[nlh]

Hello everyone,

I have a TC1g and looking for some feedback on a few questions I have.

First off, I'm looking for feedback from those of you that have taken the plunge and upgraded their BMDC board to the BMDC4.

EMI says this card will greatly improve how smooth the machine operates when running a program with many small moves, such as 3D work. I have yet to run any 3D programs on my machine since installing it about 8 months ago, but I have been running a fairly large amount of programs that use constant engagement strategies. The Advanced Adaptive Rough in BobCAD V28 gets used a lot.

I was surpised how well it actually works, not having a machine before this TorqCut that could handle this sort of code.

The problem is, even with careful optimization of the code it still can easily generate a program size of 100K or more and the motion could stand to be smoother. Sometimes it can be jerky and sometime will get a banging sounds as it changes direction. Looking for feedback from anyone on here that has the BMDC4, if it works as advertised.

Also, my machine has a gearbox that is noisy and needs to be swapped out. I have a spare that I got from a fellow member (Thanks Forhire!) late this past winter, but have been so busy I have not had time to do the swap.

I'm wondering though, considering the types of toolpaths I'm using above, if it might be worth my while to look into removing the gearbox altogether. The gearbox does increase torq by quite a bit according to the maintenance manual, but in my opinion the machine is light enough duty that any heavy roughing should be done on another machine.

Surface finish is a consideration. Currently anything over 4500 rpm will cause a "scallop" like finish. This shows up really bad on aluminum. The gearbox does seem to have a fair amount of backlash when rotating the spindle back and forth by hand (been this way since I got it) and the rattle can be quite loud when on accel/decell. I'm wondering if this backlash could be part of the poor surface finish.

Hoping someone who has a torqcut without the gearbox can chime in with some feedback and also someone who has one with a gearbox can give feedback on the type of finish they get.


Thanks,
Nate

[forhire]

The problem is, even with careful optimization of the code it still can easily generate a program size of 100K or more and the motion could stand to be smoother. Sometimes it can be jerky and sometime will get a banging sounds as it changes direction. Looking for feedback from anyone on here that has the BMDC4, if it works as advertised.

I'm pretty sure I have a BMDC3 but admittedly I've never given it much thought for the stuff I do. I'll have to check my spare controller. Did EMI indicate what needs to be changed in addition to changing the BMDC card?

I'm wondering though, considering the types of toolpaths I'm using above, if it might be worth my while to look into removing the gearbox altogether. The gearbox does increase torq by quite a bit according to the maintenance manual, but in my opinion the machine is light enough duty that any heavy roughing should be done on another machine.

While I do run at 7500 a lot and, I wish I could go faster, I can't imagine not having the gearbox for tasks like drilling. I wouldn't remove it.

Surface finish is a consideration. Currently anything over 4500 rpm will cause a "scallop" like finish. This shows up really bad on aluminum. The gearbox does seem to have a fair amount of backlash when rotating the spindle back and forth by hand (been this way since I got it) and the rattle can be quite loud when on accel/decell. I'm wondering if this backlash could be part of the poor surface finish

Can you feel the scallop with your finger or is it just visual? The backlash could contribute but I'd suspect tool deflection first. Is the tool running concentric? What is the stick out? Have you blued the spindle to see out the taper fits? I had to have mine ground.

I'd probe around with a stethoscope and see where your noise is coming from... mine turned out to be the DE bearing on my spindle motor.

[forhire]

My spare is a BMDC3. So much for digging treasure.

My program space is 256k... larger than that I have to DNC. I don't have issues doing DNC but it does introduce some lag sometimes. Doesn't the BMDC4 have larger program space? I seem to think it was 4 megs... but don't quote me. What did EMI quote you for the BMDC4 upgrade?

Understand there are some things you can do to shrink your post sized, remove spaces, comments, line numbers, non-move axis calls, redundant feed callouts, trailing zeros, and etc. You get the idea. Also make sure your post is using arcs and helical interpolation than line segments. To me it appears the machine can interpolate a lot smoother than a bunch of segments. You can also break your posts into smaller files rather than one monster large post.

[DarkStarMedia]

I have ran close to the 256k limits as my post does line segment moves instead of G12/13 for helical (anyone know Solidcam posts well enough to fix?). I would be interested in what the quote for an upgrade that can handle larger files.

