Servo Driven 4th Axis for CNC Mill

[Pete.]

Pretty cool Andrew.

Lot of effort to ID-mark your stock though

[awemawson]

Well Pete what got me re-started on this was working out how to engrave a calibrated collar for the infeed screw of my Dormer 108 drill grinder. Then I remembered that I'd never sorted the Post Procesor for the 4 th axis 

[Pete.]

Fix a tool to fix a tool - Doc Smith would have been proud!

[philf]

Andrew,

All would have been a lot easier if your 4th axis were to accept g-code.

I wrecked my diamond drag engraver by forgetting to set Z and running out of spring travel.

I ground a 120 degree point on a broken 4mm carbide end mill and have used it to do a lot of engraving on stainless steel and anodised aluminium with no signs of wear. I'm not bothering replacing the diamond now. I have made a 90 degree one to try on harder material.

For those running a 4th axis on a machine which understands g-code there's a wonderful free utility called g-code ripper which can convert flat (x&y) g-code to wrapped around a diameter and swapping either x or y for the 4th axis. It can also split g-code into smaller sections if you want to machine parts bigger than the travel of your machine as well as rotate or scale.

Phil.

[awemawson]

Phil, it will accept G code just by altering one parameter in the TNC355. But as the good post processors for the TNC controllers generate the Heidenhain conversational code I tend to use that

[gasmeter]

Hi Andrew

I replied to this post even though its old .

I'm looking at adding a 4th axis to my TNC355 (Denford quatromill).
Silly question

The fourth axis that's on input X4.. In which case its a sine wave encoder ?
X5 is quadrature square wave .

What encoder did you use on your system , I see it has to be 180000 or 36000.

I'm in s wales by the way and do get to sussex from time to time

Peter

[awemawson]

Peter, if you are passing do pop in and have a look.

My encoder came from the late lamented good chap John Stevenson, it was a sine wave Heidenhain off a Bridgeport Interact and I'll see if I can find any numbers - hang on . . . .

. . . worryingly my entire directory not only describing this build, but also all my data on the Beaver Partmaster for which it was built seems to have vanished from my hard drive 

. . .but all is not lost, as it is detailed and pictured in reply #2 of this thread - it's a ROD 450B 

(Later Note: for some odd reason the entire 'lost' Beaver Partmaster directory had become a sub-directory off my Ingersoll Rand Drier directory so just needed moving back - but why did that happen )

[gasmeter]

Hi Andrew
I would love to next time I'm in sussex . My mother in law lives in Blackboys near uckfield.

Do you have the numbers after the Rod 450B as they say what the number of pulses is. I find the Heidenhain numbering system confusing .

Pete

[awemawson]

If you read the thread I'm sure that there was long discussion regarding counts per rev and belt ratios, however I do have a picture of the encoder (also embedded in this thread) of the encoder with numbers - here it is again:

[gasmeter]

Thanks Andrew

I can't see the number of pulses on that picture but I  just order a ROD 456 1800 02S09-04 Sinusoidal Rotary Encoder.
This will be a 1800 pulse (i'm assuming 1800 is the pulse per rev .. Again not sure of their numbering system) so I can gear it to give 36000, as it was a reasonable price.


The RI+/- signal confuses me and I can't find any info anywhere .
Is it a once per rev ? And needs selective gating so it can only appear during a specific angle range ?

I think I same something about gating in th thread so I will work back through it .

Peter

[gasmeter]

Hi all

I also pulled out all the param settings for others for the 355

all the tuning params etc are  there.

Peter

[gasmeter]

I worked through all the relevant settings

summary

Andrews settings
MP 3 : 2500 - 4th axis rapid

MP 7 : 2000 - Manual feed 4th axis

MP 11 : 1800 - approach speed approaching ref marks 4th axis

MP 15 : 2  - sig eval input x4 x x interpolation

MP 19 : 0 Traverse direction approaching marks ( 0 positive)

MP 23 : axis IV ref pulse inhibit active/inactive
 on connector X10
0 - Ref pulse inhibit
2 - Ref pulse inhibit active

A ref mark is traversed every full rotation of the sensor .
If there are multiple pulses the pulse inhibit can be controlled by a cam switch or similar.


