EXP3HC homing issues, sometimes the axis doesn't stop

JA

Hi, I am trying to configure a duet3 machine in standalone mode, my setup includes 2 expansion boards: MB6HC + EXP3HC + EXP3HC .
I am runing RRF3.1.1 and on the expansion boards 3.1.0

The machine works well, the problem only comes while homing
( i've found out later the problem ocurres while using the comand "G1 H1 ... " )

Since i could not find the issue i started with a simplified sistem, using the same electronics i configured a IDEX machine with 3 of the axes on the first expansion board.

After a bunch of tests this is what i found looking at the U axis, witch is on the first expansion board

" G1 H1 U-100 F2000" works ok.
" G1 H1 U-100 F3500 " does NOT work.

If i hit the endstop while the axis is moving it doesn't stop.
if i maintain the endstop with my hand and then send the G1 H1 comand, it works ok.

If i use the "homeu.g" file:

G91               	; relative positioning
G1 H2 Z5 	F2000   ; lift Z relative to current position

G1 H1 U-100	F3500 	; move quickly to U axis endstop and stop there (first pass)
G1 H2 U5 	F360 	; go back a few mm
G1 H1 U-10 	F360  	; move slowly to U axis endstop once more (second pass)

G1 H2 Z-5 	F2000   ; lower Z again
G90               	; absolute positioning

If the speed of the first "G1 H1" comand is inside the "bad speed" range , it doesn't stop when the endstop is hited, it doesn't execute "G1 H2" and ejecutes the next "G1 H1", since this is outside of the "bad speed" range the second one works ok.

This happens in a range off speeds, and if i change the U-100 to U-50 also changes the range. Seeing that i tried changing the acceleration of the axis, and this also changes the range off speeds.

If i configure a acceleration of "M201 U500"
G1 H1 U-100 -> [F3492 - F3512] Does not work "bad speed"
G1 H1 U-50 -> [F1992 - F2005] "bad speed"
G1 H1 U-20 -> [F1092 - F1201] "bad speed"
"M201 U400":
G1 H1 U-100 -> [F5053 - F5103] "bad speed"
G1 H1 U-50 -> [F3292 - F3333] "bad speed"
G1 H1 U-20 -> [F1675 - F1700] "bad speed"

This is completlty repeatable, every time i send a "G1 H1 with the right acceleration, distance , speed combination the endstop doesn't stop the axis.

The only posibility i see is that RRF 3.1.1 or expansion firmware 3.1.0 has some incompatibility.

( i have also tried the 3.1.0+ BETA version dc42 put here: https://forum.duet3d.com/topic/17285/hiccups-on-3hc-expansion-board/24)

Just in case, here it is my simplified versions config.g

deckingman

@JA I reported a similar problem over a month ago but nobody is interested. I had to come up with a mechanical "fix" to get the Z axis somewhere near, then I can home as normal at slow speed https://forum.duet3d.com/topic/17404/z-switch-on-expansion-board-latency-issue/22

fcwilt

Hi,

Can you update the expansion boards to 3.1.1?

Frederick

dc42

There was a fix done to the main board firmware a small number of weeks ago, which may be related to this issue. There is a new main board binary at https://www.dropbox.com/sh/3azy1njy3ayjsbp/AACquxr2m00eV568RZg5QG5wa?dl=0.

JA

Thank you @dc42,
Now works as is supposed to with

MB6HC firmware version "3.2-beta" "2020-07-15b1"
EXP3HC firmware 3.1.0

But seeing that this fix is implemented since 15 of july, almost a month ago, that the latest release of RRF 3 was on the 19 of May and that there have been a bunch of commits since then raises a new cuestion:

Is RRF3.2 close?? if not have you considered releasing a intermediate point like 3.1.2 for example??

Could you comment something about it ??

JA

@fcwilt said in EXP3HC homing issues, sometimes the axis doesn't stop:

Hi,

Can you update the expansion boards to 3.1.1?

Frederick

My understanding was that there is no 3.1.1 for the expansions

fcwilt

@JA said in EXP3HC homing issues, sometimes the axis doesn't stop:

@fcwilt said in EXP3HC homing issues, sometimes the axis doesn't stop:

Hi,

Can you update the expansion boards to 3.1.1?

Frederick

My understanding was that there is no 3.1.1 for the expansions

Perhaps not.

I was under the impression they were going to try to keep all the firmware versions for all the various boards "in sync" to reduce confusion.

Frederick