Retrofitting a 1986 Maho 400E

Do you also use your CNC mill just as a manual mill, jogging around to remove metal? Every time I do this, I screw up. You can seen in this latest video, at 12:00, I figure it is a simple light cut in aluminium, so I'll jog around. Result ... big screw up!


The next step up the laziness ladder is just MDI'ing around. For me this results in a screw up about 3/4 of the time. I have sent a few end mills over the Jordan with doing that.

Moving further along the work flow path, if I sit down and write a program using conditional looping, the screw up rate seems to drop to about 25%, as we saw in my Lathe video three with the vee block, cutting at G0 instead of G01.

Tool paths generated from CAM have so far been most successful, but my natural abhorrence for work draws me to avoid CAD modelling parts and then creating a CAM path if it is only a simple part.
Mark

Hello to all, Mark I thing I have found solution to the neutral speed problem. You can either delete the “{ 0, 4 }, /* neutral 0 1 0 0 */” line in the lookup table on the mh400e_common.h file or ghange the { 0, 4 }, to { 1, 4 }, .
if the first solution wont work, in the second solution the gearbox will go to neutral only if you command 1 rpm.
John

Or make it { 0, 0 }, /* neutral 0 0 0 0 */ so it will not care for the state of the gearbox and it wil not make changes to it

static pair_t mh400e_gears[] =
{ /* rpm bitmask msb 11 10 9 8 7 6 5 4 3 2 1 0 lsb */
{ 0, 0 }, /* neutral 0 0 0 0 */
{ 80, 1097 }, /* 0 1 0 0 0 1 0 0 1 0 0 1 */
{ 100, 2377 }, /* 1 0 0 1 0 1 0 0 1 0 0 1 */
{ 125, 585 }, /* 1 0 1 0 0 1 0 0 1 0 0 1 */
{ 160, 1177 }, /* 0 1 0 0 1 0 0 1 1 0 0 1 */
{ 200, 2457 }, /* 1 0 0 1 1 0 0 1 1 0 0 1 */
{ 250, 665 }, /* 0 0 1 0 1 0 0 1 1 0 0 1 */
{ 315, 1065 }, /* 0 1 0 0 0 0 1 0 1 0 0 1 */
{ 400, 2345 }, /* 1 0 0 1 0 0 1 0 1 0 0 1 */
{ 500, 553 }, /* 0 0 1 0 0 0 1 0 1 0 0 1 */
{ 630, 1090 }, /* 0 1 0 0 0 1 0 0 0 0 1 0 */
{ 800, 2370 }, /* 1 0 0 1 0 1 0 0 0 0 1 0 */
{ 1000, 578 }, /* 0 0 1 0 0 1 0 0 0 0 1 0 */
{ 1250, 1170 }, /* 0 1 0 0 1 0 0 1 0 0 1 0 */
{ 1600, 2450 }, /* 1 0 0 1 1 0 0 1 0 0 1 0 */
{ 2000, 658 }, /* 0 0 1 0 1 0 0 1 0 0 1 0 */
{ 2500, 1058 }, /* 0 1 0 0 0 0 1 0 0 0 1 0 */
{ 3150, 2338 }, /* 1 0 0 1 0 0 1 0 0 0 1 0 */
{ 4000, 546 } /* 0 0 1 0 0 0 1 0 0 0 1 0 */
};

Thanks for looking into that. What we have been planning was to switch the control logic. Currently the gearbox.comp requests that LinuxCNC switches the spindle on and off, whereas we want to give the Gearbox.comp direct control of the spindle relays, and route LinuxCNC spindle requests to the comp. The neutral issue is something were were going to address at the same time. For these changes I am at the mercy of my nocturnal IT vampire though.

Mark

Hi,
I also used the maho gearbox component and was dissatisfied with the M5 gear shift.
So just rewrite the line above and should it work without a problem?
Thank you for your work!
Greetings from the Czech republic!

I thing it will work, disconnect the (gearbox) motors and check while you make the first attempt so you wont crush the gearbox.
John

There is also a Simulator in github, so you should first run changes to the gearbox comp through that ti test them.
Mark

I wrote a short program (attached) to surface and trim the end of piece of stock, but made a typo in the fourth to last line.instead of
When I tried to load the file, LinuxCNC tried to verify it, and got in an infinite loop. Is there are more elegant way to recover from this than hard rebooting the machine?
Mark