Those flat spots at the beginning of the torque-speed curves are due to the current limit on the drivers. The torque is proportional to the current. Once the motor speeds up there is less time to build up current before switching direction of current flow which is why the current drops at higher speeds as well as torque. This is why higher voltage systems are beneficial because they can build up the same amount of current in less time. This allows for a longer flat spot in the torque-speed curves
@willemstigter6384 exactly! I am building a system right now where I have already pushed 48V to 60V and still did not achieve the required acceleration. Rather hit saturation. Am using Elmo Application Studio and moving to an Ingenia EtherCAT driver now. Message me if you’re interested to help.
@engineerbo yep. I have always wanted to build a dyno myself. Im very impressed with yours. And your video is top notch. Cant wait to see the next one man. Thanks again for sharing all the results. (Ps: the kraken LDO might be interesting to test too 😀)
@@Vez3Dcoming from a decade of ebike design , I can see a trap with this thinking perhaps. I mean no disrespect because I love your ideas, I simply want to say I have seen the “more volts” trap before. What I am wondering is if we are in the wrong speed zone for 500mm per sec, I think it says we are in the wrong gear ratio. That’s my mechanical engineering view anyway. How can we configure the belts for a 2:1 ratio off the A/B motors? Or even 3:1. Losses aside it would seem the better step givem the other issues with higher voltages. Cheers.
@@samdekok72double edged sword at times -if I understand correctly 2:1 and 3:1 ratio for ab motors so instead of 20t stepper pulley you’ll be running a 40t or 60t pulley? If so then that might introduce stepper resonance/vibration vfa’s like seen on the early creality k1’s. The micro vibrations of overshooting each step and vibration/resonance of bouncing back to re-centre sometimes translates to stepper induced vfa’s. Afaik it’s usually more common/evident in higher torque and larger steppers where the holding torque forces are high on the rotor and higher rotor inertia results in more overshooting and a stronger pull back to micro step position.
Great video and dyno. The TMC2209 can blow up because you have no bulk capacitance on your test board. You could also add a 24V TVS in parallel as a safety mechanism. As other commenter mentioned, adding inductance to the spec table would give us a hint as to the high speed performance. LDO-42STH48-2504AC = 1.5mH LDO-42STH48-2804AC = 0.6mH Stepperonline = 2.4mH usongshine = 3.2mH jk42hs48-1684-08af = 2.8mH Its a trade off, so if you want speed you just choose a low inductance model. Doesn't need to be some fancy 3d printer brand. That and always use 24V or even 48V if possible.
Yes I was suspecting voltage spikes again, but I didn't manage to verify in time. One of my upcoming projects would be to design a more user friendly protection device to properly solve this problem.
Yeap. low inductance can eat more current and gives you higher speeds. I would also like to see more expensive stuff, like Nanotec. To actually see if price matters.
Nice video and results! Your TMC2209 drivers probably died because you weren’t using a capacitor between your 12V input line and ground before your driver. I used a 220uF capacitor without I single driver dying since.
Wow! Very in depth testing. Thank you for your service to the community. Look forward to more motor testing and would love to see some fan testing as well!
Great video! I'd just warn you, when using TMC5160 drivers in the future, to not use them above 24 V if using it in the stepstick form factor. The design pioneered by Watterott and copied by other manufacturers have a design flaw that makes them burn randomly, as many Voron owners that run their stepper motors at 48 V have discovered. Some manufacturers, like Bigtreetech and Mellow, have designed a stand alone board using this driver that correct these flaws, but they need to be connected to 3d printer boards using adapter cables. Also, I'd like to suggest including the OMC 17HS19-2504S-H stepper motor in the test, they are also rated at 2.5 A and are said to be similar or slightly better than the LDO Speedy Power ones. The only drawback is that this motor is only available in the version with wires coming out directly from its body, instead of having a connector as all the other motors tested.
Interesting, what is the design flaw? A couple other comments mention the lack of capacitance (presumably near the driver), which probably means voltage spikes killed the drivers.
@@engineerbo On page 16 of the TMC5160 datasheet there's a section specifically saying that you should not provide more than 40V on the VSA pin, it's the pin that supplies voltage for the internal 5V and 11.5V regulators. On common stepstick designs VSA is tied to VMOT, which will supply whatever voltage you are giving to the motors. So using them with 48V is basically silicon lottery. That's the flaw the standalone 5160 boards solves. Don't get me wrong, the capacitor is also needed, a good rule of thumb is 100µF for every amp of current that's given to the motor. For these ones you tested a 330µF electrolytic capacitor of 50V or more should be enough for all test cases.
@vinnycordeiro I just found some time to have a look at the TMC5160 datasheet. On page 16, the note I think you're referring to doesn't actually say not to connect VSA to >40V. Instead, it's simply recommending VSA < 40V if certain conditions are met e.g. MOSFET total gate charge > 50nC. This note is related to the text preceding it (Chapter 3.2), which talks about high power dissipation of the internal linear voltage regulators at high input voltages. So the designs by BTT etc are actually fine, at least in this regard.
@@engineerbo I oversimplified my answer but yes, you are right. It's just easier to stay under 40V than having to double check the MOSFET choice of manufacturers. BTT did that on their TMC5160T Plus V1.0 board, connecting VSA pin to 12V directly instead of VMOT.
Oooh... I know you will get lots of requests to add motors or models, or tests, but I like to think I have a decent reason to ask for another motor to be tested in a similar fashion: the Moons motor that comes in (for example) the Formbot kits, and is available on the Biqu site. That is also a pretty popular and common motor 'in the field' and I would love to know how it stacks up in actual 'accurate tests' against the LDO motors. I've got both on printers, but those are quite different in setup so I find it hard to compare the two. And I wonder if I should swap out the Moons for the LDOs, or vice versa (or simply stick with what I have since they are roughly comparable...).
@@engineerbo The ones in my kit were MS17HD6P420I-04, I know there's a -05 as well, which is a new iteration of the motor and ought to be roughly comparable. Like the LDO ones, they are relatively 'premium' in price-class.
Great results! I'd love to see some "bottom of the barrel generic" motors tested, like the ones you get in GRBL combo kits, just to see how much performance you can get by investing in higher quality motors.
That's an interesting idea! When shopping for these motors, I realised I couldn't find ultra cheap motors like I'd thought I would. Maybe random kits might be a good place to look.
Cheap doesn't really matter for performance, its a tradeoff what you want, you can get a cheap low inductance high speed motor if you want. But the reliability and accuracy might be junk.
I had bought the super power motors for a cnc router. They were absolutely awful for that application. I swapped them with the motors in my 3d printer, and now the printer is faster and the cnc works. I think they went with a heavier guage motor wire with fewer windings in the super powers. This also lowers the inductance, so you get faster current changes. By putting them in the cnc, I was effectively trying to tow a fifth wheel with a corvette.
Yes it makes sense, if you're using the motors in heavy load applications like CNC, you probably want higher torque at lower speeds, and the super power and speedy power motors aren't the best in that regard.
Nice tests! You can try DM556, DM542 branded and unbranded drivers for comparison. And higher voltages too 36V and 48V. May be you can try higher Amps then specified by manufacturers and measure the temperatures, i regulary use steppers on higher amperage without problems.
Thank you! It would be interesting to see if any of the other drivers would be able to compete with Trinamic's drivers without simply shoving more current into the motor.
Good test. You will not be disappointed with the 5160. I used 2208 (Creative board), 2130 and 5160 (BTT OCTOPUS) on an Ender 5 PRO and a 5 PLUS modified with linear rail and with StepperOnline like the ones in your test. I have not measured the torque but the reliable maximum speed results are visible: if 2208 are 100%, 2130 are 150% (bit of whine at low speed) and 5160 are 170%. Now with all 5160 the loudest things are the PSU and cooling fans.
Thanks for the link! What changes are you making while observing the phase voltages to tune your steppers? It seems to me that if we can rely on the stepper drivers to do their thing, the conclusion is that higher voltages are better (to a reasonable degree).