MP 27 : 0 Polarity Positive with traverse dir

MP 31 : 0 Integral factor (set to 0 ?)

MP 35 : 1.000 Differential factor set to 0 ?


MP 39 : +0.000 Backlash compensation

MP 43 : +0.000 Linear compensation factor (keep to 0)

MP 50 : +30000.000 software limit switch range +
MP 51 : -30000.000 software limit switch range -

MP 59 : 14 ZYX IV ref mark approach seq



MP 72 : 0 P all axis set to non inhibit


MP 90 : 0 axis designation of 4th axis 0 = A

MP 166 : 500 Feed speed for 4th axis for param 126-156

MP 180 : 2.000 4th axis KV factor value 0.1-10
 
MP 189 : +0.000 datum for 4th axis position block for M92



MP 256 : 0 allocation of encoder inpit to 4th axis
0- Standard allocation
4- Encoder x4
 
MP 299 : 0.250 4th axis accel

MP 302 : 9.000 4th axis rapid traverse voltage

Partsmaster 355 option param 158
Notes on setting to enable 4th axis
param 158 needs to be 16



1 = air conditioner
2 = flashing lamp operated with m22 / m23
4 = flashing lamp on when program run is stopped
8 = swarf conveyor
16 = 4th axis present
32 = feedrate effective in toolchange
64 = high speed spindle
128 = override spindle stall
256 = no toolchanger


add up value & enter into param 158

On my Denford Quatromill it already has 128 set for override spindle stall
so its 128 + 16 = 144 to enable 4th axis / 128 disable 4th axis

It appears a 1800 count.. Count is strange as its a sin/cos but then with the basic interpolation of x5 on the TNC355 then a 4 fold in the signal evaluation parame gives the full 36000 with no gearing.

As to enable / disable 4th axis I have yet to test bit it looks like its

either MP72  8 = IV axis inhibited / clear to enable
         MP158 +16 enables 4th axis (on denford quatro it would be 128 = No Ivth axis .. 144 Ivth axis enabled)

I will update when I test.

[ddmckee54]

"Later Note: for some odd reason the entire 'lost' Beaver Partmaster directory had become a sub-directory off my Ingersoll Rand Drier directory so just needed moving back - but why did that happen"

Andrew:

If you have a tremor like I do, your fingers probably twitched on the mouse as you were scrolling past that directory.  I lost my entire Machining Favorites folder for several days, and that's the only way I can figure it could happen.  I about had a heart attack the next time I tried to go on-line to one of those Favorites.  Like you I finally found the folder as a sub-folder in one of the adjacent Favorites folders.

Don

[awemawson]

. . . a tremor . . . Well Don only after a few too many Harvey's Best Bitter or Glen Fiddich's 

[gasmeter]

Hi Andrew could I ask a couple of questions

1/ What did you do for pulse inhibit etc ? Did you do it the same way as hazle ? Simple switch arrangement.

2/ Is there a way to tell the machine that the 4th axis isn't present without changing parameters ?

I almost have all the information sorted now .

Peter

[awemawson]

Peter I'm a bit tied up at the moment but a quick glance tells me that's covered round about reply #150 to this thread

[gasmeter]

OK to answer the inhibit part of the question.

MP23 is for axis IV on connector X10
0 - Ref pulse inhibit
2 - Ref pulse inhibit active

A ref mark is traversed every full rotation of the sensor .
If there are multiple pulses the pulse inhibit can be controlled by a cam switch or similar.

I'm lucky enough to not need that as my encoder will give 1 ref pulse  per rev, so I don't need pulse inhibit.


Peter

[gasmeter]

Hi Andrew

I added to reply #156, so I hope that quote contains most of the info in one place to help others .

Peter