Trying to use conforming vectors again [Archive]

luckettg

01-20-2017, 02:48 PM

I bought this upgrade when it first came out, tried it a couple of times, getting weird results. Mostly it was carving crazy, not moving like it should, etc. I moved the text out of the pocket with the pattern, just cutting text above and below and it worked fine, every time. Wasted a lot of wood back then, playing with it.

I decided to try it again when a customer asked me to make him a design that needs it. My first run at it, it again went nuts, right at the beginning during the pocket carving with the 1/16 ball nose. The X and Y axis seemed to be moving right, but the Z was bobbing up and down. I am attaching the mpc file and the picture of the board. I aborted the carving.

I have been looking it over and cannot see anything wrong with the design. I am carving a modified one of it now. It cut the pocket and the pattern okay, but then when reinstalling the V bit, it gave the error msg about not finding the correct depth. I reinstalled that, then same thing, so I gambled and told it to continue anyway. Usually that creates a mess with the text, but this time, as it is carving at the moment, it looks okay in the bottom of the pocket. When it is done, or I abort it for some reason, I will include the results. The mpc file is also included now for it.

I will report more in a bit.83059

luckettg

01-20-2017, 03:01 PM

8306083061Well, it worked pretty good; the best it ever has for me. It did not carve the text on the Hero's Engine very well, but the other text will work. Deeper would have been better, but that has always been a crap shoot with fonts and Designer. No depth control settings at all. I used the 1/16 ball nose and the 60° V bits.

SteveNelson46

01-20-2017, 03:21 PM

Greg,

The machine will ask for all of the bits used in the project in the setup and measuring cycle at the beginning of the carve and then again as each bit is used. The error message "Bit Depth Does Not Match" is usually caused by the bit that hasn't been inserted to the same depth as the first time it was measured. Are you using stop collars to insure the bit is inserted to the same depth each time? It has to be inserted to a depth that is within the acceptable range and to the same depth each time to avoid the error message and air carving. Also, the bit has to touch the bit plate each time so the depths are set correctly.

luckettg

01-20-2017, 03:50 PM

luckettg

01-20-2017, 03:57 PM

Darn, lost my reply to you Steve. I will try it again.
I know about how to install bits and the errors that are caused by not doing so correctly. This is not the case, but thank you for the suggestion. I have had this machine since Dec. 2007, and have both upgraded and rebuilt it several times. This error was caused by something other than miss intalling the V bit. Every now and then, since day one, the machine will give this error, and it is always with a V bit. Those are really hard to not reinstall correctly, so I have my doubts. My guess is a software glitch that measures the original operation, then does not do the calculations correctly when reinstalling the V bit.

Connie suggested I should also do another calibration check, but for a different reason. That had to do with text depth. It is a good idea, so that is next on my list.

The X axis has been having issues since putting in the rubber traction belts a couple of years ago. Intermittent in nature so I suspect board variances are causing that.

Again, I thank you for trying to help me. I appreciate that. If you have any other ideas, please let me know. :)

DickB

01-20-2017, 04:01 PM

The Centerline tool is for the 90 degree V bit and 60 degree V bits only. You selected and used the 1/16" ball nose bit for the text on the Hero's Engine - that is why it turned out so poor. Try again selecting and using the 60 degree V bit for that text.

Asking for a depth control on Centerline is asking to change the laws of trigonometry. The Centerline tool manages the depth of cut to yield the width dictated by the width of the text strokes. The only way to go deeper with a given V bit is to go wider on the text strokes, by selecting Bold or a different font with wider strokes:
83063

Or select the 60 degree V bit as you did. That bit will cut deeper than the 90 degree bit given the same font and text size (stroke width):
83064

luckettg

01-20-2017, 04:15 PM

Yes, I know how the text and depth is related and based on the angle of the bit. What I have a very difficult time with is how to determine the actual font width versus the actual cut depth. It will look fine in Designer, but that is just an illusion. I also run a Shark flat bed CNC, and do have depth control there, so it is not a simple matter of trig laws being not understood on my part. There are a lot of fonts to choose from in the drop down menu for the text, in Designer, yet no good way to determine the width or the depth. I might be wrong and just ignorant, but I struggle with it.