Love it, finally somebody going deeper into steppers. I'm very deep into cheap servo motors - this territory might be even crazier! But, 14:22 your whole setup is flexing.. not good for reliability of data :(
Thank you! The bracket design is not ideal (I've designed it to make it easier to switch motors out for testing), but I don't think the torque readings will necessarily be unreliable. If the motor is mounted on a flexible, springy bracket, the bracket will flex/deflect depending on the torque applied by the motor. The higher the torque, the greater the deflection. Yet the torque deflecting the bracket is still going to be equal the torque applied to the brake. This is going to be the same for a stiffer material. In reality, there might be a slight power loss by the motor flexing and "de-flexing" the bracket, but I think it's not a major problem.
IMO the most interesting results are "hidden" in speed and torque curves - how much (answer: yes, torque varies pretty much between full steps :) ) speed varies when micro stepping. In 3D printers we can see VFA at lower speeds (e.g. Prusa MK3, any Ender 3 clone), more strongly when the printer is designed for high speed (e.g. Creality K1). They can be fixed in various ways (0.9 degree steppers on MK4, smaller steppers and pullies on K1C). Motor design (smooth stepping) and matching torque with moving mass (less torque and micro step judder with more momentum at higher speeds) and belts (spring) might lead to better print quality. Driver current may also affect VFA (not sure if it is tested) and might need to be dynamically adjusted for speed and acceleration of movement? Silencing stepper noise on Bambu Lab printers and Prusa XL probably also provides smoother surfaces with less VFA...
Good video, really hoped to see the nema17 motors i personally use in this test. I use the steppermotor 2,1 A, 0,65nm motor. Wouldve been the highest holding force in this test while being not the highest Amp. The serial number is 17HS24-2104S - i am super happy with these motors and i can just recommend them.
I have lots of different tests planned, but not one for 60mm motors (I don't have any!). But if there's a large enough demand for this test, I can arrange for it.
This is a great video. I possess all of those motors and have played around with them not to the graphing extent that you have however. Speedy powers didn't last long for me because I'm dyslexic and I accidentally put the decimal point in the wrong area and fried both my speedy powers pretty much instantly. My superpowers however are very nice and they are connected to 5160 pro external Big tree Tech drivers . I was previously using the TV 6600 drivers and I experienced the same result
I also blew up a few btt 2209 drivers on my 24v printer, one of the capacitors failed. i think it was c6 next to the "bottom" label. i soldered a through-hole 100nf in that place and the drivers worked again.
The dyno design is great, but running stepper drivers on a solderless breadboard is just asking for problems and likely the cause of your 24v failures. When you are switching relatively high currents at high speed you need low impedance supply rails with plenty of decoupling, something that a breadboard just cannot provide. Why not use an existing stepper driver board like the SKR?
It was the first build of the dyno, the breadboard certainly isn't the best. I'll be building a PCB to replace it for the next video. I needed custom functionality for the dyno that perhaps something like an SKR could do, but it seemed more straightforward to implement it myself.
You need also to get LRC bridge so you can measure inductance values of the motors. Speed is really affected by inductance as if motor got high inductance voltage and respectively the current will not be able to go through the coil before it switches. So you put 1.77a but maximum current that goes trough is 1a. This can be changed with Voltage while increasing the voltage you mitigate this behavior as you saw with acient driver TB6600 which is terrible despite worse control Voltage was able to mitigate the effect of the inductace at the cost of thermal dissipation. You should start account for inductance bigger inductance bigger holding torque.
Yes higher voltages should also be able to overcome the back-EMF better, and definitely has an effect on motor performance. I'm also thinking of monitoring the phase currents, but perhaps adding more current sense resistors will be too invasive.
Thank you! The brake calliper is mounted on an arm that pivots around the same axis as the braking disc. The other side of the arm is attached to the load cell.
Great analysis! Love your dynamometer!! YES please microstepping vs torque! And I would love to see TMC5160 compared to an established closed loop stepper with optical encoder, e.g. Stepperonline CL57T driver.
Nice setup. It would be interesting to see the actual current waveforms of the different drivers. You have to keep in mind that when comparing smoothness of operation and noise, newer drivers are basically cheating because they don't really do steps anymore, they treat your step input more as a suggestion and generate a sort of sinusoidal waveform, trying to guess what you're doing and what you need.
@@engineerbo I'm in on a group buy, but otherwise they somehow seem to be popular in India, and 3DPrintronics has it and has reasonable international shipping rates. 2 motors looked like 48 shipped for me.
Hi. Are you still working on the next video? Would be great to see the performance of these steppers on tmc2240 at 24 volts, as its the common 3d printer standard nowadays. Great work btw. Subscribed
You did a lot of valuable work here! I had actually contemplated getting IGUS steppers because they were some of the few who published Torque/speed graphs on their website. Though curiously their published numbers mark the motors as being a lot more powerful than even the LDO stuff in your test. maybe you could test one from them as well to see how the numbers hold up on your dyno?
I looked it up, and the drylin E stepper motors don't seem to have that much higher rated torques and are pretty pricey. Are these the same ones you're referring to?
@@engineerbo depends. they have many similar products, including very pricey "industry grade" stuff. I'll try to post a link, hopefully YT will let me do it.
@@engineerbo okay apparently that didn't work. You gotta try to sort by price on their website, and just pick the cheapest one. Also to clarify: The rated holding torque isn't higher, but at least according to the first party graph they retain much more torque at higher speeds. (0.3 NM at 700 RPM / 24V)
Top work, this is really useful testing! Thanks for putting it together. Looking forward to seeing the tmc5160 data. It would also have been interesting to see how your torque speed curves compare to the ones in the datasheets for each of these motors (stepperonline have t/s curves online for their motors).
I have had success using TMC5160 with Nema 23 motors for driving a peristaltic pump (pulsating load). I used spread cycle mode to get the best results in terms of limiting heat and noise for speeds from 0.1RPM to 500RPM. However, to accomplish this it is necessary to send a new configuration datagram based on selected speed.
Hey! Few questions: Have you calibrated the dyno? Kinda afraid of belt drive efficiency changes as speed and torque change. Have you tried applying the braking torque slower? I noticed on your scatter plots that you don't have many data points near the peak, which could indicate inertia of the system causing an error, as you're not only trying to combat the torque of the stepper, but also torque from the deceleration of the system. You can also run a higher resolution encoder, as torque of the stepper is proportional to the magnetic misalignment angle within the stepper (full torque at 1 full step deflection, it follows a sine curve). A very popular method of testing the torque to speed curve of a stepper is to set and hold constant braking torque and increase the speed gradually (so the inertial effects are minimal) till you observe skipping steps. Can't wait to see further developments, good luck
Yes these are all good points that I've thought about while building and programming the dyno. The brake is being applied extremely slowly, which causes the test to take ages. There is a little stiction in the brake line that causes the brake to tighten less smoothly than I'd like, so there's still room for improvement. I also considered testing by spinning the motor up with a fixed braking force, but in the off chance the torque output really does increase with speed, this testing method might not work (probably).
Yes! The effect of microstepping on performance might be heavily dependent on the driver, but it'll be pretty useful to know the tradeoffs with microstepping if there are any.
@@engineerbo Also the effect of all the different modes and features of tmc5160. like stealthchop, spreadcycle, coolstep, SixPoint RAMP. How does changing the sense resistor affect the outcome for smaller motors like nema17? I would also recommend using one of the "pro" version from mks or btt of the tmc5160. They have much bigger FETs, can be used in standalone mode by jumper configuration and you can more easily swap the sense resistor @engineerbo
What is standalone mode? The units I purchased seem to support step/direction mode, but it isn't clear so far how the current limit is set (without using SPI). I'll be reading the datasheet properly soon, so I suppose I'll find out.
@@engineerbo i havent looked into it in younger future as i also just checked the manual to make sure they are not in standalone mode. But IIRC you can set amps and micro steps through the jumper pins
@@engineerbo On TMC jargon, standalone mode is the one that makes them compatible with older drivers like the A4988. That's because TMC can be dynamically configured through UART and/or SPI, depending on the model.
Thank you! After doing all these tests, I found some areas where the dyno needs to be improved. If there's enough interest, I don't mind cleaning up and releasing the code.
Interesting results. Would be nice to run some resonance test for each of the motors. I have StepperOnline motors on several of my builds and they perform pretty well. In my Voron 0.2 build StepperOnline resonate less than LDO motors which results in better prints quality over all.
@@engineerbo I think you could just add simple ADXL sensor to your rp2040 setup, place sensor on the motor itself and run loop changing motor frequency at the same time recording readings from the ADXL, at least I would start from there. On 3Dprinter setup it's much easier as Klipper helps to run such task semi-automatically.
