Touch Probe X Not Working Correctly with Surfacing Bit (solved, cannot probe a surfacing bit )

Accessories Touch Probe
1

Received my OneFinity with the Touch Probe Accessory. I checked the dimensions in the settings and updated based on the measurements of the touch probe with my digital calipers. I installed my 1/8" Downcut 2 Flute Spiral and everything was good.

I installed my work surface and was going to surface it so I put my surfacing bit in the router. It measures 25.49mm with the calipers. When I started the sequence in XYZ, I touched up the probe and then it prompted me to put in the bit size. I put in 25.49mm and hit set. I then started the sequence. It successfully performed the Z axis but it did not move over enough for the X and gouged the block. I have now tried this multiple times (finger on software E-stop) and same failure. Here are my questions:

  1. Is there a place to set a more generous X and Y offset so the block does not get gouged?
  2. Does putting in something other than the default 6.35mm even work?

Thanks,
Kevin

2

My understanding is that the best case scenario with the touch probe is using something like a 1/4" diameter piece of drill rod or something similar (upside down end mill, broken drill bit, etc) to probe the X and Y location. Even using a standard twist profile end mill can lead to miss-probing due to back-cuts in the grind. I’m not sure exactly what sort of face mill you are using but you may see the same relief grind behind the cutting edges which could lead to inaccurate probing.

Insofar as the programming is concerned, I assume that @charleyntexas could give you the best answer, but my guess is that the probing program was written with some of the assumptions from the previous paragraph (ie, not a 1"+ diameter bit).

If you search this forum for “probing” you may learn something about aliens. But you may also find some suggestions about probing the X and Y with one bit and then swap the the cutter you actually want to use to probe Z.

Hope this helps!

3

I would use a round bit, like a 1/4" end mill, to probe XYZ. Then switch to the surfacing bit and probe just Z. You can’t probe X or Y with an irregular shape but, like a tapered bit. I’m guessing your surfacing bit is also not going to work because of it’s shape.

4

I use a pin gage for x/y probing. And then the actual mill for whatever I’m using so that the x/y are not impacted by the variations in the diameter at the flutes.

https://www.amazon.com/gp/product/B01BHHK2M0/ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&th=1

1 Like
5

Thanks for the response. I can get a standard bit to work.

I was thinking, since the system prompts the user for the diameter, it should use it to offset the touch-up algorithm. Then, any bit should work. If it does not use the diameter to update the algorythm, why prompt? I think there is a bug in the code. Thoughts?

6

I think you’re missing the point. Even if you could change how far the 1F moves after probing the Z axis unless your bit is perfectly round it’s not going to probe either the X or Y axis with any sort of accuracy.

When you enter the bit diameter for probing I think it only uses that inform the controller how much of an offset is needed once it probes the X and Y axis. For example when you probe with a 1/4" bit the software divides that in half, so .125", and then adds it to the offset for the probe block.

7

I often refer to some bits as being Not Touch Plate Friendly. Those type of bits include large diameter bits such as waste board cutters, Fly Cutters, Face Mills, Tapered Ball Nose, and some V-Bits. As many of you have discovered, there is a easy work around which will allow you to still use your touch plate to set your work origin and use these bits for carving.

The work around is to use a bit that IS Touch Plate Friendly to establish your X Y zero origin, then switch to the bit that is NOT touch plate friendly and probe just the Z axis. This works because the X Y zero is a fixed position that will not change regardless of the initial bit used or any subsequent bits used. It is and always remains, the vertical centerline of the spindle.

So, should you go out and buy a precision gauge pin and always use that for setting your X Y origin? Absolutely not. If you are dead set on chasing the absolute most repeatable X Y zero, a much cheaper solution then grab an old bit, measure the diameter of the shank as precisely as you can. Put it in your spindle upside down and probe to the shank diameter (normal 3 axis probe sequence). Next, swap that out for you bit of choice and zero just the Z axis. The first probe sequence will establish the X and Y zero and the second will reset your Z zero without affecting the X and Y.

Is it true that a ¼ inch end mill isn’t really ¼ inch in diameter? Yes, it is true. Most ¼ inch bits start out as ¼ inch but during the grinding process a very small amount is removed from the outer diameter as the flutes are cut and ground to be sharp. You can still probe to the cutting portion of most bits but you should measure them with a set of calipers first. Even here we are looking at variations that are in the range of 0.002 to 0.006 inches. With most wood carvings, you will never see that small of a deviation and you might just drive yourself to drinking. If you are carving on metals or plastics then you may need to go down that rabbit hole. But for wooden “Welcome to my Shop” signs, don’t waste your time. The general rule of thumb is if you can’t see a problem in you engraving from 5 or so feet away, it isn’t an issue.