Using the ball nose for text was just an experiment, as it is in the drop down menu for selecting the bit, in Designer, for the text. If only the 60° and the 90° V bits should be used, why are all of those other bits selectable?

Thank you for the suggestions. I appreciate your effort to help me. Any ideas or suggestions are good.

DickB

01-21-2017, 08:13 AM

It will look fine in Designer, but that is just an illusion.
I don't agree. Designer gives a good representation of the end result for me. If the result is not as illustrated with Designer, I suspect it is because of some kind of machine problem.

What I have a very difficult time with is how to determine the actual font width versus the actual cut depth.
There are a couple of ways to do this. With the 90 degree bit, the depth will be 1/2 the width of the stroke. Draw a line across the stroke to make a measurement and read the length of the line. Even easier, place the cursor over the center of the stroke and read the depth at the bottom of the screen:
83065

I also run a Shark flat bed CNC, and do have depth control there, so it is not a simple matter of trig laws being not understood on my part.
Are you using Vcarve Desktop software? I'm not very familiar with that software, but while you can set the depth of V carve text, I believe that is used to generate a toolpath in a recessed area to produce a faithful version of the given text. If the depth setting doesn't match the depth of the surface, then the resulting character is deformed. I'm not sure why you would want that. The depth setting in Vcarve is not bending the trig laws. Using an incorrect depth setting will distort the text. In Designer, Conforming Vectors makes this depth adjustment for you, preserving the shape of the text. It's true that in Designer you cannot specify an incorrect depth to yield distorted text like you can with Vcarve.
83066

luckettg

01-21-2017, 09:29 AM

I suppose we will have to agree to not agree on your first point. I am using the latest Designer versions, always have, along with most of the add on packages. I completely disagree with you about the fonts looking like the way they actually carve.
Second point, I will give those two ideas a try, and run some test carvings, thank you.
Third point. I use Aspire, which is a souped up Vcarve. You are of course right about what happens when not set correctly, but I can tell if I am doing something the machine will not like, usually before actually carving and finding the mistakes by wasting wood and carving time. I really like being able to set feed rates, bit rpms, and a wide selection of bits to choose from, even setting up new ones. I use Aspire to create 2-1/2D PTN files too. Export from Aspire into STL to PTN.
We have digressed from the conforming vector discussion, my fault for mentioning what Connie and I were discussing about text. That first carving that went whacko, I am still unsure why. Spent more time on it in Designer last night and will do another carving today, after I double check the machine calibration. I will report what is discovered.
Thanks for the discussion and the methods for text sizing to try.:)

SteveNelson46

01-21-2017, 10:05 AM

Aspire doesn't really have a depth setting either and the same rules apply. What it does have is a start depth and a flat depth setting. This is the same as Centerline in the Designer but with much more control. The start depth setting allows for a v-carve in a carve region or pocket or even on the surface of the board and if you want the bit to carve a little deeper just set the start depth a little deeper. If the Designer had this setting one could play with the start depth to diagnose a problem and compensate, within reason, for fonts that don't carve very well.

On most CNC machines how deep the bit is installed doesn't matter as z-zero is set only once when a bit is inserted and used. If no bit is inserted at all the collet will touch the surface of the board or z-zero plate.

SteveNelson46

01-21-2017, 12:15 PM

Greg,

I think your right. We did get off topic. Your original post was about conforming vectors and somehow was hijacked to centerline text. They are two separate but similar functions with different applications. Sorry about that.