My first guess would be that every motor has its own resonance frequency, and there wouldn't be "more" or "less" resonance, so I'm not sure how I'd design a test for this.
@@engineerbo I would assume a well-made and balanced motor should have fewer frequencies it resonates at. At the same time, I think the higher the vibration level and the wider the spectrum, the harder it is for the algorithm to mitigate those vibrations, and it will end up affecting the print quality. I'd probably make two tests: the first one to see the overall noise and the second to check how high the peaks are getting.
Subbed, awesome content! I would be really interested in testing the 5160tpro with the same motor lineup. Things you could include would be how microstepping affects the curves aswell as input voltage. :)
I noticed that some TB6600 driver modules have a circuit connected to the TQ pin of TB6600 chip that chopps the current limit to 30% of its setting, during half step period. May be that is affecting your measurements; i mean, instead of its "brains". You can check by probing pin 3, whether it stays high (5v) or not.
Simple way to select a "fast" stepper is to look at the inductance (often listed) as it's the thing that will, during running, limit the current rampup and thus also torque.
@engineerbo the cm06 graph looks too good for what it is and I'm not sure what the cm08 offers over it. Should be interesting to see! On large printers I we usually run at lower rpm and higher torque (vcore 3/4) so I'm excited.
Might be worth plotting `speed * torque` versus speed. (This is just 'output power' versus speed, give or take a constant factor.) It results in a much flatter graph, which tends to be much easier to read. Also, you can get stall torque with that setup. Just lock the brake on full, then drive the stepper.
You might be right about the stall torque. The reason I've added belts to the dynamometer is so it has a little compliance, which makes it possible to measure the torque when the motor is holding its position. But I don't think it'll actually be able to reach the actual holding torque, because the motor still needs to move, even if just a little bit, to move through its torque range.
@@engineerbo Ah. I thought you meant stall torque not hold torque. Yeah, hold torque is trickier, especially if you want it to test in the same setup. One approach that can work is to replace the brake with a larger motor with a decent controller. In practice, especially if you have a bit of gear/pulley ratio between the two. Still do the actual measurement with the load cell, but replace the gradual application of brake with the gradual application of PWM (instead of open-circuit). You might want a braking resistor, depending.
Yes using an external source of torque would solve the problem. But I'm not sure the actual holding torque is that critical to know, since an extremely slow speed gets me a number somewhat close enough.
@@engineerbo Agreed, at least for this application. Pull-in torque can be substantially different than pull-out torque even at zero speed, but I don't think you care in this case. One other thing that might be nice to measure - although is somewhat terrible to measure to be fair - is the stepper motor resonant frequency and behavior around said resonant frequency. Note this depends on the inertia of your load! Doing a slide test at (near) your resonant frequency is somewhat of an interesting worst case. (Essentially: a stepper motor 'snaps' to the next pole, especially when not microstepping. But in practice it'll oscillate around said next pole somewhat before settling down. Interesting things can happen when you do another step at the quarter/half/three-quarter/whole period of this oscillation.)
It would be interesting to have a Power vs Speed curve. As torque doesn't matter so much and can be "increased" by using a different mechanical setup (smaller pulley...). As well as static torque, dynamic torque only matters at a given speed...
Power is torque × angular velocity, so actually the power curve would give you the same information, just in a different form. If you use mechanical advantage to increase torque, the tradeoff is speed. If you assume a perfectly efficient system, the speed-torque curve's axes are simply scaled up/down, but will keep its shape.
I found the "TB6600" (no such chip on them in the 1/32 microstep version, it's a TB67S109AFTG or similar) drivers to be really nice and reliable with the big heatsink, isolated inputs etc. Those tiny breakout boards can be a pain in the ass, they can get quite hot, can be really sensitive to noise and randomly die on you, just like this one.
Very nice and detailed video but what makes me curious what about accuracy and repeatability of those motors or maybe it’s not determined by the motors it self I actually don’t know so please let me know/ us know . Keep it up 👍
Great little dynamometer! Have you considered using a braking magnetic field instead of the clamping friction load? Just thinking you’d have a more consistent load since it must be vary hard to keep the variable of heat equal throughout each test. Also, a lot of the magic in stepper motors is done with a lot more microstepping than was demonstrated in this video. Can you provide the code and or lookup tables used?
Using a magnetic brake was actually my first thought. Essentially using another motor as an adjustable brake, which means there's almost no wear and tear etc. But a passive one doesn't work at 0 speed, so it can't brake the motor to a complete stall. It's also a lot more complex.
As a Materials Science Engineer, I highly appreciate this video; very good job and thanks for sharing! Just a quick clarification: when you mention the 750 RPM for the 500mm/s on the bambulab using a 20 tooth pulley, were you assuming 8 microsteps or 16? Also, why did you went with 8 microsteps instead of 16 for your tests? Tks
Microstepping isn't included in the calculations, because I was using RPM instead of steps/minute. There's no particular reason for 8 microsteps, besides it being the "default" on the TMC2209.
Nice video! Is it possible to test motor noise for VFA generation video? I am of the mind to use a motor with the least VFA and sacrifice torque for a smooth curve! Do they exist!?!
It’s unclear to me how the load cell is measuring the braking force. I would greatly appreciate if you could elaborate on that please. I love the dynamometer setup!
Thank you! The brake calliper is mounted onto an arm that pivots about the same axis as the disc, and the other end of the arm is attached to the load cell. This makes the measured force "downwards".
It’s a good dyno but the use of a padded brake and brake disc is a factor that can lead to bad numbers in the long run due to nature of wear. Maybe using centrifugal force and fixed weight would have less outside influences.
Thanks! The condition of the brake actually shouldn't matter at all, because I'm measuring the actual braking torque. For example, if the test was running, and I decided to oil the braking disc midway, this would cause the braking torque to decrease, and this change will be measured by the load cell. Oiling the disc would only mean the test would take longer, because the brake calliper would need to clamp harder to achieve the same braking torque.
Awesome video! I run the AH (high temp) version of the speedy power motors and they are brilliant. In my printers I’m using normal TMC5160’s @ 24v. However, in my little CNC machine, I’m running external TMC5160’s from MKS @ 36v. The results are amazing, I love these motors! 😂 PS, high temp means they are rated up to 180c according to the datasheet.
Thank you! I purchased a bunch of the smaller TMC5160 modules since they're rated to around 3A which is enough for these NEMA17 motors. Are you using bigger steppers with your external drivers?
@@engineerbo I’m running the LDO 42STH48-2504AH on my little CNC machine too. It started life as a 3018, so there isn’t room to use Nema 23’s in its current config.
@@engineerbo At the moment I only need 2.5 Amps, but there was an element of future proofing behind their use - if I ever decided to buy Nema 23's. TBH, I probably wont as they just arent needed on my little 300x300mm (work Area) machine. I also knew from research that many of the non-tmc external drivers were not very good. On a side note, I found it amusing that you used an A4988 on your brake - I found that I could get more torque and speed out them, than I could with TMC2109's driving the LDO's. They can actually put out 2A peak, unlike the TMC2209's - but they make a lot of noise doing it. Btw, I got mixed up, I have TMC5160's on my Printers, and external TMC2160's on my CNC machine - that latter is good for up to 4 amps and easily configurable via dip-swtiches.
I'm not familiar with the requirements for a star tracker, but with a "standard" 1.8° stepper motor without gears, 0.0007RPM corresponds to 1 full step every ~7 minutes. If you used 256 microsteps, that's 1 microstep every ~1.7 seconds which sounds better, but I think there will be accuracy/torque trade-offs. You could also consider 0.9° steppers, or steppers with gearboxes (e.g. cycloidal if you need low backlash).
Nice setup ! 😃 The TB6600 drive you are using is not a TB6600, it has a Toshiba TB67S109 chip that is better than 6600 but less current capability. The same chip can be found on small step sticks, also.
I'm very confused because speed-torque curves aren't supposed to look like this, there should be a flat torque output then a drop when the back EMF restrains the current that can be pushed though the windings
Yes we should see a flat top at low rotor speeds, corresponding with the driver's current limit. Perhaps we'll see this if the tests are repeated at higher voltages.
The data points are binned into the speed steps, since actual noisy measured speeds are used. The peak at each speed step is the highest value in the bin.