Should OneFinity modify the zeroing code to allow for larger bits by calculating how far the X and Y moves should be based on bit diameter? I had never considered that myself at Triquetra CNC but it would be an interesting idea. However, it is up to OneFinity to determine if the costs to incorporate that into their software is worth the bang for the buck. Software engineers don’t come cheap. I do know that through my many one-on-one conversations with the Owner of OneFinity, they are truly all about providing the best possible customer experience. So, if you would like to see that change, then ring that bell and make your wishes heard.

In conclusion, although the OneFinity is probably the best machine available in it’s price range (including much more expensive machines), it is not meant for manufacturing mission critical components for NASA. It is a machine designed for wood carving but capable of much more. It is the reason you see so many X-Carves and Shapeoko’s being sold used.
Lastly, my number one tip of all time is:
The Golden Rule of Probing. “THE MAGNET ALWAYS FOLLOWS THE TOUCH PLATE!!!”.

My number 2 tip of all time is:
“Keep your feed rates down to 3 Inches Per Minute or less, while probing to prevent touch plate bump.”

2 Likes
8

The gage pin I use (I have a few) was under $3 delivered and accurate to -0.0002" - I don’t think that’s too expensive :grinning: With Amazon Prime delivery was free.

A used bit is going to have two issues - potentially less accurate machining of the shaft just due to the requirements of the intended use. The collet doesn’t mind if it’s a few thou off 1/4". The gage pin is machined to a much greater tolerance (10x or better).

The other problem with used bits is the potential for out of round conditions. As it’s used, repeated deflection of the bit is going to be reflected in the bit’s vertical integrity as well as its roundness. Not usually a lot but generally in the thous.

Besides, I’ve gotten cut by the flutes putting them in upside down before :yum: Compression bits are my fumble finger downfall.

This is definitely true. :+1: But starting as precise as you reasonably can helps keep the errors from compounding and being obvious to the untrained eye. A $3 gage pin falls into my definition of “reasonable”.

Kind of like Congress, a billion here and a billion there and pretty soon you’re talking real money :wink:

2 Likes
9

Dr Al,
Thanks for the response. I understand the mechanics of the bit and was planning on rotating to get both sides set up correctly. The frustrating part is that I put in the correct size of the bit into the dialog box but it did not offset enough and bunged up the block. I was not expecting that.

 I agree, that what you described is what it should do. Because it contacted the block, one can conclude that it is not what it does.  The worst case is that it would go too far and take longer to touch up if the blades were not pointed toward the block.

Thanks,
Kevin

10

Charlie,
Thanks, this makes sense. For X and Y, I will just use a pin. For large cutters, I will do Z only.

Thanks,

Kevin

11

This seems a bit strange. If the zero routine is not using the entered dimension of a larger bit, is it using the entered dimension of the smaller bit or pin? I agree with the poster that if it cant use a larger bit it should throw an error. Or it should not ask at all and the instructions should just say to probe with a 1/4" drill rod. If course dimensions will still vary so the instructions could say use a 1/4" +/- .01" drill rod?

12

I’m not sure you got the idea. When you have a bit that is too large for the zeroing routine to handle then you can start with a smaller bit that it can handle. With the smaller bit you should zero all three axis. This will get you the correct X and Y zero location that will be valid for ANY bitt you use… but … when you change to the larger bit that you actually want to use the Z will be off. To fix that you simply zero just the z axis and not all three. When zeroing just the Z axis the X and Y are unchanged and will still be valid, only the Z is updated to the current bit.

When you design your project you also select the bit diameter you will be using for each tool path. When the design software generates the carving gcode, it takes the diameter of the bit into consideration and calculates the tool path accordingly. You can change bits as often as you like and X Y zero will remain constant, it is the tool path that adapts to the tool diameter.

So just to make it clear. Here are the steps to follow if you have a bit that is too large for zeroing with. This also works with tapered ball nose bit and V-Bits.

  1. I will assume the bit you want to use is a 2 inch Fly Cutter.

  2. Start with a straight bit such as a 1/4 inch end mill installed in your spindle.

  3. Run the zeroing sequence to zero all three axis. This will establish your X Y Zero.

  4. Replace the 1/4 inch endmill with your 2 inch Fly Cutter

  5. Position your fly cutter so that the tip of the cutter will over the target circle. ( or any other part of the touch plate for that matter)

  6. Run the zeroing sequence to zero just the Z axis only. This will set the Z height that you want without changing the X Y zero location previously set.

