@droftarts I'm on 3.2.
@tekkydave @fcwilt You're right actually, turns out I was doing it wrong. 700 mm/min is too much for my Z axis using 600 mm/min is working fine.

@fcwilt I actually set all the probing move speed a lot lower when I wasn't printing a lot but it was apparent to some extent. I kinda just lived with it because it doesn't skip steps when the rods are clean and lubed. Also, things like the dive height also affected how quickly it would start to loose steps. A short dive height < 5mm worked better. And probing a single point was not as dramatic as multiple points. My max z-axis speed is probably somewhere between 600 and 700 mm/min, so depending on how much friction the threaded rods experienced, it would z probe fine or not.
I also only truly started to investigate when I started getting "motor phase may be disconnected errors".
I checked all the motor wires and found no issue there
Then I started visibly seeing step lost in Z. Plus I also wanted to probe faster and be hands-off as much as possible in the pre-print process.
Lessons learned:

• keep threaded rod lubricated (obviously)

• Read the fine print of my gcode commands and understand how they inter-relate

• give myself some room around the max limits of the machine in the config.

Thanks to you all for helping me learn a bit and solve this issue.
Thanks @dc42 for adding the note in the documentation.
@tekkydave I second having the z lift speed independent for probing.

I'm having an issue where my part cooling fan, which is on "Fan0" per some of the recommendation I've seen, now doesn't behave as expected. I've read that the fan mosfet can fail and cause issues, but can it fail "on"?
The extruder heatsink fan is set on one of the "always on" IOs (I don't think I need to define it, if I'm using an always-on right?)

Old config lines were:

M950 F0 C"fan0" Q500                           ; create fan 0 on pin fan0 and set its frequency
M106 P0 S0 H-1                                 ; set fan 0 value. Thermostatic control is turned off
M950 F1 C"fan1" Q500                           ; create fan 1 on pin fan1 and set its frequency
M106 P1 S1 H1 T45                              ; set fan 1 value. Thermostatic control is turned on

New config lines

M950 F1 C"FAN1" Q500                            ; create part cooling fan on pin "fan1" and set its frequency
M106 P1 S0 C"Part_Cooling" H-1                  ; Thermostatic control is turned off

M950 F2 C"FAN2" Q500                            ; create case cooling on pin "fan2" and set its frequency
M106 P2 C"Case_Fan" S0.35 T40:65 H100:101:102   ; Thermostatic control turns fan 35% on when temp of mcu or stepper drivers is 40, ramps to 100% when temp rise to 65

It works fine now, but what kind of failure can cause the fan0 pin to be permanently on?
If I shift the current config to fan0 and fan1, the fan connected to fan0 stay permanently on.

@fcwilt I actually set all the probing move speed a lot lower when I wasn't printing a lot but it was apparent to some extent. I kinda just lived with it because it doesn't skip steps when the rods are clean and lubed. Also, things like the dive height also affected how quickly it would start to loose steps. A short dive height < 5mm worked better. And probing a single point was not as dramatic as multiple points. My max z-axis speed is probably somewhere between 600 and 700 mm/min, so depending on how much friction the threaded rods experienced, it would z probe fine or not.
I also only truly started to investigate when I started getting "motor phase may be disconnected errors".
I checked all the motor wires and found no issue there
Then I started visibly seeing step lost in Z. Plus I also wanted to probe faster and be hands-off as much as possible in the pre-print process.
Lessons learned:

• keep threaded rod lubricated (obviously)

• Read the fine print of my gcode commands and understand how they inter-relate

• give myself some room around the max limits of the machine in the config.

Thanks to you all for helping me learn a bit and solve this issue.
Thanks @dc42 for adding the note in the documentation.
@tekkydave I second having the z lift speed independent for probing.

@droftarts I'm on 3.2.
@tekkydave @fcwilt You're right actually, turns out I was doing it wrong. 700 mm/min is too much for my Z axis using 600 mm/min is working fine.

@fcwilt I have that, but I think it's ignoring it. Here is my config.g

; Configuration file for Omnicore
; Communication and general
M111 S0                             	; Debug off
M550 POmnicore      					; Machine name and Netbios name (can be anything you like)
;M551 Pmyrap                        	; Machine password (used for FTP)

; Wifi Networking
M552 S1								            ; Enable WiFi
M586 P0 S1                                      ; enable HTTP
M586 P1 S0                                      ; disable FTP
M586 P2 S0                                      ; disable Telnet

G21                                 	; Work in millimetres
G90                                		; Send absolute coordinates...
M83                                 	; ...but relative extruder moves

; Axis and motor configuration
M669 K1								; _RRF3_ change M667 to M669 ; set CoreXY mode

M584 X0 Y1 Z2:5:6 E3:4       		; Map Z to drivers 2, 5, 6. Define unused drivers 3,4 as extruders
M569 P0 S1                          ; physical drive 0 goes forwards  (X-motor)
M569 P1 S1                          ; physical drive 1 goes forwards  (Y-motor)
M569 P2 S1                          ; physical drive 2 goes forwards (Right Z-motor)
M569 P3 S1                          ; physical drive 3 goes forwards  (1st extruder)
M569 P4 S1                          ; physical drive 4 goes forwards  (BLTouch)
M569 P5 S1							; physical drive 5 goes forwards (Front Z-motor)
M569 P6 S1							; physical drive 6 goes forwards (Left Z-motor)