I’d have done a few different tests too. Like, with the same stepper, see how the curve changes as a function of microstepping. And maybe to compare the motors at the same current rating. More importantly though, you didn’t plot the mechanical power alongside the torque-speed curve!
Good test. Note the curve for LDO motors (at 13:45) is not marked speed-torque, but torque over frequency (hz). Would this not illustrate a claimed property for step frequency, rather than RPM?
Yes I too interpreted it as step frequency, which is the same as angular speed assuming the motor isn't stalled. E.g. for this motor, 200 (full) steps per minute is 1 RPM.
The dynamometer controller is controlling everything, including step generation, so it knows what speed the motor should be moving at and when it stalls.
I recently had a problem involving this exact setup: a (generic noname) NEMA17 stepper, a TMC2209 driver (used with factory defaults as dir/step only drive) at 1/8 microstep and a 12V supply - low speed torque seemed ok, but the motor literally stalled around 600RPM or so, with no load, no matter what I did. Tried a beefier PSU, tried maxing the current pot as far as I dared, tried different microstep, nothing helped - even with gradually trying to raise the speed, the motor just started shaking and stalled. Really weird, and extremely disappointing - 600RPM a.k.a. a measly ten turns per second is NOT all that fast...
I blew up the TMC2209s at 24V (mentioned in the video). It could be a problem with not having a capacitor near the drivers. This will be fixed for the next tests.
You should try normal stepper driver, like TMC5160 (for example Btt tmc5160 Plus), and try the 24V and 48 Voltage Your TB6600 seams to be a shit driver, that's drastically limits potential of those motors I can easily achieve 1m/s on my 500x500mm bed size printer, using LDO-42STH60-2804AC-R on 48V, that's 5000 RPM on 12mm pulley I guess! 800mm/s with 24 Voltage And its moving quite heavy aluminium profile on X axis, and toolhead assembled from two INOX plates Good video, Im waiting for more ❤
I'm a bit sad you didn't include a quick noise comparison. Put them all on the 2209 with the same microsteps at the same rpm and see which is quietest.
@@engineerbo not as much difference as the drivers, but still a considerable difference at increasing rpm. imo driver is really reducing noise at low rpm, while motor build quality reduces noise at higher rpms. just very limited experience tho.
I haven't really considered the noise factor much outside of the smoothness of the driver. I'd have guessed the higher RPM noise could be caused by bearings etc. I'll take a measurement if I notice the difference.
I'm confused why you didn't get any hybrid steppers on your lineup?! They put normal steppers to shame in pretty much every single category, except one - price. They're the ones in one piece aluminum round cases, instead of laminated steel sheet bodies. They are insanely powerful compared to a cheap laminated steel stepper
@@engineerbo Hmm, I know them just as 'hybrid steppers'. They are just normal 4-wire steppers, but they are in round aluminum cans, I got a box of them in an auction years ago (they were buried in a pallet), and I've used them for various projects over the years. The first picture for google images "EHY21512" or "23HY4604" shows the TYPE of stepper I'm talking about. The cases are round aluminum cans, not laminated steel sheets
I'm trying to figure out why you can't measure torque at zero speed. I'd think the load cell on the dyno wouldn't be the problem. Is the problem with needing to provide steps to the driver? Any insight would be appreciated. I really enjoyed the video. Thanks for all the hard work. I subscribed.
Thanks for subscribing! To measure holding torque, the motor is usually made to hold it's position, and increasing torque is applied to it (from an external source, e.g. weights) until it position holding fails. With my dynamometer, the torque comes from the stepper motor itself, so strictly speaking, it cannot be at zero speed and increasing the torque at the same time.
Man, a leadshine dm542s or something close, instead of a tb6600. They burn, make a lot of vibrations, etc etc etc, truly the worst. I did that when I built a mpcnc years ago.
![I.N "HALLUCINATION" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/n5B5q1Hwt_U/mqdefault.jpg)
Those flat spots at the beginning of the torque-speed curves are due to the current limit on the drivers.
The torque is proportional to the current.
Once the motor speeds up there is less time to build up current before switching direction of current flow which is why the current drops at higher speeds as well as torque.
This is why higher voltage systems are beneficial because they can build up the same amount of current in less time.
This allows for a longer flat spot in the torque-speed curves
Correct! That's why Im using 48V on my printers
That makes sense, I didn't have a way to verify how much current the drivers were putting out, that would be pretty good to know.
I don't think that's a significant factor, rise times are fairly low unless you're at higher speeds where bEMF from L di/dt term
@@szymonjastrzebski2909 it mainly improve the turque at higher speeds
@willemstigter6384 exactly! I am building a system right now where I have already pushed 48V to 60V and still did not achieve the required acceleration. Rather hit saturation. Am using Elmo Application Studio and moving to an Ingenia EtherCAT driver now. Message me if you’re interested to help.
Awesome dyno tester man!!! I love it. To really see the super power shining, give them 48v and 2.4amp+ 😊. Instant sub for me!!
Thank you! It'll be interesting to see how the speed-torque curve changes between 12V, 24V and 48V as well!
@engineerbo yep. I have always wanted to build a dyno myself. Im very impressed with yours. And your video is top notch. Cant wait to see the next one man. Thanks again for sharing all the results. (Ps: the kraken LDO might be interesting to test too 😀)
Just looked up the Kraken, it's quite a bit bigger so it might not be a fairest comparison, but it'll be fun to test! Thanks for the tip!
@@Vez3Dcoming from a decade of ebike design , I can see a trap with this thinking perhaps. I mean no disrespect because I love your ideas, I simply want to say I have seen the “more volts” trap before. What I am wondering is if we are in the wrong speed zone for 500mm per sec, I think it says we are in the wrong gear ratio. That’s my mechanical engineering view anyway.
How can we configure the belts for a 2:1 ratio off the A/B motors? Or even 3:1. Losses aside it would seem the better step givem the other issues with higher voltages. Cheers.
@@samdekok72double edged sword at times -if I understand correctly 2:1 and 3:1 ratio for ab motors so instead of 20t stepper pulley you’ll be running a 40t or 60t pulley?
If so then that might introduce stepper resonance/vibration vfa’s like seen on the early creality k1’s. The micro vibrations of overshooting each step and vibration/resonance of bouncing back to re-centre sometimes translates to stepper induced vfa’s. Afaik it’s usually more common/evident in higher torque and larger steppers where the holding torque forces are high on the rotor and higher rotor inertia results in more overshooting and a stronger pull back to micro step position.
Great video and dyno.
The TMC2209 can blow up because you have no bulk capacitance on your test board. You could also add a 24V TVS in parallel as a safety mechanism.
As other commenter mentioned, adding inductance to the spec table would give us a hint as to the high speed performance.
LDO-42STH48-2504AC = 1.5mH
LDO-42STH48-2804AC = 0.6mH
Stepperonline = 2.4mH
usongshine = 3.2mH
jk42hs48-1684-08af = 2.8mH
Its a trade off, so if you want speed you just choose a low inductance model. Doesn't need to be some fancy 3d printer brand. That and always use 24V or even 48V if possible.
Yes I was suspecting voltage spikes again, but I didn't manage to verify in time. One of my upcoming projects would be to design a more user friendly protection device to properly solve this problem.
How much capacitance is enough for safety?
@@saadqadeer7807 The rule of thumb is 100µF/A on the stepper motor.
@@saadqadeer7807 100uF is stated in the datasheet as a minimum. You could easily go 470 or 1000uF, whatever is on hand.
Yeap. low inductance can eat more current and gives you higher speeds.
I would also like to see more expensive stuff, like Nanotec. To actually see if price matters.
Nice video and results! Your TMC2209 drivers probably died because you weren’t using a capacitor between your 12V input line and ground before your driver. I used a 220uF capacitor without I single driver dying since.
Yes that could be it! The wires between my power supply and driver aren't very long, but might still be enough to cause an inductive spike.
Wow! Very in depth testing. Thank you for your service to the community. Look forward to more motor testing and would love to see some fan testing as well!
The dyno you built is fantastic! What a great idea!
Thank you!
Fascinating results, thanks!
Your testing setup and machine is fantastic.
Thank you! I'm glad you like it!
Excellent work! Thanks for sharing! :)
Thank you!
First time on this chan. This is a SERIOUS chan with the testing! I love it!
