[BAMCNC.COM]

Depends on length really, I've done it with aluminum foil.

[powerr]

Try here For cable. good prices, I have used the 10 conductor.
SC9-100: JAMECO VALUEPRO: Wire & Cable

[MRM RCModels]

http://allelectronics.com

I buy all of my shielded cable from here. Best prices I've seen. Built 3 machines with them, no problems.

[SCzEngrgGroup]

My luck was much improved today. I built the first successful harness, and now know exactly how to build the "real" ones, once I get more cable. I also found a few more minor "tweaks" I can make for the production units, to make installation a bit easier. I ended up spending several hours chasing a really bizarre firmware bug, which was caused by something in one of the Arduino library objects I used. But, all fixed now, and working nicely.

Here's an overview of the installation. Keep in mind, my machine is pretty extensively modified at this point, and the wiring harness in particular does not exactly represent "production", so the installation will look neater with the "real" production kit. The PDB assembly on my machine is pre-beta, so is bare aluminum, while the beta and production units are all black-anodized.

Attachment 194708

Installation will be pretty easy for new machines which will be pre-drilled for the PDB. For older machines, I'll provide a drilling template, to make it easy to drill and tap the two new bolt holes on the top of the head. The PDB "head" mounts with just two bolts, and the spindle "cap" gets replaced with a new one with the integrated spindle brake and RPM sensor, which simply screws onto the top of the spindle. The electronics mounts to the inside of the door, with four screws to hold the "control panel" in place, and four more to hold the Plexiglas panel containing the rest of the electronics.

The wiring harness runs through a flexible conduit, identical to those already present on the head. It mounts to existing screws using a provided bracket.

Attachment 194710

The middle vent grille on the E-box is removed, and replaced with a new one which holds the other end of the flex conduit. It attaches using the existing hardware. This panel also provides the exit for the 4mm plastic air line.

Attachment 194712

Inside the E-box, a small sealed battery is placed near the AC entry panel, along with its AC-powered charger.

Attachment 194714

The wiring harness connects to the VFD, and the BOB, to intercept the VFD control signals, control the VFD during PDB operation, provide an E-Stop in case of PDB malfunction (you really don't want the spindle trying to turn on with the PDB engaged, nor do you want a program to try to move the machine with the spindle disabled....), as well as the new PROBE input, which connects to one of the free BOB inputs.

Attachment 194716

Attachment 194718

Finally, here's a short video, showing the drawbar being repeatedly loosened and tightened, so you can see the speed. Not quite as fast as a Belleville PDB, but pretty darned close. And, with this one, you'll NEVER have to worry about pull-out, even on the most aggressive cuts. And, when you want to swap R8 tools, you can do it just as easily, it simply takes a few seconds longer.

Beta PDB Operation - YouTube

Regards,
Ray L.

[Hirudin]

Looks nice and fast now!

[BTP]

What limits the torque?

[SCzEngrgGroup]

What limits the torque?

Torque is controlled by motor current limiting. It is limited, by the firmware, to 25-30 ft-lbs. on the drawbar.

Regards,
Ray L.

[LeeWay]

Is that a 6 or 12 point socket? 12 point would slip on easier of course, but would wear the drawbar more I think. Not a big issue. They are easy and cheap to make.
One thing I do see is that the spindle cannot freely turn on a Torus and Pulsar, but could be coded to go to a specific tool change point on the spindle. This way a 6 point would work fine.

[SCzEngrgGroup]

Is that a 6 or 12 point socket? 12 point would slip on easier of course, but would wear the drawbar more I think. Not a big issue. They are easy and cheap to make.
One thing I do see is that the spindle cannot freely turn on a Torus and Pulsar, but could be coded to go to a specific tool change point on the spindle. This way a 6 point would work fine.

