[CROSSHATCH]

First off, hi everyone

I'm reserving this post to update/edit with new questions. :idea:

Spindle:

I just purchased a floating tap because I am going to finally machine and induce tapping during the machine program, not after. Reading through the "beta manual", it mentions a couple of times that M4 (CCW) for Spindle is not enable in the stock configuration. Is this true? If so, what would be the reason and who has enabled the use of reverse spindle? Silly question I suppose most have CCW... but the manual states it a few times.

In regards to the M4 command and a floating tap, am I going to have issues with tapping, dwell delay and reverse IF I do not have the braking resistor? I feel like it should not be a concern with inducing dwell into the cycle.

I really wish I had the braking resistor (I'll hopefully be able to get it sooner than later).

R8 Collet

Collet question; I have a bunch of really nice ER32 straight shank tool holders with the goal of eventually using pneumatic drawbar capabilities for quick and repeatable tool changes.

Novakon sells an R8 collet for use with their toolholders and Tormach as well. What are the draw-backs of not using "that" R8 collet and using another 3/4" R8 collet? I suppose it comes down to the quality of the R8 collet; I have to check my collets but I'm pretty sure it's of that German grade. Thought I would ask everyone who has had the same idea.


It's 4th of July, I'm being rushed out of the house...reserved for later thoughts/questions.


Enjoy the day fellas:cheers:

-Jason

[Hirudin]

My Torus Pro turns counterclockwise when I execute an M4 command.

Here's a thread I started with similar questions a while back...
http://www.cnczone.com/forums/novako..._backward.html

By the way: I still haven't tapped with my mill. :/ I've done a lot of thread milling, but never automated traditional tapping. I too have a floating tap holder but I haven't even tried it yet - I'm also nervous about being able to stop the spindle quickly enough to reverse direction reliably. I bought myself a braking resistor to solve that problem, but it turns out I got a mill with an incompatible VFD. Grrrr! Apparently there's a replacement coming for me in the mail - last I heard about it was three weeks ago.

As for the R8 collet: the collets Novakon and Tormach sell are designed to be nearly flush with the spindle nose when tightened. I've never actually used a "normal" R8 collet, but I get the impression that they normally stick out an 1/8 of an inch or so. The collet sticking out might prevent the tool holders from making contact with the spindle nose.

See pg. 5 of this PDF from Tormach for a helpful image.

I have a couple Tormach collets and a single Novakon collet. The Tormachs seem to work fine but the single Novakon collet I have is too tight - TTS collet chucks will not fall free from the collet, even if it's completely loosened. Of course, maybe the one I received is an outlier. I wish I could find a company selling very high precision quick-change 0.75" R8 collets but so far I only know of these two.

[Novakon]

First off, hi everyone

I'm reserving this post to update/edit with new questions. :idea:

Spindle:

I just purchased a floating tap because I am going to finally machine and induce tapping during the machine program, not after. Reading through the "beta manual", it mentions a couple of times that M4 (CCW) for Spindle is not enable in the stock configuration. Is this true? If so, what would be the reason and who has enabled the use of reverse spindle? Silly question I suppose most have CCW... but the manual states it a few times.

In regards to the M4 command and a floating tap, am I going to have issues with tapping, dwell delay and reverse IF I do not have the braking resistor? I feel like it should not be a concern with inducing dwell into the cycle.

I really wish I had the braking resistor (I'll hopefully be able to get it sooner than later).

R8 Collet

Collet question; I have a bunch of really nice ER32 straight shank tool holders with the goal of eventually using pneumatic drawbar capabilities for quick and repeatable tool changes.

Novakon sells an R8 collet for use with their toolholders and Tormach as well. What are the draw-backs of not using "that" R8 collet and using another 3/4" R8 collet? I suppose it comes down to the quality of the R8 collet; I have to check my collets but I'm pretty sure it's of that German grade. Thought I would ask everyone who has had the same idea.


It's 4th of July, I'm being rushed out of the house...reserved for later thoughts/questions.