In Aspire, "conforming vectors" can be accomplished by putting a check in the box for "Project toolpath onto 3D model" in the toolpath setup and the results are clearly shown in the preview. Correctly applied Conforming Vectors should be clearly shown in the Designer also. I'm not sure why the preview in the Designer isn't correct on your computer. Maybe it has something to do with the graphics card and the hardware acceleration setting on the preferences menu.

bergerud

01-21-2017, 01:03 PM

I wonder if many of the problems we see are due to the font. I recall the programmers having many problems with fonts which did not somehow conform to the true type font standards. So, maybe a one step in trouble shooting conforming text problems should be to try a standard, known font.

luckettg

01-21-2017, 02:07 PM

When looking at the list of fonts, I have no idea how to tell which one is standard. How is that done?
I have used Ariel and Ariel Narrow on the examples provided above.

bergerud

01-21-2017, 02:25 PM

I have used Ariel and Ariel Narrow on the examples provided above.

They are pretty standard fonts. I guess the fonts are not the problem in your case.

DickB

01-21-2017, 05:33 PM

I completely disagree with you about the fonts looking like the way they actually carve.If you have an example of a font that did not carve as previewed in Designer, could you post a photo of the carve and screen shot of the Designer view or the mpc? I've never had a mismatch unless the surface of my board was uneven.

DickB

01-21-2017, 05:39 PM

SteveNelson46

01-21-2017, 08:10 PM

I don't see how I went off topic. The mpc's in the first post use Centerline text only as the conforming vectors. Post 2 included a complaint about how the conforming text was poor. I responded to those posts.

Dick,

The post was in the Conforming Vectors section and the heading was "Trying to use Conforming Vectors Again". The topic of the text did start out as "conforming text" though. I thought it was me that went off topic.

luckettg

01-21-2017, 09:08 PM

Have you also been having difficulty with font sizes not carving as displayed in Designer? Just curious how many are experiencing this.

I wonder if many of the problems we see are due to the font. I recall the programmers having many problems with fonts which did not somehow conform to the true type font standards. So, maybe a one step in trouble shooting conforming text problems should be to try a standard, known font.

luckettg

01-21-2017, 09:13 PM

This thread is digressing to mostly text issues instead of conforming vectors, but they are so interrelated that it is probably really still on topic. My original posting was about getting conforming vectors to work at all, not just text. Until yesterday, I never got conforming vectors to work enough to get to any text problems, and that jury is still out. I am running some more conforming vector test carvings to see if I can figure out why it usually does not work for me. The guy I would usually pester with these questions left us a few years ago and I just found that out today. That would be AskBud. He was a great mentor for me and I still use his teaching files (AVI or PDF).

Dick,

The post was in the Conforming Vectors section and the heading was "Trying to use Conforming Vectors Again". The topic of the text did start out as "conforming text" though. I thought it was me that went off topic.

luckettg

01-21-2017, 09:18 PM

No, not that I know of; I have hundreds of pictures taken over the last 10 years so something might show up. I usually only take pictures of completed projects and the failed ones head to the fireplace. Next time it happens I will try to remember to take pictures.

If you have an example of a font that did not carve as previewed in Designer, could you post a photo of the carve and screen shot of the Designer view or the mpc? I've never had a mismatch unless the surface of my board was uneven.

luckettg

01-21-2017, 09:44 PM

luckettg

01-21-2017, 11:04 PM

I wonder if this was caused by coming out from under the roller? It would only take a tiny change to create the lighter carving. It would not be as noticeable in the V bit text carvings, but in an etching or very shallow text, it would look just like this.

.
Dick, I just took a picture of the carving I did yesterday, and a screen shot of the Designer depiction of what it should look like. It is using the 1/16 ball nose bit, which Designer allows to be chosen. I am not yet sure why it turned out this way, but I like the idea of using other bits. I have used end mill bits to cut out letters too, and that worked pretty good. I like the ball nose bits for the smaller size lettering. This is the first time I tried it while using Conforming Vectors.8308483085

fwharris

01-21-2017, 11:37 PM

I wonder if this was caused by coming out from under the roller? It would only take a tiny change to create the lighter carving. It would not be as noticeable in the V bit text carvings, but in an etching or very shallow text, it would look just like this.

