Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Tool-n-Around

Pages: [1] 2 3 ... 17
1
Related Softwares / Re: Freecad To CB STL's
« on: November 12, 2023, 14:19:07 pm »
Kelly I had a look into whether or not Alibre could import .stl files. Not directly, but if you use this you can convert a stl file to a step file and then import that.https://github.com/slugdev/stltostp
Dave

For the most part the things I've tried to import that were derived from stl files are an absolute mess. I'm getting proficient enough at modeling, that for most of what I do, I'm just better offer making my own solid model and working from there but there are several exceptions, mostly relating to highly organic surfaces. With all the stuff out there for free import for 3D printing, it's great but I usually need to make some mods for function or so I can actually hold and machine it. The other exception is 3D scans of objects with organic highly sculpted surfaces.

Best,
Kelly

2
Related Softwares / Re: Freecad To CB STL's
« on: November 09, 2023, 14:36:22 pm »
First off, that's a heck of a project just for grins.

.......On the Freecad side of things, the most important thing is to not use the file save as stl as provided by the main menu. Once having made the solid model and then creating an stl with the mesh workbench.... There are quite a few methods to generate the stl including different meshers, the default one with default settings is quite good and produces a far superior models to the main menu save as method......

........If you don't have a 100% error free model then you can get miscellaneous tool path errors like unexpected plunges. Even if I'm going to cut the model with CB, I load it up in Prusa slicer as it's a little more fussy than freecad regarding errors as a further check.

Dave

I'm may have to look into FC. In Alibre, although I can export stl's, I cannot import them and as far as I know, only basic mesh control settings upon export, and no way to analyze the resulting stl. For the most part, my files appear to work well in CB except at the surface boundrary where I ferquent get plunge errors, but only when I use shape outline as the bounding method. When I use the same file with a shape bounding method inside the surface boundary, no issues. So that's been my go to method.

Best,
Kelly

3
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 29, 2023, 19:08:59 pm »
That's a beautiful piece of work!  You should be proud.  I'm glad the forum could help, but it's obvious that most of the skills are yours!!!

Lloyd

Thanks Lloyd. Much appreciated. Here's a few more in-process pics.

Best,
Kelly

4
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 29, 2023, 18:58:30 pm »
Just a follow up on this project and thanks again for all the help I received from forum members along the way. The lost foam casting pattern has been completed.

https://youtu.be/42jmOe4cwHQ

Best,
Kelly

5
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 26, 2023, 19:42:02 pm »
You talk about the values in the tool library ? ("Flute Length" and "Length" properties) If it is the case, CB don't use those values, they are only used for simulation to give an alert if the depth of cut is larger than the tool length (with CutViewer, I don't know if Camotics give alerts)
++
David

Yes, that's it, I was writing my post above as you were replying. As usual, you knew right where I was going with the question.

Best,
Kelly

6
CamBam help (General usage) / CB's Treatment of Tool Depth
« on: October 26, 2023, 19:39:45 pm »
I was going to start a seperate thread but think this is really more of a 3D Machining Related topic.

It looks like Camotics recognizes the cutting length of a tool because if I simulate cuts deeper than the cutting length, it leaves uncut stock above the axial depth of cut. It got me thinking about how CB handles tool depth parameters.

I went and had a look at the CB tool library parameters and their definitions. On Axial Depth of Cut it says "not currently used for tool path generation!".

The tools I was using for this project are 4" long but only 2" flute length. Some of the 3D cutting for this project was 3" depth into the stock. When I did roughing MOPs I allowed .030" roughing clearance and then I followed with full depth finish MOPs. In some places on these parts, there were vertical faces 3" tall/deep into the stock. It looked like the unfluted portion of the bit may have been running into the uncut part of the deep vertcial faces left from the roughing clearance. So in other words, when I call for the finish MOP, CB doesn't know or account for the flute length being less than the depth of cut because a 3D Finsih MOPs just cut full depth to the surface. Is that correct?

If so, I'm fortunate I was only cutting foam. The foam thickness from roughing clearance probably just deflected at the inactive portion of the cutter. If it was a harder material there would likely have been a crash and broken bit. I wouldn't have even thought about this if it wasn't for Camotics simulation.

