Screwcutting

welderfabrod wrote:

Guys I tried the G76 and it works to a point, but there are a few odd movements so I'll give the short prog that I used to make sure I've got the G76 correct. This is for an 8mm thread 12mm long.

G21 G90 G40
G00 X5 Z1
M03
G76 P1.25 Z-12 I-1 J0.05 R1.0 K0.7668 Q30 H3
G00 X5 Z1
M05
M02
I put the M03 and M05 in but have'nt tried it without them yet.
<snip>
Other odd movements.
1)Cutter makes a X+ movement before moving to the start of the first cut. ( maybe I've misunderstood the G76 here)
2) As the cutter travels back along the drive line it moves a small amount in the X- direction also. I may not have noticed this if it was'nt for the fact that on the X axis I have a graduated handle on the rear of the stepper motor.
3) The cutter is entering and leaving the cut in a circular moition.

Are you talking about motion besides the G00 X5 Z1.
When I thread I run at 500 rpm, so I have never seen the problems at slower speeds.
Your code looks correct for an external thread, but the movement of the X axis sounds like it is trying to cut an internal thread. Is it possible that your X axis is moving the opposite direction from what you command? You probably already know this but, Zero for the X-axis should be at the center of the material being cut. When you are at X=0, X+ should move the cutter (and cutter tip) away from center. When you are at an X=positive value, X- should move the cutter tip towards the center of the material.

I would be very grateful for that Alan as I have a Denford Orac waiting to be upgraded when I purchased it the inverter was'nt working and one stepper driver was defunct. I replaced the inverter and this is working fine with a pot connected to its own 0-10v supply. I've also purchased two stepper drivers from Motion Control but have'nt installed them yet as I still need to aquire a breakout board and a relay board. I have actually managed to get a digital readout onto the Axis screen but have'nt yet got it to actually display the revs from the encoder.

Rod.

Rod,

My configs are in 'inch" mode but you should be able to make the appropriate changes. They were zipped on my Mac with its native archive function. If you can't open them let me know, I could re-zip them with another program.

Try adding L0 to the threading line (no taper) and changing R1.0 to R2.0 (constant volume). The use of constant volume can help keep the spindle from slowing down if that is a problem ( the trade off is more threading passes).
G76 P1.25 Z-12 I-1 J0.05 R2.0 K0.7668 Q30 H3 L0
Or if you would like a short taper on the starting end, E2 L1 (taper over the starting 2mm).
G76 P1.25 Z-12 I-1 J0.05 R2.0 K0.7668 Q30 H3 E2 L1

Let us know if the problem persists. (Take two aspirin and call me in the morning.)

Alan

John I do both, I change pulley ratios to the speed range I need then adjust the revs with speed control pot. This is part of the original Emco control.

Alan the material is already 8mm dia. so for the X axis I start the spindle touch off on the material and set this to 4mm. For the Z axis I basicaly use two methods, if as in this case its just a thread on a piece of bar I bring the point of the cutter in line with the end of the bar by eye and zero Z there, and thats near enough. If however I am threading up to a shoulder I bring the left side of the tool up to the end of the bar with a bit of paper in between which gives 3 or 4 thou. clearance at the shoulder. I think thats all pretty normal.
In this case I'm positioning the cutter at X5 Z2 before starting the programme so although G00 X5 Z2 is in the programme at the start I should see no movement, but what actually happens is on reaching the G76 the cutter first makes a move in X+ beyond the drive line (I think by the thread depth but am not sure as it happens to quick.) then moves to the first pass position and starts the first cut. I changed Z1 to Z2 because it appeared that 1mm from the end of the material was not enough for a 1.25 pitch. I also added L0 and R2 as suggested and changed J0.05 to J0.1.

Guy's the G76 does'nt seem to operate as I was expecting either. I thought that all the rapids would travel down the same drive line i.e. always at 1mm from the material. However what is happening is that after the first cutting pass the cutter retracts with a semi-circular path beyond the drive line, by I think the thread depth then rapids back towards Z2 on a slightly tappered path in X- this can clearly be seen on the graduated handle on the end of my stepper motor, this is probably by J0.1. If I zoom in on graphics its visible there also. It then makes a 90deg circular path to Z2 then a straight move in X- to start the next cut. this motion occurs on all subsequent passes. So what you have on the screen is not only a set of cut passes in X- but also a set of rapids in X+ beyond the drive line. This seems most strange to me and does'nt follow what the manual appears to be saying i.e. 3) A spindle-syncronized Z feed along the thread. 4) A traverse to the original X. (I take this to mean in my case the intial start point X5 after each pass).