.

If that is the case that you are carving with out staying under the roller I would agree.

I was just getting ready to say that there was an issue a couple of years ago with a few machines that went. They were air carving or carving very shallow. The problems was with some type of machine code. LHR issued a patch that had to be flashed to the card and then installed on the machine.

DickB

01-22-2017, 07:48 AM

I am running some more conforming vector test carvings to see if I can figure out why it usually does not work for me. Greg, some photos of the test results and mpcs might help diagnose the problem. Other than the text issue it's not clear to me what problem you are experiencing.

DickB

01-22-2017, 08:01 AM

Dick, I just took a picture of the carving I did yesterday, and a screen shot of the Designer depiction of what it should look like. It is using the 1/16 ball nose bit, which Designer allows to be chosen. I am not yet sure why it turned out this way, but I like the idea of using other bits. I have used end mill bits to cut out letters too, and that worked pretty good. I like the ball nose bits for the smaller size lettering. This is the first time I tried it while using Conforming Vectors.
8308483085
In Designer I read that using the 1/16" ballnose bit for your text the depth will be .017". If you used the 60 degree V bit instead the depth would be .067" - almost 4 times deeper. Of course that would be much more tolerant of any board movement, if that is what is happening here. Personally, I would be using the 60 degree V bit for that text and actually any small lettering. It will cut significantly deeper than the ballnose bit. But if you want to stick to the ballnose bit I would make sure that the board is not moving and stays under the rollers. That is a possible cause of this problem.

I think I will try a couple of test carves with a portion of your mpc myself and see what results I get.

If you're not using a sled, I highly recommend that you do for a number of reasons. I use them on virtually all my projects.

I'm not sure why Designer allows the use of other bits, but this is how Centerline is described by LHR:
The Centerline Text Feature adds an additional Rout Mode for the Text Tool feature. This feature uses the 60° and 90° V-groove bits to follow the center of the letters for a clean engraved, v-cut look.

SteveNelson46

01-22-2017, 10:37 AM

Greg,

Although the Designer will allow you to select other bits when using the centerline function it was designed to use only v-bits. I think this is a programming glitch to even allow other bits to be selected. If you stop to think about it the Designer would have to compensate for the shape of the bit and I don't think it's in the programming to do that. Kearning and other factors would have to be taken into account also. A good example would be to select the 1/2" ogee bit. The bit could not plunge deep enough to show the profile so how would the preview display it? The Designer program just isn't written the same as Aspire. In Aspire, you can even design the shape of your own bit and the preview will display that shape at any depth. The Designer will not. Even in Aspire you can select any bit you want for a v-carve but it just doesn't make sense. A v-carve is, well, v carving and to use a bit other than a v-bit may not give you the expected results.

SteveNelson46

01-22-2017, 12:03 PM

Most fonts are made with parallel lines instead of one singular line. The spacing between the lines that make up the letter is varied. They get closer together or further apart to achieve that tailored look. As it's carving, v-carve or centerline, the bit is raised or lowered to fill the space between the lines. Otherwise one could just assign a bit to a vector in the shape of a letter but it wouldn’t look the same. How would it look if a bit other than a v-bit is used. It might look okay if a tapered ball nose is used but how deep would the bit need to go to fill the space. That would depend on the size of the tip and the taper on the sides of the ball nose bit and if the program knows what those measurements are. If a straight bit was used how deep would the bit go? If the tip of the bit is wider than the space between the lines it wouldn't carve at all and if it’s narrower, ostensibly it would plunge all the way through the board, There are many factors to consider if a bit other than a v-bit is selected for a v-carve and I don't think the Designer is programmed in detail enough to allow for these factors. V-carving is meant to be used with a v-bit.