If I have such an instance in the future, I suppose I could use SL Roughing with no roughing clearance and set the depth of cut to the flute length, but it's really the only vertcial faces that require the attention, not the entire surface.

Best,
Kelly

7
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 26, 2023, 13:58:38 pm »
.....Yes, it is one reason to use negative boundaries ; I don't know why CB do those unwanted plunges on the side, but I already seen this before. Not sure if it is related to the 3D model or the boundary polylines.

I think you may be right about that. I get these errors frequently but only at the surface boundary and never in the interior, and only when I use shape outline bounding method. On the same surface, if I create and use a bounding shape the excludes the edge of the surface, then there is no problem so there must be some anomoly at the edge of the stl files. I use to create surfaces so I could just use them as is with shape outline bounding but now I just accept that I'm better off creating a good bounding shape. In this case, I just went back to my original larger (not the reduced area version) surface with a polyline boundary and all is fine.

With Camotics, sometime you also may have bad simulation if you're working with Z0 at the bottom of the part, because when Camotics run simulation, it always start with the tool at Z=0, unfortunately (for all I know) there is no way to set the Z position for the tool start position as in Cutviewer and its FROM command.

Noted. Thank you.

You become an expert of 3D machining ;) ... nice job.

I'm trying!

I had a little mishap while machining the second part (lower half) related to this post. I had already completed the machining on one side and was running the WL Rough MOP on the second side when one of the linear bearings on my Z-axis became fouled and caused loss of steps late in the MOP. Consequently, it cut too deep afterward and I thought I had scrapped the part.

So I had to completely dissassemble the Z Axis and then take on the tedious task of fully dissassembling the 4 linear bearings, cleaned, re-lubed, and reassembled them. While I had the Z-Axis apart, I attended to a couple other annoying design features so this turned into a half day project itself, but I was back in business.

I decided to run the SL Finish MOPS just to see if there were any other surpises or things I had overlooked. After I ran the first SL Finish Horizontal MOP, I discovered the eroneous cuts were quite shallow and had not broken through to the exterior surface on the other side. Being this is a lost foam casting pattern and I routinely glue pieces together, I just glued in patches and ran the last SL Vertical finish MOP which it trimmed them to shape so the art is saved!

I have discovered one other area I need to better understand with regard to how CB treats cutting length versus total tool length. I'll make a seperate  post on this subject.

Best,
Kelly

8
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 24, 2023, 17:12:21 pm »
I think I've answered my own question at least in part. I discovered what was causing the apparent plunges on the simulated example in post 18. The length of the .25 BN tool in Camotics tool libray was only 2" but the depth of cut was nearly 3". Since these weer finish MOPs they were full depth. I changed the tool length in the library to 3.5" and all the things that looked like errors dissappeared.

However, there are a number real plunge holes on the the boundary of the part I cut yesterday and they show up both in the CB tool paths and the Camotics simulation. But, this part was cut with the "Reduced Surface" model using shape outline as the boundary, whereas the simulation above uses the full size surface model with a bounding shape. I have seen this before when I use shape outline for bounding surface so I still need to have a harder look at this. This is what I was referring to in post 19 and using the negative boundary margin to eliminate the plunges.

Best,
Kelly

9
CamBam help (General usage) / Re: Tool Plunge Errors
« on: October 24, 2023, 14:14:50 pm »
One other tidbit that may be related to the above, when I used the "Reduced Surface" version of the model, I used the shape outline bounding method with 0 boundary margin. When I examined the toolpath in CB, I observed a number of areas where it looked like the tool was falling off the edge of the surface. I was quite surprised by this because with a boundary marging of 0, isn't the center of the tool on the boundary and thus half the tool diameter still within the boundary?

To eliminate this, I had to set a negative boundary margin and the plunges didnt not dissappear in CB until I increased the negative BM to -0.02"! This is why I previously inquired about the use of negative BM.