Enjoy the day fellas:cheers:

-Jason

The Torus PRO does have reversing capability. I will have to look at the beta manual again for the references about CW only. This may have been a carry over from the older NM-200 manuals. Thanks for the heads up.

The tapping head, as I understands it, has auto reverse with a torque clutch built inside. I am not sure if there is a need to reverse the spindle for a tapping head. Also, the speed at which you tap is fairly low and their may not be a reason to worry about a braking resistor at this time. It can easily be added though, and it is nice for dropping speed fast from 6,000 rpm.

I would highly suggest the new PDB before you invest too much into the TTS type tooling. With this being said, our PDB functions with R8 capture system and doesn't need the TTS tools for its use. This is why we selected this technology because it is so versatile. If you have an assortment of very good R8 tools, check this out first.

Regards,
Novakon Team

[MichaelHenry]

The Procunier and Tapmatic tapping heads turn the tap clockwise when feeding into the work and an internal clutch/gear arrangement reverses the the tap rotation when the tap is retracted from the work. The spindle rotates clockwise both while feeding and retracting. Hence, there is no need to reverse spindle direction when using these tapping heads. The tension-compression tapping head, on the other hand, has no way to internally reverse tap direction so the spindle direction must be reversed when the tap is retracted from the work. These types of heads usually have a significant amount of vertical float to allow some slack while the spindle is stopping and reversing. Until the spindle stops following a feed, the tap will continue turning and will therefore cut a few more threads after the spindle is commanded to stop and before it is commanded to reverse.

With the Tormach system (mill, newer VFD, and T-C head) I usually specify 500 rpm for a #4-40 tap and a 0.2-0.3 sec dwell time and that seems to work well enough for me. It can be a bit tricky to figure out the proper tap depth to specify on blind holes with this type of head as it is difficult to determine exactly how far the tap will keep going down before it reverses and starts retracting. The clutch-type can probably be used to tap depths a bit more precisely since you only need to reverse feed direction to stop tapping. One downside to the clutch type heads is that they generally take up more of your vertical travel and another is that they are not well suited to ATCs.

Mike

[CROSSHATCH]

My Torus Pro turns counterclockwise when I execute an M4 command.

Here's a thread I started with similar questions a while back...
http://www.cnczone.com/forums/novako..._backward.html

By the way: I still haven't tapped with my mill. :/ I've done a lot of thread milling, but never automated traditional tapping. I too have a floating tap holder but I haven't even tried it yet - I'm also nervous about being able to stop the spindle quickly enough to reverse direction reliably. I bought myself a braking resistor to solve that problem, but it turns out I got a mill with an incompatible VFD. Grrrr! Apparently there's a replacement coming for me in the mail - last I heard about it was three weeks ago.

As for the R8 collet: the collets Novakon and Tormach sell are designed to be nearly flush with the spindle nose when tightened. I've never actually used a "normal" R8 collet, but I get the impression that they normally stick out an 1/8 of an inch or so. The collet sticking out might prevent the tool holders from making contact with the spindle nose.

See pg. 5 of this PDF from Tormach for a helpful image.

I have a couple Tormach collets and a single Novakon collet. The Tormachs seem to work fine but the single Novakon collet I have is too tight - TTS collet chucks will not fall free from the collet, even if it's completely loosened. Of course, maybe the one I received is an outlier. I wish I could find a company selling very high precision quick-change 0.75" R8 collets but so far I only know of these two.

Kind of wish I had the breaking resistor, but who knows...I might be able to get it done nicely with the right dwell settings. Only way to find out is perhaps break a couple of taps Just kidding, well maybe not.

That's what I pretty much thought in terms of the collet (collet "stick-out"). I wonder if I just faced it flat, again gotta try. Maybe it will work, maybe it won't. I wouldn't mind a little stick because I can see myself trying to juggle the collet while I'm loosening the drawbar...hate when tools fall and hit the table, or the work piece.