I have the 8k spindle (no gearbox) and have not ran into any load problems that were not my own fault.

I have not drilled any bigger than 7/16 and my largest tool is a 3.15" facemill (glacern FM45). I have been working in aluminum... Getting ready to try some steel in the future though.

One thing I can think of that the gearbox would be handy for is tapping but on the larger sizes you can threadmill.

Richard

[forhire]

I have ran close to the 256k limits as my post does line segment moves instead of G12/13 for helical (anyone know Solidcam posts well enough to fix?). I would be interested in what the quote for an upgrade that can handle larger files.

Does the BCDM4 do look ahead?

When I setup helical G12/G13 on Visualmill I emailed a scan from the manual along with a working sample and tech support sent me the changes. For fun I posted some code using interpolation and the same using line segments. What I can do in two lines using helical interpolation took 500+ lines as segments.

Here's the deal with line segments. If I set the arc fitting tolerance at 0.01" it is pretty choppy but less code, 0.005" was smoother but more lines of course. Using arcs was the best. Arcs and helical interpolation 17k, 52k and 66k for the others.

Choppy 52k 2000 lines.
Attachment 331856

Less choppy 66k 2650 lines
Attachment 331858

Helical and arc interpolation 17k 1000 lines.
Attachment 331860

[nlh]

Thanks for the reply's and discussion guys,

It's been pretty hectic since my first post so I apologize for taking so long to get back.

When I last talked to EMI about the BMDC4 upgrade, it was probably about 6 months ago, so working from memory...I believe the price was around $1700-$1800 plus ~$900 or so core charge. At least if you go to bptparts.com that is what they have listed. Actually when compared to a regular BMDC3S, it's not a huge difference in price. Search - bmdc - Bridgeport & CNC Machine Replacement Parts

It has 2M of memory and according to Bart has better motion processing capabilities. I don't believe there is any software upgrade to go along with it, just a direct swap but I can't confirm that until I talk with them again.

I use BobCAD V28 Pro with thier advanved 2D rough path. The post actually does a pretty good job of breaking the code down into arcs, much better IMO than V26. I've noticed the same thing with the tolerance settings, it look like a really choppy toolpath but it's just for roughing too so not a big deal most of the time.

It's interesting but I have been able to use standard G2/G3 helical moves (at least the machine doesn't alarm out) but be careful becuase when the machine reaches aprox .01-.015 above the programmed depth, the head drops with a small bang, which doesn't make for very good tool life. I don't have my post setup to use G12/G13 just yet, so I usually will predrill or do a linear ramp.

For me and the BMDC4 it's really more about smoothing out the motion rather than program size, at least for now. I have instances where for no aparent reason, the machine will just make a banging sound when switching direction. I have seen this on just simply X,Y linear moves like milling around a square...just a random corner, "Bang", weird.

That's a good idea using the stethescope, I can borrow one from a friend. I will take a video sometime soon and post it so you can see what I mean by backlash (very noticeable when orienting the spindle) and spindle noise.

I did have some issues a few months ago with tools fretting and sticking in the taper which is a pretty big red flag that the tool isn't seating, however upon inspection (as close as I can get with what I have for tools) could not find anything obvious. I did rent a drawforce gauge and found my drawbar pressure to be about 1250-1300psi (manual calls for 1700 IIRC). I installed a new spring washer stack and got the pressure up to around ~2000 but still have occasional sticking issues.

I have been using a special type of pullstud from JMperformanceproducts.com called thier HT line which is supposed to eliminate the swelling on the small end.

I am going to call a company soon about having them come check my spindle and possibly regrind on the machine. It's possible some damage was done prior to me getting the machine, although visually the taper looks very good.

On another note, this machine threadmills very well!

[forhire]

It's interesting but I have been able to use standard G2/G3 helical moves (at least the machine doesn't alarm out) but be careful becuase when the machine reaches aprox .01-.015 above the programmed depth, the head drops with a small bang, which doesn't make for very good tool life. I don't have my post setup to use G12/G13 just yet, so I usually will predrill or do a linear ramp.

G2/G3 on this machine is 2 axis only. This likely explains your drop despite your lack of error.

