To describe a little bit better my setup, I have a test motor connected to a load motor. The load motor sets the load torque (via another controler). The test motor is speed controlled via an Odrive S1.

Here, for instance, I show a picture of my oscilloscope. In yellow, the voltage between 2 phases of the inverter (hence terminals A & B of my motor). At the begining we can clearly see a sinusoidal PWM, with the correct electrical frequency (motor spinning at 250rpm and 5 pole pairs, so fe=250/60*5=20.8Hz, T=50ms). But then the PWM turns off, I can’t understand why. Furthermore, the current (in cyan) is unbothered. This behaviour repeats periodically over time.

Can anyone elaborate on this behaviour? Is it normal?

Thanks in advance!

.jpeg)

I believe we’ll be restocking the magnets within a month or two. Meanwhile, you can use e.g.:

8mm x 3mm from Guysmagnets
6mm x 3mm from Magnetstore

Those will both be lower field strength than the 6x5mm ones we sell on the shop, but that’s perfectly manageable. I’d recommend giving the aforementioned magnetic encoder design guide a close read, and the examples section is useful.

I’d probably recommend getting an XT60 extension cable, then cutting that and connecting it to the S1. That way, you can easily connect/disconnect the battery.

One thing to note is that when you connect the battery, it’ll dump current into the capacitors on the S1, which can cause a pretty big spark and degrade the connectors on the battery (as well as stress the battery and S1 a bit). So I’d generally recommend using a specialty “antispark connector,” which has an additional component in the connector to prevent this (and cause a smoother startup of the current flowing into the S1 when connected). That being said, it takes a bit of soldering to connect that to the battery, which you may or may not be comfortable with. There’s also standalone antispark power switches like these.

As the shop is currently out of stock of the encoder magnets, has anyone found a UK supplier ? The exact dimension magnet (6x5mm) is difficult to find, I suspect that it may still work with 4 or 5mm diameter magnets? but can anyone confirm if that would cause issues

Regards,
Martin

I was wondering what kind of battery I can get to support the motor, I wasn’t sure which one would fit during my research online

Good question! Really depends on the the maximum speed and torque you need. It’s usually easiest to just start with a power supply, we sell this one, but it’s a bit chunky – anything over 12V/60W or so should be fine to get started and testing – higher voltages and powers are really only needed for higher torques and speeds, respectively.

Does the M8325s 100KV need a CUI encoder to function?

Nope! All ODrive systems need some sort of encoder, but the M8325s + S1 kit uses the onboard magnetic encoder on the S1, makes things nice, easy, and compact

The encoder is single turn, shall I set biss_c_multiturn_bits to 0?

Yep!

After these this change, shall I click the Call button to the Save_configuration? Then go back to the configuration tab and run the remaining steps?

The “Save & Reboot” button in the configuration tab will also save configuration. So you could run save_configuration in the inspector tab, or back in the configuration tab, both works

Thanks for your support.

1. Select SPI encoder, leave protocol blank
   

   

2. Run the first two steps on the final tab:
   

   

3. Go to the inspector tab, search spi_encoder0, and double click on (or drag to middle “controls” pane):

• config->mode
• config->biss_c_bits
• config->biss_c_multiturn_bits

Change mode to BISSC, and biss_c_bits/multiturn_bits to the number of singleturn and multiturn bits from your encoder.

4. Go back to the configuration tab, run the remaining steps (starting with “Save & Reboot”)
   
   

DocT:

but it seems it only works with the V3 ODrive, as the firmware wouldn’t be compatible with the ODrive S1. If anyone has had success in

Can you guide me on how to do this?

My plan is to just connect the A+, B+, and Z+ wires to the ODrive and ignore all the negative wires (A-, B-, Z-) and the U/V/W commutation wires. I want to make sure this won’t cause issues with the ODrive’s hardware or the startup calibration.

Encoder Specs
AM26C31 Line Driver.
Differential A, B, Z plus U, V, W commutation .
Resolution 2500 C/T or 4096 C/T.
Voltage: DC 5V.
Logic Leve VHigh​≈3.5V and VLow​≈0.1V.
Speed/Accuracy: Up to 20,000 RPM with 0.1∘ linearity.

My Main Questions:
Logic Levels: 0.1V(Low) and 3.5V (High) work fine with the ODrive’s internal 3.3V pull-ups on its 5V-tolerant pins?

Skipping U/V/W: Can I still do a normal ENCODER_MODE_INCREMENTAL startup without the U, V, and W wires connected?

Sounds like it will do the job, we’ll get some part ordered.

I’m also planning to use this encoder for a new AMR model, for this new machine we need to have the multiturn information. Do you have any update on this topic?

Regards,
Abel