In the previous post above, the plunge errors shown in gcode simulation did not show up in CB tool paths like they do in the attached picture

Best,
Kelly

10
CamBam help (General usage) / Re: Tool Plunge Errors
« on: October 24, 2023, 13:07:24 pm »
Sometimes when I use 3D MOPs on surfaces I get these random tool plunge errors. Such is the case on these two parts. They are real as far as the gcode goes because they show both in simulation and the actual cut. Fortunately, they are always on or more usually just outside of a boundary so they don’t scrap or damage my actual machine part. This is true whether I use a bounding shape or shape outline. Also quite fortunately, I am machining polystyrene foam or these anomalies would very likely cause a crash or broken bit in harder materials.

Here’s the strangest thing, using the same CB file, surface, and boundary, if I individually enable the 3D MOPs, they (the plunge holes) appear in different places on different MOPs or they don’t appear at all. The pictures below are an example. You can see plunge errors in different places on the SL Horizontal and Vertical MOPs, yet no plunge errors on the WL Rough or when all of the MOPs are run in a single simulation.
I tried resaving the CB File, regenerating the tool paths, and ran a new G-code files. The plunge errors recur in the same places.

Any ideas what could be the possible causes of this and how to correct it?

Sorry, CB file is too large to post. I forgot to mention the surfaces are imported .stl files. I can post the mesh settings I selected if needed.

Best,
Kelly

11
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 24, 2023, 13:05:24 pm »
So I’ve done some experimenting with the previous suggestions and here’s what I’ve concluded on reducing processing time.

Back face culling has always been set to true/activated so there were no gains in processing time to be had there.
I changed the CB Worker Threads Setting to 4 instead of the previous 2.
I did go to the CAD model and removed a portion model surface that wasn’t used in the cutting path as shown in the attached pictures.
For the SL scalloping on the oval features, I also generated a bounding area one tool width wider than the oval 8-eared plenum for a WL F MOP, but I used the same reduced surface model as opposed to creating a whole new (further reduced) surface for just the WL F area, so I did not reduce processing time any more than the that for the entire (new but reduced) surface.
I do have an email and web browser open, along with a couple Camotics windows, and the computer is not off line so I suppose there could be all kinds of crap running in the background.

In summary, there may be some improvement but if so I cant tell because the processing time WL Finish MOPs is still quite excessive….several hours. I don’t think any of things I did above were worth the effort nor time to try to apply them to any of the other CB CAM files in an effort to reduce processing time the other 3 sides for this project. 

On the other hand, processing time for SL MOPs is quite reasonable and usually less than 1 minute. WL Rough only slightly longer.
With respect to run/cut time, at the speeds and settings I’ve selected, it takes about 40 minutes of run time for any finish MOP whether SL or WL.

It takes about 15 minutes to execute a WL Roughing MOP vs 40 minutes for a SL R MOP, so even though it’s much cruder roughing, for foam it’s sufficient prep for the finish MOPs and that is a good improvement in cut time.

I can’t see any reason to continue to torture myself with WL Finish MOPs processing times, at least for this project. So I ran the first part with just WL Rough, and both SL Finish Horizontal and Vertical, and the results are just fine with very minimal scalloping that will easily clean up with some minor hand finsihing.

But, there were a couple 0f very strange tool plunge errors in locations that nearly severed my holding tabs. Luckily not and the part is fine, but it caused me to go look at the simulations for the second part and see what I discovered in the post below.

Best,
Kelly

12
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 23, 2023, 17:26:12 pm »
Re Threads.

Read this to see how many threads (Logical Processor) your machine has, then put that number into CamBam settings.

https://www.intel.com/content/www/us/en/support/articles/000029254/processors.html

It says mine has 12 cores and 20 threads so I have 20 in the CamBam Worker Threads setting.

It basically means the CPU can do 20 things at the same time rather than one after the other, i.e. synchronous as opposed to asynchronous

I'm a little confused on the terminology. My old laptop that I run CB on says it has 2 cores and 4 logical processors......several thousand threads. My newer computer has 4 cores and 8 logical processors and displays a similar number of threads. How do the cores/logical processors/threads relate to the best CB settings for CB worker thread setting?

A search indicated there are typically 2 CPU threads for physical core.

So would the appropriate CB worker thread setting be 4 for my CB laptop?