The next thing is the software does'nt seem to be syncronizing the Z axis with the spindle speed correctly, while it appears to be seeing the index pulse ok if I run at 500revs as Alan does the thread pitch is too small, if I reduce the spindle speed the thread pitch increases until I get to a point where the pitch is to large, reduce it much further and I reach a point where I get the stop starts in the Z axis again. Surely it should syncronize at whatever spindle speed I set. I feel that as all this odd behavior occurs in graphics whether or not the steppers are enabled this has to be a software issue. Feel free to shoot me down if I'm wrong.

Also what I did is to boot another PC from the live CD and run G76.ngc in sim/lathe.ini and the above motion is there also. But it does'nt seem right to me. This version is Ubuntu 8.04 Emc2 2.2.7 this is the version I am running.

Alan I'm fresh out of Asprin!!!!

Rod

acondit wrote:

welderfabrod wrote:

Guys I tried the G76 and it works to a point, but there are a few odd movements so I'll give the short prog that I used to make sure I've got the G76 correct. This is for an 8mm thread 12mm long.

G21 G90 G40
G00 X5 Z1
M03
G76 P1.25 Z-12 I-1 J0.05 R1.0 K0.7668 Q30 H3
G00 X5 Z1
M05
M02
I put the M03 and M05 in but have'nt tried it without them yet.
<snip>
Other odd movements.
1)Cutter makes a X+ movement before moving to the start of the first cut. ( maybe I've misunderstood the G76 here)
2) As the cutter travels back along the drive line it moves a small amount in the X- direction also. I may not have noticed this if it was'nt for the fact that on the X axis I have a graduated handle on the rear of the stepper motor.
3) The cutter is entering and leaving the cut in a circular moition.

Are you talking about motion besides the G00 X5 Z1.
When I thread I run at 500 rpm, so I have never seen the problems at slower speeds.
Your code looks correct for an external thread, but the movement of the X axis sounds like it is trying to cut an internal thread. Is it possible that your X axis is moving the opposite direction from what you command? You probably already know this but, Zero for the X-axis should be at the center of the material being cut. When you are at X=0, X+ should move the cutter (and cutter tip) away from center. When you are at an X=positive value, X- should move the cutter tip towards the center of the material.

I would be very grateful for that Alan as I have a Denford Orac waiting to be upgraded when I purchased it the inverter was'nt working and one stepper driver was defunct. I replaced the inverter and this is working fine with a pot connected to its own 0-10v supply. I've also purchased two stepper drivers from Motion Control but have'nt installed them yet as I still need to aquire a breakout board and a relay board. I have actually managed to get a digital readout onto the Axis screen but have'nt yet got it to actually display the revs from the encoder.

Rod.

Rod,

My configs are in 'inch" mode but you should be able to make the appropriate changes. They were zipped on my Mac with its native archive function. If you can't open them let me know, I could re-zip them with another program.

Try adding L0 to the threading line (no taper) and changing R1.0 to R2.0 (constant volume). The use of constant volume can help keep the spindle from slowing down if that is a problem ( the trade off is more threading passes).
G76 P1.25 Z-12 I-1 J0.05 R2.0 K0.7668 Q30 H3 L0
Or if you would like a short taper on the starting end, E2 L1 (taper over the starting 2mm).
G76 P1.25 Z-12 I-1 J0.05 R2.0 K0.7668 Q30 H3 E2 L1

Let us know if the problem persists. (Take two aspirin and call me in the morning.)

Alan

File Attachment:

File Name: EMC2_Ini.zip
File Size:22 KB

Alan thanks for those files I uzipped them ok and opened them with Open Office which I have on my Windows box indoors.

Rod.

welderfabrod wrote:

John I do both, I change pulley ratios to the speed range I need then adjust the revs with speed control pot. This is part of the original Emco control.

