Remora - ethernet NVEM / EC300 / EC500 cnc board

I have an ec500 with rt1052

zmrdko wrote:

Thanks Scott. It was set to 0, but I changed it to 0.0001 and it seems to work fine now. I have now issue with motor accuracy. The scale is set to 160 and I have set jumpers on JMC ihsv57 motor to 800 steps per rotation so 5mm jog should be exactly one full circle. However its a little less than that. After few 5mm jogs its obvious. So do I have problem with step pulse lenghts?

Sounds like some calibration is needed to dial in the step scale.

forum.linuxcnc.org/38-general-linuxcnc-q...rating-axis-movement
www.forum.linuxcnc.org/plasmac/45808-scale-axis#242146

Formula is: (distance commanded / distance moved) * old step scale = new step scale

 

meister wrote:

source/baseThread.c:

Code:

void makeSteps() {     static int8_t i;     static int32_t stepNow;     if (rxDataFlag == 1) {         rxDataFlag = 0;         memcpy(rxDataCopy.rxBuffer, rxData.rxBuffer, BUFFER_SIZE);     } .... but i think its better to move this part to the end of the makeSteps() function  

Yes, I saw that. But where is rxDataCopy being filled with the packet data?

Thanks Scott for patiently replying to all my queries.

i know this formula for correcting travel, but I was expecting the exact numbers to work. If the motor needs 800 pulses to make one turn, I would expect 160 scale times 5mm to work.

I have read that it might be caused by too short pulses. This is diagram I found in motor manual:

is there a way to set up these values?


 

Currently there is no step parameter settings configuration. Something in the pipeline though.

Your drive needs a 2.5 microsecond step pulse, the step pulse output form the Remora firmware is 25 microseconds in length. So should be ok. I'd do a step/mm calibration and run it though some test moves to check repeatability.

Nice motors BTW!

Yes, I saw that. But where is rxDataCopy being filled with the packet data? 

memcpy(rxDataCopy.rxBuffer, rxData.rxBuffer, BUFFER_SIZE); 

Following is the declaration for memcpy() function.void *memcpy(void *dest, const void * src, size_t n)

the memcpy command can be a bit confusing, because the first argument is the destination

What is the likelyhood of the 1052 firmware running on an imx rt1062? I've asked about Teensy 4.x firmware in the past which didn't get much traction but since there's ethernet firmware for a board running an rt1052, firmware for a Teensy 4.1 can't be too far off.

Hi Scott

Could I get you to check this
I am running an NVEM RT-1052, when I activate remora.output.00 OUT-2 comes on.
And vice versa. remora.output.01 turns OUT-1.

This seems to be backwards. Is this whats intended?

Thanks
Alex

frayja2002 wrote:

Hi Scott

Could I get you to check this
I am running an NVEM RT-1052, when I activate remora.output.00 OUT-2 comes on.
And vice versa. remora.output.01 turns OUT-1.

This seems to be backwards. Is this whats intended?

Thanks
Alex

Hi Alex,

No, nots what intended. I'll check it out but I'm thinking there may be a very subtle difference between the EC300 and NVEM. I developed on the EC300 as it has LEDs on the output. Everything else align numbering wise?

I have now checked them.
All the other outputs seem to be in their natural order.

Alex.