Board for very high step rates

Hi,

so my printer (custom) is a corexy with 0.9 degree steppers running @1/64 microstepping (for quietness)
this results in very high step rates required for speed… i currently run it on a SKR 1.4 Turbo which can handle up to 160mm/s movement speed. While this is fine for printing, it’s not that fast for travel, so i’d like to upgrade to a better board, BUT:

i already upgraded (back in the day) from the SKR 1.3 to 1.4 turbo and it literally did not make any difference. As i understand, the MCU clock is not really an indication of how fast the pins can be toggled, and it kinda shows in the current benchmark page of the documentation, where the top performer seems to be the RP2040 by a HUGE margin, despite the modest clock speed.

unfortunately, there is only one board with a RP2040 which i don’t like. Can you recommend a board (or MCU) that can actually step considerably faster than the LPC1769? from the benchmark page is not really that obvious which kind of step rates you could get

1 Like

Hello @Bracco !

The klippy.log would be of great help here.

Haven’'t you read the template before you deleted it?

besides the fact that i can’t provide a log as i’m running a long print right now, i don’t see how that would help in any way. i just need to know if there are any board with an mcu that can reach substantially higher step rates than the lpc1769 of the skr 1.4, considering the fact that the mcu frequency and benchmark page do not really tell the full story

Sorry, I misread the title.

Have you looked at the step benchmarks? The RP2040 is nowhere near the top performer, and is almost identical to the LPC1769. Look for something with the STM32H743.

https://www.klipper3d.org/Features.html#step-benchmarks

i was looking at the benchmark page, where the rp2040 sits on top. also, the rates reported in the benchmark page differ greatly from those in the features page. perhaps it is not the same number?

anyway, as i said, i had a 100mhz LPC1768 board and upgraded to a 120Mhz lpc1769 board (years ago) and the performance gain was basically 0, unlike what is stated in the benchmark page. To be fair, in that page it is explained that the result is obtained by overclocking the 1768, which i presume is NOT the same as having the true 1769 (either that, or the firmware for the 1769 makes it run at 100mhz)

anyway, my point is that the benchmark data is not really an accurate representation of the real world performance. sure, i see that STM32H7 can reach 550Mhz which is really fast, it can for sure compute stuff much faster than anything else but can it really toggle the pins as fast as you expect? that’s not an easy answer

i don’t quite need processing power, rather pin toggling speed…

All I can say is I’ve done testing with a STM32F446 and I achieved much higher step rates than what you described in your first post.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.