;Leadscrew locations
M671 X262:-85:262  Y390:180.5:-70 S25  ;Front left, Rear Left, Right  S7.5 is the max correction - measure your own offsets, to the bolt for the yoke of each leadscrew

; Endstops
M574 X1 S1 P"!xstop"                            ; configure active-high endstop for low end on X via pin xstop
M574 Y1 S1 P"!ystop"                            ; configure active-high endstop for low end on Y via pin ystop
M574 Z0                                         ; configure Z-probe as endstop
M564 H0											; Limit Axes
M557 X80:300 Y30:295 P7:7                       ; Define mesh probing grid              

M350 X16 Y16 Z16 E16 I1	                        ; set 16x microstepping for axes& extruder, with interpolation
M906 X1000 Y1000 Z1000 E1000 I30	            ; Set motor currents (mA)
M201 X3000 Y3000 Z100 E1500                     ; Accelerations (mm/s^2)
M203 X24000 Y24000 Z700 E3600                  	; Maximum speeds (mm/min)
M566 X800 Y800 Z80 E800                         ; Maximum jerk speeds mm/minute
M92 X80.00 Y80.00 Z1600.00 E98.00               ; set steps per mm

; Axis Limits
M208 X300 Y345 Z375                             ; set axis maxima and high homing switch positions (adjust to suit your machine)
M208 X0 Y0 Z-1.0 S1                             ; set axis minima and low homing switch positions (adjust to make X=0 and Y=0 the edges of the bed)


; Thermistors
M308 S0 P"bedtemp" Y"thermistor" A"bed_temp" T100000 B3950 R4700 H0 L0 		;_RRF3_ Bed thermistor, connected to bedtemp on Duet2

;If you have a Slice Engineering thermistor, comment out the next line
M308 S1 P"e0temp" Y"thermistor" A"Extruder_temp" T100000 B4725 R4700 C7.06e-8 H0 L0	;_RRF3_ duet3 e3d

;Additional Temperatures/heater settings
M308 S3 Y"mcu-temp" A"MCU" 						; configure sensor 3 as thermistor on pin e1temp for left stepper
M950 H0 C"bedheat" Q25000 T0				    ;_RRF3_ define Bed heater is on bedheat
M950 H1 C"e0heat" T1				            ;_RRF3_ define Hotend heater is on e0heat
M140 H0 										; Configure bed heater
M143 H0 S120	


M307 H0 R1.099 C332.207:332.207 D1.97 S0.50 V12.2 B0 ; Bed Heater
M307 H1 R2.473 C159.682:159.682 D6.35 S1.00 V12.1 B0 ; Nozzle heater
M570 H0 P5 S180 T10								; detect bed heater fault after 180 seconds 
M570 H1 P5 S180	T15								; detect end heater fault after 180 seconds 
M143 S285										; Max hot end temperature

; Fans
M950 F1 C"fan0" Q25000                          ; create fan 1 on pin fan1 and set its frequency (Part Cooling)
M106 P1 S0.5 H-1 T45 C"Part_Cooling_Fan"        ; set fan 1 value. Thermostatic control is turned on

; Tool definitions
M563 P0 S"E3Dv6" D0 H1 F1                      	; Define tool 0
G10 P0 S0 R0                        			; Set tool 0 operating and standby temperatures

;BLTouch - comment out the following 3 lines if using a IR Probe
M558 P9 C"^zprobe.in" H4 R1 F600 T2400 A4 S0.02                  	; _RRF3_ BLTouch connected to Z probe IN pin
M950 S3 C"duex.pwm1"				                                ; _RRF3_ Define BLTouch Servo (S0) on duet pwm1
G31 X31.225 Y10.5 Z2.6 P25                                          ; Customize your offsets appropriately - do a paper test, and put the probed value in the Z value here


T0					; select first hot end
M501				; Read stored parameters

@o_lampe It's not working completely as intended. Sure I can set the speeds to be different but what's happening is that I'm able to get the fast XY "T4000 "moves and when the bed lift to the probe it uses the good feed speed F600. But then when the bed pulls away from the probe, it uses the XY speed again.

M558 P9 C"^zprobe.in" H6 R0.1 F600 T4000 A4 S0.02 

@o_lampe Thanks, I'll try that.

@o_lampe Thanks this solved it, I can't believe I did not think of that. My only question is, how to do something similar for mesh conpensation moves? I came across M558, but I have firmware 3 and I don't think I can use that.

I have a DIY CoreXY printer with 3 independent Z-axes. My Z lead-screws are M10, and they move slower than XY axis. Thus, in my M558 command, I had to define a speed F500 so that the Z axis doesn't dive faster than the hardware can handle. But that also means that my XY moves are super slow. Is there a way to set different speeds for Z vs XY moves? Below is my bed.g file.
M561
G28 ; home all axis
G30 P0 X300 Y290 Z-9999
G30 P1 X76 Y162.5 Z-9999
G30 P2 X300 Y30 Z-9999 S3
G1 X0 Y0 Z10 F2400 ; Move extruder back to corner

@jens55
Thanks, I'll dig around and rewire to try that.

I have a bunch of 24v fans that I want to use as a part cooling fan.
I tried using a buck converter to step up the voltage but it would not work. when I connect the converter alone to the 12v pwm, I can measure 24V out. but when I connect the fan, that drops to 1.8v. Is it even possible to do this in any way?