definitly test more extruders like the clockwork 1,2 orbiter ect.
kewl test rig
This is gonna sound kind of obvious but I love when printer/accessory manufacturers use printed parts in their designs
Oof the flow rate is so disappointing. Usually you go super light to print fast but if you can’t print fast then what’s the point? My mk3 regularly is flow limited 17mm^2/s with my CHT
Have you ever thought to create a custom extruder? I think you have everything needed to start such a project.
00:18 the hand of a giant :)
Kleiner Motor....war abzusehen. trotzdem schnuckelig. Wo gibt es den Hoodie zu kaufen?
What voltage is the PSU on the test jig? 12V or 24V?
Running a stepper motor at higher RPM requires a proportionally higher drive voltage to produce a high enough di/dt in the windings.
If you're using a 12V power supply in the jig, you'll probably find it'll extrude much better through the 0.6mm nozzle if you try a 24V supply.
And if you're using a 24V power supply, you'll probably find the Roto will run like garbage at 12V - which would be interesting information for people considering fitting one on a 12V printer.
Another possibility is the super high step rate might be hitting a limit of the driver IC or 3D printer firmware that you're running - try halving the microstepping value of the driver and see if that makes a difference.
Cheers!
24V for this one - at this point, is it fair to consider 12V setups "legacy"?
@@MadeWithLayers Definitely fair, but there's lots of 12V printers still out there - Prusa MK2/2.5 for example - that people might try putting one of these extruders onto and run into problems as a result.
How about slapping a 12V power supply on your jig and repeating some tests to see how bad it gets? Might be useful information for people.
Why don't you compare it with hemera XS ?
why in the world is it mm/s instead of the mm3 that we are used to?
This might be a valid option for a delta-printer.
The word "roto" means "broken" in Spanish. It may not be the best name for Spanish-speaking countries. On the other hand, E3D is a great brand. I have several machines with hemera and I am very happy with this extruder.
Why would I want to have such a light extruder, when don't want to print very fast?