DickB

01-22-2017, 01:24 PM

I ran a test using a portion of your mpc (to save carving time) and the result looks like a good match to Designer in my opinion, but you be the judge. Take a look at the detail, such as the squiggly crossbar on the "T" in both Designer and the carving:

83092

Here is the same thing using the 60 degree bit for comparison. Notice that the "T" is perfectly straight. The depth also makes the text more visible as there is more shadow (you can see this comparing the two thumbnail pictures - the V text pops):

83093

I think that these results suggest no issues with Designer, but rather with your machine or your setup. BTW I used a sled for these carvings.

luckettg

01-25-2017, 07:32 PM

Dick, I made a simple sled a couple of years ago to carve on acrylic, but have yet to really make a sled. I agree that I should do so and have had the plans and notes about making them for years now. I would like to learn more about what sleds you are using, if you are willing to share that.

By the way, I modified the carving and it worked fine. I did not do any lettering on the Hero's Engine; I like your idea of the 60°V-bit.

Thanks for the help and advice.

DickB

01-26-2017, 07:10 AM

Greg, now we are getting off topic, but no worries.

I have a few sleds of different sizes that are constructed like this: http://forum.carvewright.com/showthread.php?28480-More-accurate-machining-when-using-a-sled&highlight=sled. I use 3/4" plywood or MDF as a base, 3/4" x 1" rails top and bottom, and 4" ends. This ensures that the sled always stays under the rollers AND has at least 2" of surface to grip. It keeps the cutting bit at least 1" away from the brass roller which is also a good thing.

At the top of the sled are two or three thumbscrews. These are used to clamp the workpiece in place. I have a stack of scrap strips of wood to use when my workpiece is narrower than the sled. These are laid in place between the workpiece and the thumb screws. The thumb screws clear the sliding plate, but can get in the way of finding the bit depth if you're not careful. I have my machine set to jog to touch position so that I can touch on the work piece and not the rails, so I simply jog to a position clear of the thumb screws.

Due to the thickness of the base the machine is unable to determine the board thickness, so you have to enter the thickness when prompted.

The sleds can accommodate 1/2" workpieces by laying a 1/4" piece of plywood down first. I have a similarly-constructed sled for 1/4" plywood but use masking tape rather than thumbscrews to hold the workpiece in place.

luckettg

01-26-2017, 07:30 AM

Thanks Dick. This is pretty much what I had evisioned making and your details have encouraged me. I probably have everything needed to quickly make one too.

Last night I discovered the Universal Carrier Base and am considering purchasing the plans for it. Have you tried one? I am not sure exactly how it will work for me but it does look interesting. I watched some videos and scanned through the PDF about it. I would appreciate your advice. PM me if you would like to be more private about discussing things (or email me?)
The coffee is ready here now.....it has a sirens call. Thanks again for helping me.

luckettg

01-26-2017, 07:33 AM

Steve,
I discovered (rediscovered?) how to find True Type Fonts on my PC, doing a search for the *.TTF files. I am struggling with which fonts to use, especially when they are one that I have not used in the past. Your advice will be used, along with Dick's. I do appreciate it and thank you.

DickB

01-26-2017, 07:48 AM

Last night I discovered the Universal Carrier Base and am considering purchasing the plans for it. Have you tried one? I haven't tried one myself. Dan developed it after I had made my own sleds, so I did not have a strong need. It is interesting and I have followed the threads on it. Others have posted about it so you can get their views. I don't recall any negative comments.

DickB

01-26-2017, 07:58 AM

I am struggling with which fonts to use, especially when they are one that I have not used in the past. I think the main thing to watch out for is very thin strokes. For example, look at the capital A in this font (Frederick text):
83125

There are some very thin lines meaning that the bit is going to be shallow. If your board has even a slightly uneven surface, or your sled rails are slightly taller and you touch the bit to the rails, the stroke might not carve at all. The smaller the lettering, the more difficult to carve well. I like to highlight text with a darker stain or spray ink, sanding off the excess. It's easy to sand too much and destroy the lettering with a font like this. I would be very cautious about using this font.