Hi,
They're shity but I haven't found anything better for the required voltage 150VDC. And they are relatively cheap.
FYI: SanyoDenki driver were offered to me for €1500.
They're hard to tune. Maybe it's problem of experience with the servidrivers tuning. But after weeks of experiments were have never reached the motions so smooth like with original drivers. It's easy to set them and have perfect hig speed and accuracy, but when you test it at small jogs - you loose a lot if steps. When you tune them to have it being good at small steps then long motions are drama etc etc.
I've seen some Serbian drivers having auto-tune mode friendly to use at the beginning at least. The have looked promisely but I haven't bought them finally.
If you will accept much more noise from motors and instability of positions (little trembling which you can't see but hear) - you can work with them as I do. But it's very very far from original solution. Also speed and "culture" of moving.
They hangs sometimes or loose settings sometimes.
Misleading are some settings in the interface that are not clearily described what are they for, or work other than general servodrivers' docus are used to say. Also some parameters are visible, described but do nothing. After weeks I've got info from distributor "ach, sorry, factory has forgotten to remove it. It worked in older versions only but not now".
Manufacturer is useless. When I've contacted to them, I was given some answers but useless. And stopped to get response when digged deeper. My impresdion is that they seems to be maybe good electronic engineers but with poor servomotors or steppers practical experience. I've thought with very low motions and dynamics and got zero advise why it's so shity. Much later I've found that voltage for motors powering must be much higher than nominal motors voltage parameter is, to get reliable speed. Manufacturer didn't know about it, but SanyoDenki engineer does, and some steppers' engineer either. So if motors are 75VDC, you must power them with 150VDC and control the current only. And only then you'll get it working as it should be.
Smoothieboard is too slow for these drivers. It has only 100kHz STEP output and it's too low to use full motors encoders resolution and full speed of motors. You must choose: max speed and lower resolution (0.02) or half speed and full resolution 0.01. Despite 0.02 is enough in practice. But to get full parameters you need 200kHz motion controller, so not the Smoothie (and no matter which version) but Marlin having other disadvantages. Compromises, compromises...
Today, maybe I'd use some AC servomotors and drivers. But who knows what the issues then with Chinese drivers.
You can talk to Jaroslaw Karwik user about it. He's experienced with servomotors problems and designing some advanced motion controller for it and Openpnp machines. But be prepared on higher costs as Jarek is using advanced and not cheap solutions rather. But I think it's worth to observe his progresses (not to big lastly or not reported only).
@Jarek, how does it go? I'm sure you read it :-).
brgds
M/