Alan the material is already 8mm dia. so for the X axis I start the spindle touch off on the material and set this to 4mm. For the Z axis I basicaly use two methods, if as in this case its just a thread on a piece of bar I bring the point of the cutter in line with the end of the bar by eye and zero Z there, and thats near enough. If however I am threading up to a shoulder I bring the left side of the tool up to the end of the bar with a bit of paper in between which gives 3 or 4 thou. clearance at the shoulder. I think thats all pretty normal.
In this case I'm positioning the cutter at X5 Z2 before starting the programme so although G00 X5 Z2 is in the programme at the start I should see no movement, but what actually happens is on reaching the G76 the cutter first makes a move in X+ beyond the drive line (I think by the thread depth but am not sure as it happens to quick.) then moves to the first pass position and starts the first cut. I changed Z1 to Z2 because it appeared that 1mm from the end of the material was not enough for a 1.25 pitch. I also added L0 and R2 as suggested and changed J0.05 to J0.1.

The next thing is the software does'nt seem to be syncronizing the Z axis with the spindle speed correctly, while it appears to be seeing the index pulse ok if I run at 500revs as Alan does the thread pitch is too small, if I reduce the spindle speed the thread pitch increases until I get to a point where the pitch is to large, reduce it much further and I reach a point where I get the stop starts in the Z axis again. Surely it should syncronize at whatever spindle speed I set. I feel that as all this odd behavior occurs in graphics whether or not the steppers are enabled this has to be a software issue. Feel free to shoot me down if I'm wrong.

Also what I did is to boot another PC from the live CD and run G76.ngc in sim/lathe.ini and the above motion is there also. But it doesn't seem right to me. This version is Ubuntu 8.04 Emc2 2.2.7 this is the version I am running.

Alan I'm fresh out of Asprin!!!!

Rod

Rod,

I would suggest updating EMC2 to the current version. I don't remember for sure but it seems like there were some bugs in lathe threading that were fixed in 2.3 or 2.3.1.

You also have to be careful in touching off with the threading tool. It should have a radius (small but there none the less) on the tip. So if you touch off the tip, the sides won't be where they are expected by the software and it will cut deeper than you want. You almost have to use the tool table to get it right. You have to cut an initial thread and see how much too deep it cuts ((Desired diameter - actual diameter) divided by 2). Save this offset and when you set up your tool table, remember to include this offset as part of your tool offset. For example my threading tool sticks out farther than my turning tool, so my offset is computed from the extra length and the distance from the radius tip to where the point would be if it had a full 60 degree point. I hope this makes sense.

Alan

Deleted-- I used a special character for degrees in the last post and it then refused to display.

Yes Alan your quite right, but surely the software has no idea what my tool tip radius is unless I define it in the tool table, which I have'nt as most of the time the thread pitch's that I use are quite small because I am often more concerned with thread depth and calculate my pitch accordingly much of the time my thread depth is 0.5mm so the point radius is not really worth worrying about and a rub with a stone on the cutter is adequate. (metric for a bolt Rad.= 0.1443 x Pitch the nut has no radius just a flat at the core= pitch/8). Also this should have no effect on the pitch I cut. When I say the pitch is too small at 500revs. it is less than 1mm instead of 1.25mm. This is just a test to get syncronization right.
And as I said when running the G76 sample from the live CD in simulation the odd toolpath is still there. I'll download 2.3.1 tomorrow and have a try with that as you suggest.
I'm sure we'll get there in the end, just wish I had more knowledge of hal though. Rod

Are you sure you scale is correct for both axii? One of the ways to get incorrect pitch is to have the wrong scale set up. The other thing that can cause problems, is if you don't have enough acceleration for the distance away from the starting thread. Also on coarser pitches, there can be a problem with whether the z-axis actually can move fast enough to cut the thread at a given rpm.

Alan

Alan my scales are correct, I did think of acceleration when I first set up the encoders and tried G76 which is why I started at a low revs (150). Anyway I decided to bite the bullet and upgrade to 2.3. Nothing is ever that simple though is it. Went to the Installing off-line (32-bit systems) downloaded the Emc2 deb ok but when I tried to download the the two Ubuntu debs I.E. reports that the web pages cannot be found. So I had to download the live CD and have now installed that and this all appears to be running ok I hope now to get into the workshop this weekend and have a go with that.

Let you know how I get on in the next couple of days

Thanks again for persevering with me on this, I did think with the number of views this topic has had that we would have had a little more input from others as we can't be the only two people that have set up encoders for thread cutting.

Rod.