Lee,

It's a 6-point hardened impact socket. Why do you say "the spindle cannot freely turn on a Torus and Pulsar"? When the spindle servo is not powered, the spindle should turn freely, just like on any other machine. In any case, the spindle MUST turn, to engage the brake, as the servo, by itself, cannot withstand 30 ft-lbs torque. It is not possible to "know" the orientation of the drawbar head relative to spindle position, nor is it necessary to. If the socket and drawbar are not aligned when the PDB comes down, they will be shortly after the PDB starts rotating. The key to minimizing drawbar, and socket, wear is to minimize down-force. And, both the drawbar, and socket are quickly, easily, and inexpensively, replaceable, should they wear over time.

Regards,
Ray L.

[LeeWay]

With the servo energized, it will turn a bit by hand, but pulls itself back where it was. Meaning not fully rigid, but keeps it's position. You would have to kill the power to get it to relax.
I agree that the drawbar position cannot be known. Even with the servo, threads on R8 collets may be different.
The drawbar would have to allow for some spin on the PDB.

I considered doing this on my 80/20 mill with a butterfly impact. I don't have aspirations for an ATC on it though, so it was almost as fast just doing it by hand.
I'd love to have an ATC on the Torus though and that would require a PDB.

[SCzEngrgGroup]

With the servo energized, it will turn a bit by hand, but pulls itself back where it was. Meaning not fully rigid, but keeps it's position. You would have to kill the power to get it to relax.
I agree that the drawbar position cannot be known. Even with the servo, threads on R8 collets may be different.
The drawbar would have to allow for some spin on the PDB.

I considered doing this on my 80/20 mill with a butterfly impact. I don't have aspirations for an ATC on it though, so it was almost as fast just doing it by hand.
I'd love to have an ATC on the Torus though and that would require a PDB.

The servo spindle is not a problem - the PDB can command the spindle to move to the correct position, just as it does on the Torus Pro. Though you don't see it in the latest video (you can in some of the earlier ones) because the brake is already lined up, the spindle is rotated as the PDB comes down, to ensure the spindle brake is engaged before the socket reaches the drawbar. The socket and drawbar come into alignment when the PDB starts rotating. The only difference between the Torus Pro and the Torus is the exact mechanism used to induce the spindle rotation.

Regards,
Ray L.

[SCzEngrgGroup]

Been too long without an update.... I finally got in new cable a few days ago, and made up three harnesses for the beta test units. Today I'm doing functional test on those units, prior to shipping. I'll definitely need to build some kind of test fixture for doing this more efficiently in the future.

I was surprised to find new motor/gearbox units seem to need several minutes of burn-in to get motor current down to a respectable level, and avoid over-current trips in the firmware. I don't know yet why I'm seeing this issue now, when I haven't on any of the previous units. The gearboxes arrive here partially assembled, and I have to completely disassemble, and re-assemble them to their final configuration. I suspect the change might be due to the slightly different way I greased these gearboxes when I assembled them, and perhaps due to slightly higher friction in the PDB hardware external to the motor/gearbox, due to design improvements. In any case, after a few minutes of continuous full-speed operation, the current on all three units came down to expected levels, and the motor speed went up, enough to eliminate the over-current trips. All three units now seem to be operating properly, and will no doubt loosen up a bit more in the first days of real use.

Here's the first completed wiring harness:

Attachment 195946

And the beta mechanicals, nearly ready to go:

Attachment 195948

The firmware is working really nicely, and I believe it can now correctly detect, and react to, virtually any potential failure mode. At least I have yet to see, or even manually create, a failure mode that it does not deal with correctly. In many cases, it is able to simply mitigate the failure itself, by performing extra steps in its sequencing. When that doesn't work, it disables the machine by throwing an E-Stop (you really don't want a g-code program starting when the PDB is not in its dis-engaged, "idle" state), and displays a clear, simple error message on the LCD, alerting the user to the exact nature of the failure, the manual recovery action being pretty obvious at that point. The failures it can detect are:

Lift Down Error (i.e. - PDB lift did not move down when commanded at the beginning of a PDB operation cycle)
Lift Up Error (i.e. - PDB lift did not move up when commanded at the end of a PDB operation cycle)
Spindle On Error (i.e. - the operator attempted to activate the PDB while the spindle is on)
Loosen Drawbar Failed (i.e. - The drawbar did not loosen properly, so the tool is likely still retained in the spindle)
Tighten Drawbar Failed (i.e. - The drawbar did not tighten properly, as when there is no collet or tool in the spindle)
Over-Current Error (i.e. - PDB motor current and/or run-time exceeded safe limits)
Under-Current Error (i.e. - PDB motor current is lower than expected)

And, of course, in all cases, the exact nature of any error is communicated to the ATC, so it can also react appropriately, by pausing any toolchange in process until the problem is corrected by the user.