Thank you!
Dude thank you for this comparison, what a gold mine with all these charts and test!
Thank you!
Great video! I'd just warn you, when using TMC5160 drivers in the future, to not use them above 24 V if using it in the stepstick form factor. The design pioneered by Watterott and copied by other manufacturers have a design flaw that makes them burn randomly, as many Voron owners that run their stepper motors at 48 V have discovered. Some manufacturers, like Bigtreetech and Mellow, have designed a stand alone board using this driver that correct these flaws, but they need to be connected to 3d printer boards using adapter cables.
Also, I'd like to suggest including the OMC 17HS19-2504S-H stepper motor in the test, they are also rated at 2.5 A and are said to be similar or slightly better than the LDO Speedy Power ones. The only drawback is that this motor is only available in the version with wires coming out directly from its body, instead of having a connector as all the other motors tested.
Interesting, what is the design flaw? A couple other comments mention the lack of capacitance (presumably near the driver), which probably means voltage spikes killed the drivers.
@@engineerbo On page 16 of the TMC5160 datasheet there's a section specifically saying that you should not provide more than 40V on the VSA pin, it's the pin that supplies voltage for the internal 5V and 11.5V regulators. On common stepstick designs VSA is tied to VMOT, which will supply whatever voltage you are giving to the motors. So using them with 48V is basically silicon lottery. That's the flaw the standalone 5160 boards solves.
Don't get me wrong, the capacitor is also needed, a good rule of thumb is 100µF for every amp of current that's given to the motor. For these ones you tested a 330µF electrolytic capacitor of 50V or more should be enough for all test cases.
Thanks for the heads up! It'll be pretty annoying if I had to modify the boards to test at 48V.
@vinnycordeiro I just found some time to have a look at the TMC5160 datasheet. On page 16, the note I think you're referring to doesn't actually say not to connect VSA to >40V. Instead, it's simply recommending VSA < 40V if certain conditions are met e.g. MOSFET total gate charge > 50nC.
This note is related to the text preceding it (Chapter 3.2), which talks about high power dissipation of the internal linear voltage regulators at high input voltages.
So the designs by BTT etc are actually fine, at least in this regard.
@@engineerbo I oversimplified my answer but yes, you are right. It's just easier to stay under 40V than having to double check the MOSFET choice of manufacturers. BTT did that on their TMC5160T Plus V1.0 board, connecting VSA pin to 12V directly instead of VMOT.
Great analysis! Subscribed for more driver tests.
Thank you!
Fantastic video editing and data presentation! Nice work, and thank you!
Thank you, glad you enjoyed the video!
Oooh... I know you will get lots of requests to add motors or models, or tests, but I like to think I have a decent reason to ask for another motor to be tested in a similar fashion: the Moons motor that comes in (for example) the Formbot kits, and is available on the Biqu site. That is also a pretty popular and common motor 'in the field' and I would love to know how it stacks up in actual 'accurate tests' against the LDO motors. I've got both on printers, but those are quite different in setup so I find it hard to compare the two. And I wonder if I should swap out the Moons for the LDOs, or vice versa (or simply stick with what I have since they are roughly comparable...).
Thanks for the tip! I'm keeping a list on all potential top performers. Do you have the part/model numbers for the motors you're referring to?
@@engineerbo The ones in my kit were MS17HD6P420I-04, I know there's a -05 as well, which is a new iteration of the motor and ought to be roughly comparable. Like the LDO ones, they are relatively 'premium' in price-class.
I'll look them up, thank you!
Great results! I'd love to see some "bottom of the barrel generic" motors tested, like the ones you get in GRBL combo kits, just to see how much performance you can get by investing in higher quality motors.
That's an interesting idea! When shopping for these motors, I realised I couldn't find ultra cheap motors like I'd thought I would. Maybe random kits might be a good place to look.
Cheap doesn't really matter for performance, its a tradeoff what you want, you can get a cheap low inductance high speed motor if you want. But the reliability and accuracy might be junk.
Awesome work. Appreciate your efforts to investigate and show the true torque capabilities across all he speed range
Thank you!
Great job and very interesting results! I was surprised to see how much of a difference thr stepper controllers made.
I had bought the super power motors for a cnc router. They were absolutely awful for that application. I swapped them with the motors in my 3d printer, and now the printer is faster and the cnc works. I think they went with a heavier guage motor wire with fewer windings in the super powers. This also lowers the inductance, so you get faster current changes. By putting them in the cnc, I was effectively trying to tow a fifth wheel with a corvette.
Yes it makes sense, if you're using the motors in heavy load applications like CNC, you probably want higher torque at lower speeds, and the super power and speedy power motors aren't the best in that regard.
Nice tests! You can try DM556, DM542 branded and unbranded drivers for comparison. And higher voltages too 36V and 48V. May be you can try higher Amps then specified by manufacturers and measure the temperatures, i regulary use steppers on higher amperage without problems.
Thank you! It would be interesting to see if any of the other drivers would be able to compete with Trinamic's drivers without simply shoving more current into the motor.
@@engineerbo I think the DM542 has been replaced by the EM542S, even already a couple years ago.
Good test. You will not be disappointed with the 5160.
I used 2208 (Creative board), 2130 and 5160 (BTT OCTOPUS) on an Ender 5 PRO and a 5 PLUS modified with linear rail and with StepperOnline like the ones in your test.
I have not measured the torque but the reliable maximum speed results are visible: if 2208 are 100%, 2130 are 150% (bit of whine at low speed) and 5160 are 170%.
Now with all 5160 the loudest things are the PSU and cooling fans.
Looking forward to see similar test for 48 and 60V
Congrats to 4K subscribers. Just saw that it flipped. Didn’t even knew RUclips refreshes that. 😂
Thank you! It just happened today!
Great video! Now I have to watch your other ones too. Thanks
Thank you! I hope you enjoy the other videos too.
You can use oscilloscope to get Voltage graph of the coils of the stepper. I use it when I tune up my steppers for my CNC.
Interesting, how do the phase voltages help with the tuning?
@@engineerbo ruclips.net/video/RWDZLtUeg6o/видео.htmlsi=zZrla6DPicAKmCRr
Thanks for the link! What changes are you making while observing the phase voltages to tune your steppers? It seems to me that if we can rely on the stepper drivers to do their thing, the conclusion is that higher voltages are better (to a reasonable degree).
Very in-depth and quality explanation! I've wondered about real world testing on this for quite some time.
Love it, finally somebody going deeper into steppers. I'm very deep into cheap servo motors - this territory might be even crazier! But, 14:22 your whole setup is flexing.. not good for reliability of data :(
Thank you! The bracket design is not ideal (I've designed it to make it easier to switch motors out for testing), but I don't think the torque readings will necessarily be unreliable.
If the motor is mounted on a flexible, springy bracket, the bracket will flex/deflect depending on the torque applied by the motor. The higher the torque, the greater the deflection. Yet the torque deflecting the bracket is still going to be equal the torque applied to the brake. This is going to be the same for a stiffer material.
In reality, there might be a slight power loss by the motor flexing and "de-flexing" the bracket, but I think it's not a major problem.
Quality content! Well done and keep it up! Love your work
Thank you!
Nice Video! I am looking forward to the next Tests Resuls of the LDO Motors combined with a TMC5160.
Thank you!
IMO the most interesting results are "hidden" in speed and torque curves - how much (answer: yes, torque varies pretty much between full steps :) ) speed varies when micro stepping.
In 3D printers we can see VFA at lower speeds (e.g. Prusa MK3, any Ender 3 clone), more strongly when the printer is designed for high speed (e.g. Creality K1). They can be fixed in various ways (0.9 degree steppers on MK4, smaller steppers and pullies on K1C).
Motor design (smooth stepping) and matching torque with moving mass (less torque and micro step judder with more momentum at higher speeds) and belts (spring) might lead to better print quality. Driver current may also affect VFA (not sure if it is tested) and might need to be dynamically adjusted for speed and acceleration of movement?
Silencing stepper noise on Bambu Lab printers and Prusa XL probably also provides smoother surfaces with less VFA...
Good video, really hoped to see the nema17 motors i personally use in this test. I use the steppermotor 2,1 A, 0,65nm motor. Wouldve been the highest holding force in this test while being not the highest Amp. The serial number is 17HS24-2104S - i am super happy with these motors and i can just recommend them.
