I am running LinuxCnc with an Octopus Pro board and Remora firmware. All works fine except that the servos motors make cracking noises when reversing the load. With the oscilloscope I see that when the DIR Pin changes also the first Puls starts (~120ns). I assume that the problem comes from there. But I can not find a setting where I can set the values for Dirhold and Dirsetup.
I have tried:
setp remora.joint.0.dirhold 10000
setp remora.joint.0.dirsetup 10000
Remark from the Servo Manuel：
(1) t1: ENA must be ahead of DIR by at least 5μ s. Usually, ENA+ and ENA- are NC
(not connected).
(2) t2: DIR must be ahead of PUL active edge by 6μ s to ensure correct direction;
(3) t3: Pulse width not less than 2.5μ s;
(4) t4: Low level width not less than 2.5μ s.
My hardware:
Rasperry Pi4 4Gb with the RT image.
Octopus Pro Board via SPI
JMC IHS57 Servos
Software:
LinuxCNC 2.8.3
Remora 1.1.0
Currently there is no setup and hold settings available in Remora as it has not been an issue.
Looking at your servos, they are expecting a differential pair driving them. Not sure if this is the issue but maybe worth a try to use some adaptors like
The long standing issue crops up enough to make the lathe unusable. You can move the system on either axis and when it stops or nears stopping, it (sometimes) suddenly jumps into motion (step change) and starts slowing down. This can be seen below. The position feedback value changes and the system compensates.
-Rafal
I'm curious as to whether we've got a similar issue that was a bug in the Remora-NVEM version where the data move was being interrupted and corrupted by the threads. It was fixed by disabling interrupts before moving the data.
Code:
// ensure an atomic access to the rxBuffer
// disable thread interrupts
__disable_irq();
// then move the data
for (int i = 0; i < BUFFER_SIZE; i++)
{
rxData.rxBuffer[i] = rxBuffer.rxBuffer[i];
}
// re-enable thread interrupts
__enable_irq();
Doing the same in RemoraComms.cpp at lines 165 may do the trick.
Ive seen this in sim, but running into this issue on one of my physical machines, so I would like address this. Based on this post, scott, is this the change you would like to make to address the issue? I made a separate repo with this change, if this is what you were thinking I can push it through.
Recently things with mbed have spurred a community fork of mbed called mbed-ce, with our friend jojo as an active developer. I think the goal being actual updates to hardware bugs.
github.com/mbed-ce/mbed-os
Basically they are working imxrt1062 aka teensy4 support for mbed, and need people to help test. Since Im stupid as shit, I will be fumbling my way through, but if anyone else is interested, here are a few links to the stuff
Haven't ever used a forum like this before sorry if I am doing this incorrectly.
I am trying to get remora working on an SKR 1.3 board with an RPI4. I'm sure this is something stupid simple, everything appears to be working as far as SPI between the RPI4 and the SKR 1.3 but when I take it out of estop in linuxcnc and click to power it on in linuxcnc everything seems good. As soon as I try to move any axis I get a joint follower error.
I am using the remora xyz config for linuxcnc and it is as-is from when i downloaded it. I am clueless as to where to start looking for the issue.