G2, G3 — CIRCULAR INTERPOLATION
When the control is in the Circular Interpolation mode (G2, G3), a circular arc in the selected plane can be generated by
the coordinated motion of 2 axes. The format is:
X-Y Plane (G17)
G2X__Y__I__J__F__; CW
G3X__Y__I__J__F__; CCW
Z-X Plane (G18)
G2X__Z__I__K__F__; CW
G3X__Z__I__K__F__; CCW
Y-Z Plane (G 19)
G2Y__Z__J__K__F__; CW
G3Y__Z__J__K__F__; CCW

For me and the BMDC4 it's really more about smoothing out the motion rather than program size, at least for now. I have instances where for no aparent reason, the machine will just make a banging sound when switching direction. I have seen this on just simply X,Y linear moves like milling around a square...just a random corner, "Bang", weird.

Looking at the two the BMDC4 appears to have a larger processors. Good to know the swap just requires a hardware change. My machine has banged a few times also. I haven't really dug into the cause.

I did have some issues a few months ago with tools fretting and sticking in the taper which is a pretty big red flag that the tool isn't seating, however upon inspection (as close as I can get with what I have for tools) could not find anything obvious.

If you blue the taper it will show you the contact area. Take a look at this post that shows my before and after grind.
New (to me) VMC - Page 5

[nlh]

Thanks again Forhire,

The thread you referenced from the welding forum is packed full of info, seems like I find something new every time I go thru it. :-)

After finally taking the time to do some more investigation into my spindle issues, it would appear I have almost the exact same issue with mine as you did on your TC3. Blueing show's a ring at the top (small end) and a very slight ring of contact toward the bottom (big end). But no (or extremely little) contact in the middle.

I've attached 3 photo's of my results. Photo 1 (blueing is a little more pronouced) was taken before turning spindle on, the other 2 are what the holder looked like after running spindle for a few seconds.

This will likely explain why my machine seems to be so finicky about speeds/feeds, even in aluminum. It just seems on the verge of chatter or other weird harmonics all the time. I can't run anything with an even # of flutes and get good results, has to be either a 3 or 5 flute.

Would probably explain the poor surface finish at higher rpms as well. Rather than gearbox backlash.

Now to find someone willing to travel to northern Maine. Current quotes are around $1750. Wouldn't be as bad except i'm in the middle of a large building expansion. Never convenient timing on these things!

[Jonp]

nlh, are you sure you are using the right pull studs ? I've got a TC3G like forhire and I use the short pull studs like in his posts, i've often wondered if im using the right ones maybe our machines use different ones I don't know. Our machine cuts amazingly well but I've often wondered about the pull studs as some bridgeport literature i have shows ones like in your picture although it is referencing a VMC760 not the TC or torqcut series. Maybe some one who knows there pull studs will chime in ??

[nlh]

Jonp,

That's a great question, and one I've spent many hours trying to answer definitively. I did end up finding a service bulletin, I believe it was bulletin 206 IIRC and it was recently mentioned on a another post from a day or so ago. According to this I do have the right ones but there are 2 different styles, both the same stem length. The difference is the angle where the grippers (or in my case balls (BT40), grip the stud. At this point I am 95% sure I have the right ones. Do you use CAT or BT holders? In my opinion, the CAT is a better setup simply because the drawbar uses 4 grippers rather than 4 balls. I believe this would grasp the pull stud better.

[nlh]

Here are a few pics of the marks left on my spindle and tool taper. These were taken a few months ago.

[Jonp]

OK well after a little more investigation i think,( i have been wrong before) the longer pull studs like you have are for BT spindles and the shorter ones are for CAT spindles

[forhire]

After finally taking the time to do some more investigation into my spindle issues, it would appear I have almost the exact same issue with mine as you did on your TC3. Blueing show's a ring at the top (small end) and a very slight ring of contact toward the bottom (big end). But no (or extremely little) contact in the middle.

The grind procedure was really pretty simple. I'll likely do it myself next time. Basically the tool post grinder is trammed in and an incremental program is run while making very light cuts. He'd take a few tenths and then check the bore. I think it took a thou and a half to get it cleaned up if I recall. He used a pre-dressed wheel but I'd think it would be trivial to dress it on the machine.

Here's the secret sauce.

G91
G80
S500M3
G1X0.005F20.0
G1X0.5378Z-3.6875
G1X-0.005F20.0
G1Z3.6875X-0.5378