Brother TC-225 / TC-229 adventure!

PCW wrote:

It sounds like something basic is wrong with your LinuxCNC setup
Its unlikely there is a major error in the TP that others have not noticed.

Can you post your current hal and ini files?

ill see if I can get that tomorrow night when I go to the shop.

I imagine its possible I changed a setting that's causing this, but ive not changed anything relating to that I don't think.

its almost like a threshold. if I run a circle (2 g3's) at some speeds and radii it respects g64, and at others is stops between arcs.

I wish there were more examples online of successful tuning of analogue servos. theres only a handful, and they don't tell you anything - or are on low performance machines.

gcofieldd wrote:

I have not shipped the machine yet. I am waiting for my employer to setup their warehouse so I can move it in. It should happen in a couple of months. The industrial controller I was looking at is a Delta NC311. When combined with the A2 style drives and motors it is supposed to auto tune, which is appealing. I have not heard much about the controller, so I am a little worried about support. I was hoping since it setup for drill and tap centers it would include some optimization of the tool change process.

Either way, I plan on replacing the motors and drives. I am just not sure what I am going to use for a controller.
LinuxCNC is appealing because of its flexibility and cost, but I don't want to spend a huge amount of time tinkering.

ive looked at the delta control a bit. I am seriously considering the asd a2r servo drives for the tc229 but they wont autotune the stock motros

theres also the siemens controls, which at least here in Toronto have good support. they don't cost a lot relative to equivalent controls, but all of them are "kinda bit bit too much to spend on a $1500 machine" so I am trying to exhaust all the cheap as free options first.

Note that with properly tuned velocity mode servos, pretty much all the "high performance" aspects
are controlled by the drives, not LinuxCNC, so major difficulty tuning velocity mode drives really points
some kind of setup issue.

PCW wrote:

Note that with properly tuned velocity mode servos, pretty much all the "high performance" aspects
are controlled by the drives, not LinuxCNC, so major difficulty tuning velocity mode drives really points
some kind of setup issue.

that sounds logical, (especially since we know the drives are fine under the brother control)

.... but whats is the setup problem!!!!



can you just come to my house and fix it? hahaha.

Well actually if I did stop by your house to try and debug this issue,
the very first thing I would do is look at the hal/ini files...

haha. sorry i havent been to the shop this week to get the files. maybe tonight, if i remember.

Hi,
I read through the whole thread. It was a lot of helpful information. Reading it earlier would have saved me some time. I am currently converting a Brother TC215 I bought two months ago. The TC215 is pretty similar to the TC225 that ihavenofish is converting. It has less travel and a smaller spindle (only 1.5kW with either 10krpm or 6krpm). Kind of the small brother of the TC255. They even share the same documentation.
I have started to reverse engineer the connections from the old control board to the relay board, which does all the switching.
Some of the pinout can be found in my github repo: github.com/kpufahl/tc215Retrofit

I have been able to move the first axis, but haven't done any extensive tuning yet. There was a lot of other work. The previous owner was lubricating but the tubes behind the grease fittings all broke off. So the grease didn't get very far. I still need to put in some new tubes. Similar to ihavenofish I found a lot of chips under the covers. I really don't now how they got there.

@ihavenofish I am quite interested in you config files. Will you share them? How far did you get with the ATC code?

oooh, sounds fun, i want to see where you get with servo tuning.

ive yet to get my files together to send to pcw after he looks at them to see if they are obviously wrong, ill send you them (or post them here)

ive been a bit distracted, i have to figure out where to move my shop as they are ripping the building down for condos. yay condos! i mean, we need more of those for sure cause only 65% of them in the city are empty held by flippers.

sigh.

oh, that's bad news!

Have you made a list of the pinout of the relay board? I am have not yet found all remaining signals. Also I am still lacking the alarm signals from both the monitoring circuit and the servo drivers.

your relay board will be completely different from mine unfortunately.

but: the way to check the relays is quite simple. the relays are labelled, and you actually only need i think 2 to run the machine. one relay is for servo enabling, which was labelled SV (ill have to double check) you can trace its signal input (pin with the diode) back to its connector. the other relay you need is the z brake, to release it, which has to be done AFTER servo on. this one is likely labelled BRK.

then you have the ATC relays. one for forward/reverse selection, and one for run.
the relay labelled M8 is for the coolant pump, which will of course become useful, but isnt mandatory for getting started.

the rest of the relays are for things that are not required, like alarm output (for flashing lights) M codes for customs things, external outputs for robots, powered doors, indexers, etc etc.

the alarm INPUT is different. the drives alarms are on each drive connector on mine (different drives than you have). the alarm for the safety unit goes straight into the motherboard, which i never connected though id like to cause its useful.

on the atc side, mine "works" in that i check its at atc home, read the encoder, spin the carousel and land on the desired tool in the shortest bidirectional movement with M6 Tx i used a m6 remap to a gcode subroutine for all of this.
i have not yet implemented z axis movement though, partly because my z never moved reliably due to tuning issues. partly because i cant figure out how to swap limits to allow it to have 2 zones of travel - 1 for atc, one for milling.