  7. Now you have a X Y Z zero location set correctly for the Fly Cutter.

2 Likes
13

I agree with you that error validation in the 1F controller should be better handled in this and a number of other cases.
However it’s a small team, they work from a third-party/open source basis and so firmware updates are not that frequent.
We have to accept these rough edges as part of the affordability of the machine I’m afraid. Despite these limitations, it’s still a pretty advanced package!

1 Like
14

That is an understatement if ever I heard one!. While anyone could argue that the OneFinity is not a perfect machine, I would argue that it outperforms every single machine in it’s class by leaps and bounds. I remember watching a video where Ben Meyers and his son were both standing on top of the OneFinity Gantry! You won’t find that test of rigidity attempted on any other cnc machine I have seen.

2 Likes
16

Hmmm… I’m not native speaker, as you may know, and I thought that “pretty advanced” expressed a very positive assessment of a non-professional device.

One learns every day :face_with_monocle: and I thank you for giving me the opportunity to correct myself.

17

When people truly love something they can become extra critical of details because they want it to be perfect :wink:

Thanks for the step by step, no doubt some new folks will benefit from the description and your continued commitment to education.

2 Likes
18

I do not disagree with you. In fact I do agree with your assessment. I just added to it, that’s all.

1 Like
19

Thank you for this explanation.
My question:- After the zeroing of just the Z with larger bit (anywhere on the plate), do I just load the toolpath and the origin would be unchanged or do i need to move the machine to the origin? If so how do I move it back to the origin without throwing off the X,Y already set? Thank you

20

This is a little bit of a long reply. If you (the reader) do not have a good understanding of how to zero odd shaped bits and how it all works, please read the entire message. It’s not rocket science but it is CNC, and as we all know by now, CNC is not all common sense. If you have questions please don’t hesitate to ask.

Some bits are what I refer to as NOT being “Touch Plate Friendly”. These include larger diameter bits such as waste board cutters, Tapered Ball Nose bits, V Bits with a cutting diameter larger than the shank, bits with indexable inserts, and so on. This is why Z axis Only probing is a valuable tool to have.

When using these types of bits, it is still critical to be able to have an established X Y Z zero that conforms to the same origin as the rest of your project. @Olan, you are on the right track in accomplishing this. The only thing your missing is confirmation that what you are thinking is correct. Here are the answers to put your mind at ease. (Just so you know, I have been making touch plates and writing zeroing code for the last 7 years so if you can have faith in what I’m saying is factual).

When you probe X Y and Z together before your first cut you establish your origin that all cuts are referenced from. Under normal circumstances the X and Y origin will never change throughout the entire series of cuts on a single project. The Z axis however, may change multiple times depending on if you change bits and how many times. This is where zeroing Just the Z axis comes in. When you change a bit you need to re-zero the Z axis because the tool length is likely to be different. You do not need to re zero you X and Y axis after a bit change. You can jog your machine all you like, just be sure you don’t manually move the X or Y using force during the bit change. Your X and Y Zero is the centerline of the spindle/bit. This never changes even though the bit diameter might. You design software uses the bit diameter assigned to a tool path to compensate for this. Most design software packages never ask for bit length though, only bit diameter. This is where Z only zero comes to the rescue. When zeroing just the Z axis, the X and Y axis (origin) is unaffected. Z only zeroing simply updates the current job setup with the current length of the tool.

Z Only zeroing also makes it possible to achieve a accurate and repeatable X Y Z zero when using those pesky but valuable bits that are not touch plate friendly. First zero all three axis with any bit that is touch plate friendly at the beginning of your project. Then switch to your bit of choice and zero just the Z axis to get that updated information in and your all set. If you are chasing down the most accurate zero possible for X and Y then yes you can use a precision dowel initially to set all three axis but to be honest, most all straight endmills will not have any detectable error in woodworking.

There are some straight endmills that have an extremally high helix that could create significant errors in X Y zero and I put those in the category of being Not Touch Plate Friendly. You can still use them, just use a normal endmill to set your initial origin and then switch to the bit you want to cut with and update Z zero with the Z Axis only zeroing.

4 Likes
21

Personally, I don’t take it lightly whether someone goes out of there way or not, to help others.
So I appreciate your time to explain. Don’t mind feeble and dull minded being. And you’re right; common sense will error out before irregular bits could zero. Anyways, I did it. I changed the 120 (Amana RC-1146) and 90 degrees (Amana Tool RC-45711). I actually used the 120 to probe X,Y,Z by placing it closer to the tip of the plate. Everything works fine after changing it to 90.
Thanks.

1 Like