Regards,
Ray L.

[SCzEngrgGroup]

Finally! Beta units going in the mail to the testers today! Woo Hoo! :-)

Regards,
Ray L.

[LeeWay]

All right. When do we get to see some of these in action?

[MRM RCModels]

All right. When do we get to see some of these in action?

As soon as we receive them

[SCzEngrgGroup]

The PDB beta test is going pretty well so far. Both kits were received by our testers late last week, and both were installed over the weekend. One is working nicely, the other seems to have suffered a minor electronic failure in-transit (the spindle jog function is not working), so a replacement electronics assembly is en-route, hopefully to be delivered tomorrow. There were only a couple of pretty minor issues in the two installations, mostly due to documentation errors, and the expected minor machine-to-machine differences. I'm sure our intrepid testers will post some pictures and comments here before long.

I've made a few minor, but significant, changes as a result of the beta test. Most important is I've finally come up with a wire management scheme that I really like. Don't know why I didn't come up with this approach much sooner.... The PDB wiring will now terminate at the top of the head door, eliminating the need for ALL the wiring to flex and extend when the door is opened. The ONLY wires that now need to cross the hinge area are the ones that go over to the PDB head - this is the two motor wires, and two small sensor harnesses. Plus, of course, one 4mm air hose.

Here are a couple of picture of my machine as it now stands. The production harnesses will not have all the extra wire length bundled up across the top - this just makes development easier for me. So, the production wiring will be very "tidy", and very well constrained.

Attachment 197106

Attachment 197108

So, the PDB is now VERY close to "production ready"!

Regards,
Ray L.

[MRM RCModels]

The PDB beta test is going pretty well so far. Both kits were received by our testers late last week, and both were installed over the weekend. One is working nicely, the other seems to have suffered a minor electronic failure in-transit (the spindle jog function is not working), so a replacement electronics assembly is en-route, hopefully to be delivered tomorrow. There were only a couple of pretty minor issues in the two installations, mostly due to documentation errors, and the expected minor machine-to-machine differences. I'm sure our intrepid testers will post some pictures and comments here before long.

I've made a few minor, but significant, changes as a result of the beta test. Most important is I've finally come up with a wire management scheme that I really like. Don't know why I didn't come up with this approach much sooner.... The PDB wiring will now terminate at the top of the head door, eliminating the need for ALL the wiring to flex and extend when the door is opened. The ONLY wires that now need to cross the hinge area are the ones that go over to the PDB head - this is the two motor wires, and two small sensor harnesses. Plus, of course, one 4mm air hose.

Here are a couple of picture of my machine as it now stands. The production harnesses will not have all the extra wire length bundled up across the top - this just makes development easier for me. So, the production wiring will be very "tidy", and very well constrained.

Attachment 197106

Attachment 197108

So, the PDB is now VERY close to "production ready"!

Regards,
Ray L.

Sweet. I'll be making a video review this week.

[slowtwitch]

So, the PDB is now VERY close to "production ready"!

Regards,
Ray L.

Ray, what is the expected cost of this PDB ????

[SCzEngrgGroup]

Ray, what is the expected cost of this PDB ????

I don't know. You'll have to ask Novakon.

Regards,
Ray L.

[Novakon]

Thank you for your inquiry about our Power Draw Bar (PDB).

The PDB is available in kit form which can be added to the purchase of either Torus PRO Servo or Stepper CNC mills. Older NM-200 series mills may also be upgraded with this kit with slight variations to the wiring harness.