Advanced Search

Hello everyone,As I will soon be starting the conversion of an amazing milling machine myself, I’ve been looking for information about it.
Since there is very little available about the materials and components used, I’m sharing the information I’ve gathered, in the hope that it can help others as well.It’s a machine that uses Bosch equipment, and I’ve uploaded the complete manual, schematics, and everything I have on it.

 Kind regards,
Ronny 

Hi, just wondering if you ever made progress on this? Also looking at EtherCAT drives from that manufacturer.

Are the QA, QB, IDX pins on the 7i85 equivalent to the A, B, Z signal in line driver encoders?

Hello Tommylight, yes this advices fixed my problem the THC is now working how it should.
Thanks once more and Happy new year!

Hello Tommylight thanks for your quick reply. Need to check it out, let you know. BR

This is the download link

drive.google.com/file/d/1pmV587gXFEvznQB...00x/view?usp=sharing

Something has changed on linuxcnc between then and now, but not much.
In particular, the function for using GPIOs which was previously hal_pi_gpio and is now hal_gpio
linuxcnc.org/docs/devel/html/drivers/hal_gpio.html
forum.linuxcnc.org/24-hal-components/504...-with-raspberry-pi-4

The image of RaspiOS with Linuxcnc is available in this link

forum.linuxcnc.org/9-installing-linuxcnc...pi-5?start=40#303752

Let me know if you need further guidance

I would think (in general) encoder resolution, axis acceleration and trajectory servo update rate all have potential to improve the tracking.

but changing the speed significantly is not going to track in the same place with the current code. I wouldn't think it would be hard to code absolute position tracking, but I bet it has some corner cases.

For your homing issue, set a home_offset to move the axis off the switch when homing otherwise the home/limit switch will be homed when homing of that axis is completed

Metric threads are pretty simple. My guess it would be a metric fine thread 17mm x 1.5mm pitch
So if you don't have a thread gauge, measure 10 threads and see if it is 15mm long.
Then you just need  to buy a 17x1.5mm tap.
to size the drill for a metric tap, just deduct the pitch from the diameter. In this case, 17-1.5 = 15.5mm

So my laptop running win 11 won’t do it correctly

My laptop running win 10 was totally different acting when downloading the file.
I already had etcher installed.
Worked flawleslly .

B

Height Per Volt?
Leave at default 0.1 or whatever the default is, i never had to change that.
VAD at 65% or 70% should be fine, 90% is to much
THC delay i usually have at 0.1-0.2, only for slow cutting speeds use more if needed
And disable "use auto volts" until you get the machine working properly.
From your pictures, i would venture a guess at Z axis acceleration being way to low, or scaling being way off from reality.
If all of the above is set and verified to be OK, then set the P value to 8 or 7 or even lower, but if you end up any lower than 5, something is terribly wrong.

They require a lot of rework to get inputs. Swapping out chips, add jumpers, messing with power supply voltages, can be done but fiddly.

Best starting with one of the more common boards discussed on this thread.

Just keep this is mind that it may end up costing close to the cost of original Mesa board (less postage if you are outside the locality of a local supplier). Most of these projects turn out that way.

BTW Linuxcnc-RIO is a great project, well supported and actively developed.

Could you plot the encoder position and velocity to see if the error is from the encoder side or LinuxCNC?

The jagginess may be from encoder wheel inaccuracies and inherent velocity quantization
errors from the 1 ms time stamp of the software encoder (this would be improved by using the hardware encoder)