Best,
Kelly

13
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 20, 2023, 19:50:45 pm »
.....Also, you can set "Back Face Culling" to "true" so the mop will ignore triangles that have the face that are not pointing to the top. (as those on the bottom face of the model)

Ah, this is some low hanging fruit too, because there is a lot of additional unneeded backside surface because I just exported the solid as an .stl.

Quote
It basically means the CPU can do 20 things at the same time rather than one after the other, i.e. synchronous as opposed to asynchronous.......Yes and with CamBam, if possible, each MOP will be calculated by a separate thread so in Eddy's example, if you have 20 MOP, they will be calculated at the same time.

It will be a couple days before I can work with this again, but I will apply these tips. Thanks you.

Best,
Kelly


14
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 20, 2023, 16:13:43 pm »
Re Threads.

Read this to see how many threads (Logical Processor) your machine has, then put that number into CamBam settings.

https://www.intel.com/content/www/us/en/support/articles/000029254/processors.html

It says mine has 12 cores and 20 threads so I have 20 in the CamBam Worker Threads setting.

Ah, thanks Eddy. I was way off. I'll do so. Thank you.

Best,
Kelly

15
CamBam help (General usage) / Re: Waterline Finish 3D MOP - Again
« on: October 20, 2023, 14:25:24 pm »
.......If you go to the tools\options and look at the bottom of the page, there is a property box where you can allocate the number of threads for CB to use. Ideally the PC should be offline, and not running anything else. Then run as many threads as possible.

I've been using the term instances of CB. I gather that's incorrect. What I really mean is just the number of windows I have open working active/different CB projects. Are those CB threads? I presume you mean the setting with the red arrow. It is and has been set to 2. I don't think I'm understanding what this actually does and what about the green arrow worker thread setting? The help explanation isn't clearing it up for me.

......If your chasing down run time and have picked all the low hanging fruit,then you can try (effectively)horizontally slicing the model and use multiple mops, essentially copies of each other and adjust the depth increments etc, accordingly.

It's not so much run (cut) time as it is the duration of processing time so I can experiment with applying different MOPs to different portions of the surface. Rather than run/cut time, what's more important to me is the quaility of the cut surface to reduce hand finishing and high first pass yield to avoid scrap.

When it takes hours to get to the point of merely viewing a simulation, that's a bit defeating. Keep in mind, I'm cutting polystyrene lost foam patterns for casting, and making them one at a time. It doesn't make much sense for me to invest hours into improving the program to reduce minutes of run time. However, I do expect to make more of this part, slight variations on the part, and similar parts, so as I do, I can incorporate improvements.....thus this post.

I have added many features to the CAD model to facilitate ease of CAM such as the machining frame, holding tabs, parting surface, offsetting the parting surface so I get full depth of cut beyond the ball nose radius, etc, so additional mods to the surface model to facilitate ease of future CAM programimg are no problem. I shoot for a surface such that when I import into CB, I can just use the shape outine as the bounding surface and the task for the intial cut at the CB program is just to position the surface and calling the 3D MOP. Then maybe I add a bounding shape to cut down the total surface area to be cut and reduce run time.

The first MOP is usually just a finish cut so I can see what I have. If that looks good then I add a MOP for roughing and stock removal before the finish MOPs. I usually give myself .030-.060" of roughing clearance. It really doesn't matter much with foam nor does direction of cut so I use mixed, and I use the same ball nose bit for roughing (though specify end mill in CB for roughing) just to avoid tool change so I can let it run and go do something productive. I use 2D and 2.5D MOPs where possible.

As opposed to WL F with a bounding box as David previoulsy suggested, for speed, I think I'd just produce a toolpath to geometry polyline 1/2 the tool diameter larger than the eared/oval perimeter, project that tool path to the surface, and then use an Engrave MOP to clean up the entire perimeter of the vertical faces of the oval with 8 ears in a single full depth pass (after roughing)....make sense? But why go to the effort if calling a single WL F MOP would do the entire surface?  The answer as to why for me is because it's taking 2-3hrs to get the tool paths and another 2-3 hours for the GCode, and I have to do this four times because there is two sided machining required on two halves of the the part.

-That's my thought process. Seem reasonable?

Best,
Kelly

Pages: [1] 2 3 ... 17