Novakon:

Maybe the beta manual has a few things from the older machine? I just thought I remember seeing it a couple of times. Just for the record, I am not using a reversible tapping head. I'm simply using a floating tap to take up any slack in a un-synchronized spindle, again hopefully I don't break too many taps and I'll mainly be doing thru-holes.

I'm the kinda guy that loves to design something on my own and make it, being an engineer...I can't help myself. If it's something that I have to have over time spent designing and prototyping, I will absolutely consider the PDB. It looks to be a very nicely designed system and I'm positive it will work great, I'll read the thread again in the morning with a nice cup of coffee.


MichaelHenry

Good bit of info in that last post in terms of your dwell and RPM. Makes absolute sense in which a blind hole will be tricky, especially dealing with smaller taps. As of now, I don't see doing too many small holes on any upcoming projects. I'd be in the #10 size hole going upwards and on other future projects or if it was work that I use to do, 4-40 and 6-32. So we'll see, good info though.


Thanks all.

-Jason

[rlockwood]

Jason,

Might consider thread milling, especially if these are relatively low-quantity projects. Its just about always slower; but generally won't scrap the part if it breaks, and doesnt require the high degree of synchronized spindle & axis accuracy that tapping does. It also gives much more control over the thread spec (can easily dial-in a tighter or looser thread, etc.) Mostly applicable to relatively shallow tapped holes, etc.

Further, a single point thread mill can do virtually any thread, though for small threads I recommend a dedicated pitch.

[CROSSHATCH]

I think it's something to consider but I'd almost be really disappointed if I couldn't properly tap on the machine using the common floating tap. I just feel that it's plausible and should be allowed/doable on all accounts.

I will certainly look into a single point thread mill and it's quite interesting actually.

Thanks for the insightful input, you get a little more control over the threads, I like that.


-Jason

Jason,

Might consider thread milling, especially if these are relatively low-quantity projects. Its just about always slower; but generally won't scrap the part if it breaks, and doesnt require the high degree of synchronized spindle & axis accuracy that tapping does. It also gives much more control over the thread spec (can easily dial-in a tighter or looser thread, etc.) Mostly applicable to relatively shallow tapped holes, etc.

Further, a single point thread mill can do virtually any thread, though for small threads I recommend a dedicated pitch.

[LeeWay]

The problem with it as I see it is that if it does take a long time to spin down, then wouldn't you overshoot easily?
I mean I tap pretty slowly when I do rigid tapping, so that wouldn't take long to stop and reverse, but mine has the servo spindle of course. Not sure you would have the torque at that low of a speed. Like 100 rpm. Now I sometimes tap at 300 rpm.

Of course, I don't actually know how a floating tap works.

[rlockwood]

Yes, a normal floating tap will largely over-travel if the spindle doesn't stop when the axis does. The floating tap holder allows a certain amount of travel by the tap to accomidate the extra revolutions; but they usually only float half an inch or so, so it truly needs to be minimized. If you have a half inch overshoot, it becomes quite difficult (impossible actually) to tap shallow blind holes.

Ideally, the Z-axis is slaved off of the spindle rotation, so each pulse of a spindle rotation allows a certain movement of the z-axis. When the spindle begins to slow, the z-axis does as well, and reverses when the spindle does. This is rigid tapping, and allows the elimination of the floating tap holder altogether. I'm not sure how implementation of this would happen in mach; though i'm certain its possible. Virtually every CNC lathe requires this, as there's not really any "floating tap holder" equivalent for turning threads.

At the least, it requires a 1ppr sensor on the spindle, which I believe ray is adding with the PDB.

I agree, tapping is a must have. I suggest thread milling as an alternative simply because it's easy to implement. OTOH, it would be worthwhile for the community for someone to truly sort out whats required to achieve acceptable floating tap performance; likely achievable through the addition of a braking resistor & some accel/decel VFD settings (I wonder if its possible to pass an alternate set of accel/decel settings to the VFD when a tapping cycle is issued?) Alternately, same deal with a rigid tapping setup..