Thank you! The motor you specified has a longer body length (60mm vs 48mm), which wouldn't have been great for comparison in this video.
@@engineerbo will there be a Test with 60mm Body lenght? Would be excited to see it. Really liked your test setup. 👍
I have lots of different tests planned, but not one for 60mm motors (I don't have any!). But if there's a large enough demand for this test, I can arrange for it.
This is a great video. I possess all of those motors and have played around with them not to the graphing extent that you have however. Speedy powers didn't last long for me because I'm dyslexic and I accidentally put the decimal point in the wrong area and fried both my speedy powers pretty much instantly. My superpowers however are very nice and they are connected to 5160 pro external Big tree Tech drivers . I was previously using the TV 6600 drivers and I experienced the same result
Thank you!
I also blew up a few btt 2209 drivers on my 24v printer, one of the capacitors failed. i think it was c6 next to the "bottom" label. i soldered a through-hole 100nf in that place and the drivers worked again.
Yes there are three capacitors between VM and GND. The caps are rated to 50V though, so they have a better than surviving than the TMC2209 chip.
Very nice experiments and Dymo tester!
Thank you!
The dyno design is great, but running stepper drivers on a solderless breadboard is just asking for problems and likely the cause of your 24v failures. When you are switching relatively high currents at high speed you need low impedance supply rails with plenty of decoupling, something that a breadboard just cannot provide. Why not use an existing stepper driver board like the SKR?
It was the first build of the dyno, the breadboard certainly isn't the best. I'll be building a PCB to replace it for the next video. I needed custom functionality for the dyno that perhaps something like an SKR could do, but it seemed more straightforward to implement it myself.
You need also to get LRC bridge so you can measure inductance values of the motors. Speed is really affected by inductance as if motor got high inductance voltage and respectively the current will not be able to go through the coil before it switches. So you put 1.77a but maximum current that goes trough is 1a. This can be changed with Voltage while increasing the voltage you mitigate this behavior as you saw with acient driver TB6600 which is terrible despite worse control Voltage was able to mitigate the effect of the inductace at the cost of thermal dissipation. You should start account for inductance bigger inductance bigger holding torque.
Yes higher voltages should also be able to overcome the back-EMF better, and definitely has an effect on motor performance. I'm also thinking of monitoring the phase currents, but perhaps adding more current sense resistors will be too invasive.
@@engineerboUse clamp style current sensors, ideally with a scope.
Great video! How does the load cell interact with the brake disk?
Thank you! The brake calliper is mounted on an arm that pivots around the same axis as the braking disc. The other side of the arm is attached to the load cell.
Great analysis! Love your dynamometer!! YES please microstepping vs torque! And I would love to see TMC5160 compared to an established closed loop stepper with optical encoder, e.g. Stepperonline CL57T driver.
Thank you! Microstepping vs torque is definitely a video I'll do.
Nice setup. It would be interesting to see the actual current waveforms of the different drivers. You have to keep in mind that when comparing smoothness of operation and noise, newer drivers are basically cheating because they don't really do steps anymore, they treat your step input more as a suggestion and generate a sort of sinusoidal waveform, trying to guess what you're doing and what you need.
Leadshine 42cm06 might interest you for power and speed :)
Those look nice (and potentially expensive)!
@@engineerboThey can be had for about $18, plus shipping. I'm looking at getting them to run my remote direct drive extruders at over 2500 rpm.
Where do you get them? $18 seems reasonable.
@@engineerbo I'm in on a group buy, but otherwise they somehow seem to be popular in India, and 3DPrintronics has it and has reasonable international shipping rates. 2 motors looked like 48 shipped for me.
Alright, I'll keep an eye out for this, thanks!
Cool tests. I"m looking for steppers for random projects other than 3D printing so this is helpful
Glad it was helpful!
Amazing video, it must have taken a lot of time to do everything!
It would have been cool to see performance of Moons motors, as they're very popular
Thank you! I've got Moons' on my to-do list now.
Nice video, testing motors at same current setting would be better idea
Hi. Are you still working on the next video? Would be great to see the performance of these steppers on tmc2240 at 24 volts, as its the common 3d printer standard nowadays. Great work btw. Subscribed
Yep I'm actively working on it!
You did a lot of valuable work here! I had actually contemplated getting IGUS steppers because they were some of the few who published Torque/speed graphs on their website. Though curiously their published numbers mark the motors as being a lot more powerful than even the LDO stuff in your test. maybe you could test one from them as well to see how the numbers hold up on your dyno?
I looked it up, and the drylin E stepper motors don't seem to have that much higher rated torques and are pretty pricey. Are these the same ones you're referring to?
@@engineerbo depends. they have many similar products, including very pricey "industry grade" stuff. I'll try to post a link, hopefully YT will let me do it.
@@engineerbo okay apparently that didn't work. You gotta try to sort by price on their website, and just pick the cheapest one.
Also to clarify: The rated holding torque isn't higher, but at least according to the first party graph they retain much more torque at higher speeds. (0.3 NM at 700 RPM / 24V)
Ok thanks, I'll have a look!
Top work, this is really useful testing! Thanks for putting it together. Looking forward to seeing the tmc5160 data. It would also have been interesting to see how your torque speed curves compare to the ones in the datasheets for each of these motors (stepperonline have t/s curves online for their motors).
I have had success using TMC5160 with Nema 23 motors for driving a peristaltic pump (pulsating load). I used spread cycle mode to get the best results in terms of limiting heat and noise for speeds from 0.1RPM to 500RPM. However, to accomplish this it is necessary to send a new configuration datagram based on selected speed.
Id you control Junction resistances on The bread board? At 2 A they might Be significant.
You're right, the breadboard is rated for 2A, so there would have been some power loss there. That would be fixed for the next video!
Great video and amazing work!
Would you be releasing the design of the dyno open source by any chance?
Hey! Few questions: Have you calibrated the dyno? Kinda afraid of belt drive efficiency changes as speed and torque change. Have you tried applying the braking torque slower? I noticed on your scatter plots that you don't have many data points near the peak, which could indicate inertia of the system causing an error, as you're not only trying to combat the torque of the stepper, but also torque from the deceleration of the system. You can also run a higher resolution encoder, as torque of the stepper is proportional to the magnetic misalignment angle within the stepper (full torque at 1 full step deflection, it follows a sine curve). A very popular method of testing the torque to speed curve of a stepper is to set and hold constant braking torque and increase the speed gradually (so the inertial effects are minimal) till you observe skipping steps.
Can't wait to see further developments, good luck
Yes these are all good points that I've thought about while building and programming the dyno. The brake is being applied extremely slowly, which causes the test to take ages. There is a little stiction in the brake line that causes the brake to tighten less smoothly than I'd like, so there's still room for improvement.
I also considered testing by spinning the motor up with a fixed braking force, but in the off chance the torque output really does increase with speed, this testing method might not work (probably).
Would like to see the popular Wantai motors used in high-speed printers.
I would like to see a test with microstepps
Yes! The effect of microstepping on performance might be heavily dependent on the driver, but it'll be pretty useful to know the tradeoffs with microstepping if there are any.
@@engineerbo Also the effect of all the different modes and features of tmc5160. like stealthchop, spreadcycle, coolstep, SixPoint RAMP. How does changing the sense resistor affect the outcome for smaller motors like nema17?
I would also recommend using one of the "pro" version from mks or btt of the tmc5160. They have much bigger FETs, can be used in standalone mode by jumper configuration and you can more easily swap the sense resistor @engineerbo
What is standalone mode? The units I purchased seem to support step/direction mode, but it isn't clear so far how the current limit is set (without using SPI). I'll be reading the datasheet properly soon, so I suppose I'll find out.
@@engineerbo i havent looked into it in younger future as i also just checked the manual to make sure they are not in standalone mode. But IIRC you can set amps and micro steps through the jumper pins
@@engineerbo On TMC jargon, standalone mode is the one that makes them compatible with older drivers like the A4988. That's because TMC can be dynamically configured through UART and/or SPI, depending on the model.
Amazing work. thank you!
Wish there was a section for precision/position reliability also
Any plan to make that dyne open source?
Thank you! After doing all these tests, I found some areas where the dyno needs to be improved. If there's enough interest, I don't mind cleaning up and releasing the code.
would be very appreciated
Interesting results. Would be nice to run some resonance test for each of the motors. I have StepperOnline motors on several of my builds and they perform pretty well. In my Voron 0.2 build StepperOnline resonate less than LDO motors which results in better prints quality over all.
