Ajax error after bed probe

bendy

Hi All,

When ever I run a bed probe (via G32 S2), once the probe completes, I get an AJAX error on the web interface and am unable to reconnect.

The error I get is:

[[language]]
Communication Error

An AJAX error has been reported, so the current session has been terminated.

Please check if your printer is still on and try to connect again.

Error reason: SyntaxError: Unexpected token a in JSON at position 65

```I am running 1.16 firmware with 1.13 web interface.

If I restart the controller, I can get back in.

Any ideas?

bendy

Note that this issue also occurs after using the "Auto Bed Correction" button from the Machine control page (which I am guessing just does some flavour of a G32 in the background)

Sniffle

the "Auto Bed Correction" enacts your Bed.g file so yes it does whatever you set it to.

StephenRC

@bendy:

Note that this issue also occurs after using the "Auto Bed Correction" button from the Machine control page (which I am guessing just does some flavour of a G32 in the background)

Check your bed.g file, there may be a bad gcode in there.

dc42

What we really need to see is the json message that provoked the error. Can you provide a Wireshark pcap file?

bendy

Here are the contents of my bed.g -

[[language]]
; bed.g
; called to perform automatic bed compensation via G32
;
; generated by RepRapFirmware Configuration Tool on Sun Nov 27 2016 09:59:20 GMT+1100 (AEDT)

; Clear any bed transform
M561

; Probe the bed at 4 points
G30 P0 X75 Y110 H0 Z-99999
G30 P1 X345 Y110 H0 Z-99999
G30 P2 X345 Y355 H0 Z-99999
G30 P3 X75 Y355 H0 Z-99999 S

I will get a packet trace done later on and post it up.

bendy

Quick update - I have manually executed the commands in bed.g and the first 4 commands work fine, it is only when the last line -

[[language]]
G30 P3 X75 Y355 H0 Z-99999 S

```Is executed that the system crashes. The probe works as expected but I am guessing when the calculation of the Z plane is done it is causing an issue. I have tried it with both S and S4 with the same result. Packet trace still to come.

Jackal

Your probe points have to been in a clockwise order, starting nearest to 0,0

bendy

OK, Another update - the Ajax error is masking some other error here related to z plane calculation by the looks of it - If I run the probe for 3 points it works fine, it only fails if I do it for 4 points (am running a cartesian setup so 4 points should be supported?)

Qdeathstar

@Jackal:

Your probe points have to been in a clockwise order, starting nearest to 0,0

Is that true? DC42's calibration calculator suggestions points starting at the z-tower, then moving clockwise moving towards 0,0 with the last probe being 0,0…. should i rearrange the probes?

lignumaqua

Your sequence is anti-clockwise instead of clockwise, I suggest trying:

G30 P0 X75 Y110 Z-99999		; Bottom Left
G30 P1 X75 Y355 Z-99999		; Top Left
G30 P2 X345 Y355 Z-99999	; Top Right
G30 P3 X345 Y110 Z-99999	; Bottom Right
G30 P4 X210 Y232 Z-99999 S0	; centre, and calibrate compensation

Jackal

@Qdeathstar:

@Jackal:

Your probe points have to been in a clockwise order, starting nearest to 0,0

Is that true? DC42's calibration calculator suggestions points starting at the z-tower, then moving clockwise moving towards 0,0 with the last probe being 0,0…. should i rearrange the probes?

The calibration calculator is for Deltas. For Cartesian machines, you need them in clockwise order.
I stumbled upon a similar discussion before, you can find them by searching 'clockwise' in the forum.

bendy

OK, looks like going clockwise fixed it