Our great sponsors
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
I used 2 link paired 25kg servos per joint. I went this route because I had quite a few lying around, but it didn’t work well because the servos would fight each other due to imperfections in the servo assembly, so even though they were set to the same position that actual position didn’t match between the 2 of them. If I were to do it again, I would either go with a single servo, or more likely go with a single BLDC motor and gearing because I have more control over the specs I’m looking for like speed and positional control resolution needed as well as backdrivability which allows for better response to uneven terrain at high speeds. You can also get the needed control boards (may be easiest to just design / order a custom board like this one and solder on the components as this is actually not as expensive as one might think, I’m currently doing this for a large humanoid project) and encoders (hall effect specifically) for relatively cheap (while not as cheap as servos, when considering the features you’d get it, it’s more fair to compare them to something like a dynamixel servo which, depending, you may actually come out cheaper).