Thank you! That's an interesting suggestion. How did you test for resonance in your build?
@@engineerbo I think you could just add simple ADXL sensor to your rp2040 setup, place sensor on the motor itself and run loop changing motor frequency at the same time recording readings from the ADXL, at least I would start from there. On 3Dprinter setup it's much easier as Klipper helps to run such task semi-automatically.
My first guess would be that every motor has its own resonance frequency, and there wouldn't be "more" or "less" resonance, so I'm not sure how I'd design a test for this.
@@engineerbo I would assume a well-made and balanced motor should have fewer frequencies it resonates at. At the same time, I think the higher the vibration level and the wider the spectrum, the harder it is for the algorithm to mitigate those vibrations, and it will end up affecting the print quality. I'd probably make two tests: the first one to see the overall noise and the second to check how high the peaks are getting.
Subbed, awesome content! I would be really interested in testing the 5160tpro with the same motor lineup. Things you could include would be how microstepping affects the curves aswell as input voltage. :)
Thank you! The tests will be done in the near future!
I noticed that some TB6600 driver modules have a circuit connected to the TQ pin of TB6600 chip that chopps the current limit to 30% of its setting, during half step period. May be that is affecting your measurements; i mean, instead of its "brains". You can check by probing pin 3, whether it stays high (5v) or not.
I love thorough, informative, testing
Simple way to select a "fast" stepper is to look at the inductance (often listed) as it's the thing that will, during running, limit the current rampup and thus also torque.
This was a great vid.
So you recommend the high torque motors for perhaps the Z motors and the high rpm for the X and Y motors?
It'd be great if you could add leadshines to the list 42CM06 are quite common in 3dprinting too
Yep I have added some Leadshine devices to my list, thanks!
@engineerbo the cm06 graph looks too good for what it is and I'm not sure what the cm08 offers over it. Should be interesting to see!
On large printers I we usually run at lower rpm and higher torque (vcore 3/4) so I'm excited.
Might be worth plotting `speed * torque` versus speed. (This is just 'output power' versus speed, give or take a constant factor.) It results in a much flatter graph, which tends to be much easier to read.
Also, you can get stall torque with that setup. Just lock the brake on full, then drive the stepper.
You might be right about the stall torque. The reason I've added belts to the dynamometer is so it has a little compliance, which makes it possible to measure the torque when the motor is holding its position. But I don't think it'll actually be able to reach the actual holding torque, because the motor still needs to move, even if just a little bit, to move through its torque range.
@@engineerbo Ah. I thought you meant stall torque not hold torque. Yeah, hold torque is trickier, especially if you want it to test in the same setup.
One approach that can work is to replace the brake with a larger motor with a decent controller. In practice, especially if you have a bit of gear/pulley ratio between the two. Still do the actual measurement with the load cell, but replace the gradual application of brake with the gradual application of PWM (instead of open-circuit). You might want a braking resistor, depending.
Yes using an external source of torque would solve the problem. But I'm not sure the actual holding torque is that critical to know, since an extremely slow speed gets me a number somewhat close enough.
@@engineerbo Agreed, at least for this application.
Pull-in torque can be substantially different than pull-out torque even at zero speed, but I don't think you care in this case.
One other thing that might be nice to measure - although is somewhat terrible to measure to be fair - is the stepper motor resonant frequency and behavior around said resonant frequency. Note this depends on the inertia of your load! Doing a slide test at (near) your resonant frequency is somewhat of an interesting worst case.
(Essentially: a stepper motor 'snaps' to the next pole, especially when not microstepping. But in practice it'll oscillate around said next pole somewhat before settling down. Interesting things can happen when you do another step at the quarter/half/three-quarter/whole period of this oscillation.)
It would be interesting to have a Power vs Speed curve. As torque doesn't matter so much and can be "increased" by using a different mechanical setup (smaller pulley...). As well as static torque, dynamic torque only matters at a given speed...
Power is torque × angular velocity, so actually the power curve would give you the same information, just in a different form.
If you use mechanical advantage to increase torque, the tradeoff is speed. If you assume a perfectly efficient system, the speed-torque curve's axes are simply scaled up/down, but will keep its shape.
Very interesting! Would be great to also include a rough price of the motors next time
I found the "TB6600" (no such chip on them in the 1/32 microstep version, it's a TB67S109AFTG or similar) drivers to be really nice and reliable with the big heatsink, isolated inputs etc. Those tiny breakout boards can be a pain in the ass, they can get quite hot, can be really sensitive to noise and randomly die on you, just like this one.
Id like to see the difference between TMC2209 and TMC5160 at the same settings and currents
Very nice and detailed video but what makes me curious what about accuracy and repeatability of those motors or maybe it’s not determined by the motors it self I actually don’t know so please let me know/ us know . Keep it up 👍
Thank you! I've started planning to test for position accuracy, so there'll be a video of that in the future.
Great little dynamometer! Have you considered using a braking magnetic field instead of the clamping friction load? Just thinking you’d have a more consistent load since it must be vary hard to keep the variable of heat equal throughout each test.
Also, a lot of the magic in stepper motors is done with a lot more microstepping than was demonstrated in this video. Can you provide the code and or lookup tables used?
Using a magnetic brake was actually my first thought. Essentially using another motor as an adjustable brake, which means there's almost no wear and tear etc. But a passive one doesn't work at 0 speed, so it can't brake the motor to a complete stall. It's also a lot more complex.
You should also include the fancy stuff, like Nanotec. And compare similarly specked models for this to make any sense.
I'd be more than happy to test the fancy stuff if I could afford them!
As a Materials Science Engineer, I highly appreciate this video; very good job and thanks for sharing!
Just a quick clarification: when you mention the 750 RPM for the 500mm/s on the bambulab using a 20 tooth pulley, were you assuming 8 microsteps or 16?
Also, why did you went with 8 microsteps instead of 16 for your tests?
Tks
Microstepping isn't included in the calculations, because I was using RPM instead of steps/minute.
There's no particular reason for 8 microsteps, besides it being the "default" on the TMC2209.
Nice video! Is it possible to test motor noise for VFA generation video? I am of the mind to use a motor with the least VFA and sacrifice torque for a smooth curve! Do they exist!?!
Thank you! There've been several comments talking about positional accuracy, so I'll look into make a video on this after my speed and torque series.
Can you compare tmc2209 vs running with FOC (like simpleFOC)?
In a previous video I did a comparison between open loop (TMC2209) with closed loop and a DIY servo which uses FOC. Feel free to check that video out!
It’s unclear to me how the load cell is measuring the braking force. I would greatly appreciate if you could elaborate on that please. I love the dynamometer setup!
Thank you! The brake calliper is mounted onto an arm that pivots about the same axis as the disc, and the other end of the arm is attached to the load cell. This makes the measured force "downwards".
@@engineerbo Aah! That makes sense. Fantastic, thank you!
It’s a good dyno but the use of a padded brake and brake disc is a factor that can lead to bad numbers in the long run due to nature of wear. Maybe using centrifugal force and fixed weight would have less outside influences.
Thanks! The condition of the brake actually shouldn't matter at all, because I'm measuring the actual braking torque.
For example, if the test was running, and I decided to oil the braking disc midway, this would cause the braking torque to decrease, and this change will be measured by the load cell. Oiling the disc would only mean the test would take longer, because the brake calliper would need to clamp harder to achieve the same braking torque.
@@engineerbo ok you’re compensating the loss of torque with time. Thanks for the explanation
Awesome video! I run the AH (high temp) version of the speedy power motors and they are brilliant. In my printers I’m using normal TMC5160’s @ 24v. However, in my little CNC machine, I’m running external TMC5160’s from MKS @ 36v. The results are amazing, I love these motors! 😂
PS, high temp means they are rated up to 180c according to the datasheet.
Thank you! I purchased a bunch of the smaller TMC5160 modules since they're rated to around 3A which is enough for these NEMA17 motors. Are you using bigger steppers with your external drivers?
@@engineerbo I’m running the LDO 42STH48-2504AH on my little CNC machine too. It started life as a 3018, so there isn’t room to use Nema 23’s in its current config.