[LeeWay]

On the Torus and the Pulsar, the BOB is what does the syncing. There is a switch on it that you use when you intend to do tapping. Not sure why the switch is there, because it acts normal when you aren't tapping. Maybe for machine setup convenience?

It does rigid tapping by syncing the spindle and the Z axis. Spindle becomes the A axis then when in tapping mode. It uses a subroutine so far and it is very effective, but it needs to be filled in by hand. I just broke a 5/16" tap because I neglected to change the tool offset call numbers. I had the correct tool number with the old tools offset. :nono:

Sheetcam is working with us on a post that should act like this as well when a tapping mode is called. So far it is very close. Les told me something else to try. Just haven't had a chance to do that yet.
The servo on the Torus stops immediately and reverses, so there isn't any dwell at all in the bottom. Just spins in and back out smoothly.

[rlockwood]

I bet the switch is switching the servo amplifier mode from speed\velocity mode to position mode..

IMO, the setup on the pulsar/torus is great; but controllers aren't generally setup to control a servo as a spindle very well. What they have going appears to work well; but its pretty dependent on quite a bit of g-code to make it happen. I'm betting the subroutine is producing a z-axis move equal to the pitch of the tap, with an a-axis rotation of 360* (and then the opposite upon retract)? Its actually a neat way of doing it. As long as the z-depth per revolution matches the pitch, the feedrate directly controls the spindle "RPM" to match. While it requires a bit of programming each time, that really shouldn't be an issue; I could write the post processor code needed for an HSMWorks post up in a few minutes. A deep hole will require a few dozen lines of code (where a normal subroutine would require one..) but who cares? Perhaps post up the subroutine they're having you use.. i'm a little interested. I'd be perturbed if I needed to walk to the back of the machine and flip a switch whenever I wanted to run a tapping cycle though; surely this could be commanded by an m-code?

FWIW-- On the swiss lathes we have, you can enter a threading cycle, kill the spindle, and manually rotate the spindle around; the z-axis will continue to move in sync with the spindle rotation. This is what I consider a truly "rigid tapping" setup; the second axis is electronically being driven by the encoder position of the major axis (and a given ratio)

[CROSSHATCH]

Will post my thoughts shortly. Good info going back and forth.

-Jason

[LeeWay]

Here is the tapping part of the code John sent to me initially.

N2510 (Operation: Tap, TAP_1, T11: Rigid tap, 0.167 in x 32 TPI, 0.5 in Deep)
N2520 S150 G00 Z0.1000
N2530 M09 (Coolant off)
N2540 M05
N2550 (Rigid tap, 0.167 in x 32 TPI)

N2610 G00 X-0.2027 Y-0.7467 Z0.1000

M98 P10000
Z.5

N2710 (Operation: Tap, TAP_2, T11: Rigid tap, 0.167 in x 32 TPI, 0.5 in Deep)

N2750 X-0.5295 Y-0.1793
N2760 Z0.0394

M98 P10000
Z.5

N2850 (Operation: Tap, TAP_3, T11: Rigid tap, 0.167 in x 32 TPI, 0.5 in Deep)

N2890 X-3.5374 Y-0.2100
N2900 Z0.0394

M98 P10000
Z.5

N2990 (Operation: Tap, TAP_4, T11: Rigid tap, 0.167 in x 32 TPI, 0.5 in Deep)

N3030 X-4.7454 Y-0.1933
N3040 Z0.0394

M98 P10000
Z.5

N3050 M30



O10000 ( S/R Tap 32TPI .5" Deep )
M49
F 200 ( Arbitrary Feed Rate USE F200 )
G91 ( Switch to Incremental Mode )
G1 Z-0.5 A16 ( Plunge 1/2", Rotate Spindle 16 Times CW )
G1 Z 0.5 A-16 ( Retract 1/2", Rotate Spindle 16 Times CCW )
G0 Z 0.1 ( Raise Spindle 0.1" )
G90
M48
M99 ( End Subroutine )
%


It is for tapping 4 holes using the same tap.
When you want to use two size taps, just add the subroutine again with a different line call.