I see, did you need more than 3A? Or were the bigger/external TMC5160s more for ease of mind?
@@engineerbo At the moment I only need 2.5 Amps, but there was an element of future proofing behind their use - if I ever decided to buy Nema 23's. TBH, I probably wont as they just arent needed on my little 300x300mm (work Area) machine. I also knew from research that many of the non-tmc external drivers were not very good.
On a side note, I found it amusing that you used an A4988 on your brake - I found that I could get more torque and speed out them, than I could with TMC2109's driving the LDO's. They can actually put out 2A peak, unlike the TMC2209's - but they make a lot of noise doing it.
Btw, I got mixed up, I have TMC5160's on my Printers, and external TMC2160's on my CNC machine - that latter is good for up to 4 amps and easily configurable via dip-swtiches.
Thanks for the information! I was wondering if I should get the bigger TMC5160 modules as well for NEMA 17 motors, but it's probably overkill.
how do I get the slowest possible rotation? I wanted to use it on a star tracker so around 0.0007 rpm 😅
I'm not familiar with the requirements for a star tracker, but with a "standard" 1.8° stepper motor without gears, 0.0007RPM corresponds to 1 full step every ~7 minutes. If you used 256 microsteps, that's 1 microstep every ~1.7 seconds which sounds better, but I think there will be accuracy/torque trade-offs.
You could also consider 0.9° steppers, or steppers with gearboxes (e.g. cycloidal if you need low backlash).
Nice setup ! 😃
The TB6600 drive you are using is not a TB6600, it has a Toshiba TB67S109 chip that is better than 6600 but less current capability. The same chip can be found on small step sticks, also.
Nice, there was a general lack of documentation regarding the TB6600 when I was looking it up, so this information is useful. Thank you!
wow great test machine
Thank you!
Why not testing them against 50kg load cell ? What cell is used?
This is a small load cell designed for weighing scales. AFAIK the load cell shouldn't matter too much as long as the loads are within spec.
@@engineerbo Yeah, I also used for small weighing scales (max. 5kg ±1-2gr precision) but they tend to make errors above 2 kilos
I see, I didn't check the load cell's linearity or accuracy, but I'll take note of this, thanks!
This is really SOMETHING!
Can you do TMC5160 at 48 volts?
Yep, that's the plan!
I'm very confused because speed-torque curves aren't supposed to look like this, there should be a flat torque output then a drop when the back EMF restrains the current that can be pushed though the windings
Yes we should see a flat top at low rotor speeds, corresponding with the driver's current limit. Perhaps we'll see this if the tests are repeated at higher voltages.
How did you extract the peaks from the rest of the data to get the motor dyno curve?
The data points are binned into the speed steps, since actual noisy measured speeds are used. The peak at each speed step is the highest value in the bin.
I’d have done a few different tests too. Like, with the same stepper, see how the curve changes as a function of microstepping. And maybe to compare the motors at the same current rating. More importantly though, you didn’t plot the mechanical power alongside the torque-speed curve!
There's lots to test, too much to put into one video of reasonable length. I'll be testing the microstepping in the near future!
Good test. Note the curve for LDO motors (at 13:45) is not marked speed-torque, but torque over frequency (hz). Would this not illustrate a claimed property for step frequency, rather than RPM?
Yes I too interpreted it as step frequency, which is the same as angular speed assuming the motor isn't stalled. E.g. for this motor, 200 (full) steps per minute is 1 RPM.
poderia Usar o TMC2160. Excelente driver da trinamic, com capacidade de 4.1A e 36V.
Excellent video
Thank you!
You should have done these with 5160 drivers... The 2804 gets it's performance at 2.5A+@24v and 2504 at 2.0A+@48v
Yes, but I first need to setup the firmware to configure the TMC5160 via SPI.
@@engineerbo just use a BTT board and Klipper, doesn't this test just need to move the motor in different feed rates?
The dynamometer controller is controlling everything, including step generation, so it knows what speed the motor should be moving at and when it stalls.
I recently had a problem involving this exact setup: a (generic noname) NEMA17 stepper, a TMC2209 driver (used with factory defaults as dir/step only drive) at 1/8 microstep and a 12V supply - low speed torque seemed ok, but the motor literally stalled around 600RPM or so, with no load, no matter what I did. Tried a beefier PSU, tried maxing the current pot as far as I dared, tried different microstep, nothing helped - even with gradually trying to raise the speed, the motor just started shaking and stalled. Really weird, and extremely disappointing - 600RPM a.k.a. a measly ten turns per second is NOT all that fast...
I love the pun in the title screen 😂
why r u testing 2209 at 12v? It can go up to 24v as well
I blew up the TMC2209s at 24V (mentioned in the video). It could be a problem with not having a capacitor near the drivers. This will be fixed for the next tests.
In my experience, Oriental Motors PKP series are the highest quality stepper motors you can buy.
Sanyo Denki steppers are also nice but not as nice.
Thanks for the tip! It'll be fun to test them but the Oriental Motors models are ~3x the price!
Torque is not everything, also how true the 1.8° steps are is important. With the encoder you can measure this, would like to see this :-)
I'll add this to my todo list, thank you!
You should try normal stepper driver, like TMC5160 (for example Btt tmc5160 Plus), and try the 24V and 48 Voltage
Your TB6600 seams to be a shit driver, that's drastically limits potential of those motors
I can easily achieve 1m/s on my 500x500mm bed size printer, using LDO-42STH60-2804AC-R on 48V, that's 5000 RPM on 12mm pulley I guess!
800mm/s with 24 Voltage
And its moving quite heavy aluminium profile on X axis, and toolhead assembled from two INOX plates
Good video, Im waiting for more ❤
Yes! I've bought a few TMC5160s that are now waiting to be tested.
need a 48v test with tmc5160
Yep that should be in the follow-up video.
I'm a bit sad you didn't include a quick noise comparison. Put them all on the 2209 with the same microsteps at the same rpm and see which is quietest.
I agree that would be very nice to have. Maybe also include a 0.9° stepper and 48V with the TMC 5160.
In your experience, would the model of the stepper motor make a substantial difference to the noise?
@@engineerbo not as much difference as the drivers, but still a considerable difference at increasing rpm. imo driver is really reducing noise at low rpm, while motor build quality reduces noise at higher rpms. just very limited experience tho.
I haven't really considered the noise factor much outside of the smoothness of the driver. I'd have guessed the higher RPM noise could be caused by bearings etc. I'll take a measurement if I notice the difference.
I'm confused why you didn't get any hybrid steppers on your lineup?! They put normal steppers to shame in pretty much every single category, except one - price. They're the ones in one piece aluminum round cases, instead of laminated steel sheet bodies. They are insanely powerful compared to a cheap laminated steel stepper
All these motors are hybrid stepper motors. Are you referring to something else?
@@engineerbo Hmm, I know them just as 'hybrid steppers'. They are just normal 4-wire steppers, but they are in round aluminum cans, I got a box of them in an auction years ago (they were buried in a pallet), and I've used them for various projects over the years. The first picture for google images "EHY21512" or "23HY4604" shows the TYPE of stepper I'm talking about. The cases are round aluminum cans, not laminated steel sheets
Ok, hang on a minute, how did you burn out 2x 2209 drivers when 99% of printers all run on 24V's?
Someone rightly pointed out I should have placed a capacitor near the driver, which would have helped absorb the voltage spike!
I'm trying to figure out why you can't measure torque at zero speed.
I'd think the load cell on the dyno wouldn't be the problem. Is the problem with needing to provide steps to the driver?
Any insight would be appreciated.
I really enjoyed the video. Thanks for all the hard work. I subscribed.
Thanks for subscribing! To measure holding torque, the motor is usually made to hold it's position, and increasing torque is applied to it (from an external source, e.g. weights) until it position holding fails.
With my dynamometer, the torque comes from the stepper motor itself, so strictly speaking, it cannot be at zero speed and increasing the torque at the same time.
Man, a leadshine dm542s or something close, instead of a tb6600. They burn, make a lot of vibrations, etc etc etc, truly the worst. I did that when I built a mpcnc years ago.
The TB6600s are dirt cheap though! Probably good enough for casual projects.
Epic video
Thank you!
Жаль, что в сравнении не участвовали моторы Leadshine. 😔
I got in contact with Leadshine, we might be able to arrange something!
Это было бы замечательно! 👍