I first generate the full code in Sheetcam. Then using the coordinates it gives, just swap and change things according to those and the taps size, depth and speed.

Mach 3 cannot really do rigid tapping yet.
It thinks it can, but doesn't. The timing at the bottom of the hole is off and that is where the Z and spindle become out of sink just enough to screw up the threads.
I've heard Mach 4 has this corrected though.

You also don't have to flip that switch every time.
I switched it the very first day and haven't touched it since.
Now when I get Mach 4, I can simply switch it back.

[rlockwood]

Ah ha. Interesting, it's not setup quite how I expected, a axis is specifying rotations rather than degrees. Now I'm curious how they did that. Lol.

[LeeWay]

They did it with a switch on the BOB.
It's all a little over my head at this point, but John knows what is going on. I know just enough about Gcode to get myself into trouble.

Maybe I should mention it here.
I have a brand new Shars tapping head that I was going to use on my HF mill. With the Torus though, I won't be needing it. It is this one.
6 1 2" Tapping Tap Head Self Reverse Drilling Milling | eBay

I intend to list it on Ebay if no one here wants it.
Price will be $200 shipped on Ebay. $180.00 for a Zone Member.
It has never been mounted. The only thing missing is the outer cardboard sleeve.
Still packed in the styrofoam box and also comes with the adapter so it will fit a 3/4" collet. It looks and feels pretty substantial.

That may help solve someone's tapping woes.
The collets and holders for each size rigid tap I use is around $100 each. This one will hold taps from #6's to 1/2" in size. Heck-u-va lot cheaper.

[Novakon]

I am intrigued by the investigation to reveal the secret to Novakon's rigid tapping process. While the BOB technology is a closely guarded secret, the result is a very coordinated movement through out the tapping cycle.

To do a good tapping job, especially when you want to do a bottom tap, you need precise dual axis control. Most motors simply can not do a good job to stop where you want to. The servo motor does this very well. We can stop within .001" of the bottom and back out. So, the result is very clean thread.

We do have a video showing the operation and control that might answer some additional questions.

Rigid Tapping Tutorial

We will be more than happy to entertain more questions on our unique rigid tapping capability.

Regards,
Novakon Team

[LeeWay]

Here is another threading video I did this afternoon.
My Video - YouTube

[BAMCNC.COM]

Do you take trade-ins for a tormach?

[CROSSHATCH]

Little tired from the past few days, catching up on the reading.

Thanks for posting up the code, but no rigid tapping here with the Torus Pro...will be using a ER20 floating head (not the clutch one though).


-Jason

[CROSSHATCH]

Servo Edition Torus Pro here!

I have some harmonics going on in my x-axis after setting and messing with the mill tonight. The machine was awfully grimy and had particulate almost everywhere, I did run the axes a little before wiping off as much as I could off the ballscrews. I oiled her up with my Mobil #2, I don't have the time to take apart the entire x-y assembly; I would really like to because I know they could use a gooooood cleaning.

I registered about .001 backlash in the X today. It just doesn't have a great sound right now, the Y-axis is silky smooth but I did not check the backlash on the Y-axis.


Also, the Z-Axis was making noise after moving. It's as if the encoders were fighting gravity? I would move and the Z-axis would hum, I think it happened in + - but once settled it would HUM until I moved again. This is mainly after the Z-axis was lubed up nicely the past 48 hours.


I'm more concerned with the x-axis and the harmonics. I might mess with the Gibs on the X-I wonder if it's too tight or It's set that the ballscew is enduring axial load and "harmonizing". It's hard to tell right now...

I'm also super tired, it's 3:10 am in NY right now and I have been working on the machine since 8 Pm. Sigh, I'll edit this post in the morning to make more sense I suppose.

Thanks,

-Jason