PDA

View Full Version : New Scanner-Is this right??



c6craig
06-28-2010, 11:31 PM
Hi all,

I ran my first scan with the probe tonight and I'm not too sure if my results are typical. Maybe someone with more experience can tell me if this is what I can expect?

The scan was of a Woodcraft knife blank. I specifically wanted to scan something metal to see if it scratched the surface.

What suprised me...and what I am wondering if is normal...when the probe moved along the knife surface, I expected more of a dragging motion. This was more of a sewing machine style rapid paced tapping. As a result, instead of the smooth surface I expected, I ended up with this.

Any advice?? Am I doing something wrong or is this typical?

Thanks in advance,
Craig

3496434965

bjbethke
06-29-2010, 12:29 AM
Hi all,

I ran my first scan with the probe tonight and I'm not too sure if my results are typical. Maybe someone with more experience can tell me if this is what I can expect?

The scan was of a Woodcraft knife blank. I specifically wanted to scan something metal to see if it scratched the surface.

What suprised me...and what I am wondering if is normal...when the probe moved along the knife surface, I expected more of a dragging motion. This was more of a sewing machine style rapid paced tapping. As a result, instead of the smooth surface I expected, I ended up with this.

Any advice?? Am I doing something wrong or is this typical?

Thanks in advance,
Craig

3496434965

It should be smooth, the last time I had a problem like that was the item was not mounted tight to the sled.

geekviking
06-29-2010, 12:49 AM
That's how some of my knife scans came out too. I used blur or degauss or something in the pattern editor to smooth them out.
By the way, thanks for the pattern, would you mind sharing any others you do, I've been trying to get a good wooden knife collection going... :)
Thanks again!

eromran
06-29-2010, 10:23 AM
Try taking a small mirror or piece of glass make sure and clean it good then scan it. It can be a really small piece so the scan wont take long (2 or 3 inches) be sure when you scan you are picking normal or best and the scan should come out smooth on top. If it does not post the results. Let us know how it comes out any how.

c6craig
06-29-2010, 10:59 AM
Thanks, I will try the mirror trick tonight. I have the machine doing a 4 hour scan now on a different knife blank to see if it does the same thing. I really made sure this one was solidly attached to the sled and not moving in any way. I can pretty much already hear the thing tapping along the surface though so I expect the same result...

Truck seems to move along smoothly until it contacts the item, I even removed the flexshaft from the truck as recommended.

I do select Best quality and set the depth just off the board, using a credit card to touch down on.

Thanks,
Craig

CarverJerry
06-29-2010, 11:22 AM
Did you make sure the probe was tight in the chuck, and are you using the QC, Rock or Carvetight.

CJ

c6craig
06-29-2010, 11:27 AM
Did you make sure the probe was tight in the chuck, and are you using the QC, Rock or Carvetight.

CJ

Yes the probe hasnt shifted in the chuck, I took measurements before and after the scan because I was worried about that. This one is in a Rock Chuck with the adapter sleeve. I made sure to measure because I didnt want to torque too hard on the plastic, so I was a bit worried about it shifting.

It's weird, it just seems to kind of chatter along the top instead of the sweeping dragging motion I expected. When it isnt in contact with the piece, on the air scan parts it moves perfectly smooth. I guess I will see what the scan gives me, and also try the mirror test.

Thanks,
Craig

Router-Jim
06-29-2010, 11:32 AM
I am under the impression that the probe is suppose to sound like a minature jack hammer.

c6craig
06-29-2010, 11:38 AM
I am under the impression that the probe is suppose to sound like a minature jack hammer.

Well then that's what I have. I just don't understand how people get such smooth looking scans if this thing is tapping along the top. Yes, for a flat surface like a knife blank it is pretty easy to clean up, but what about a detailed scan, which is what I am hoping to do..

Maybe I misunderstood. I thought the probe would scan a pretty close likeness into a pattern without much cleanup. Do you guys have to go through the whole photo editing thing to make your scans usable?

Thanks,
Craig

spalted
06-29-2010, 11:38 AM
I am under the impression that the probe is suppose to sound like a minature jack hammer.

I have only done one test scan so far, but that is what mine sounded like.

BJB's suggestions sounds very logical to me.

eromran
06-29-2010, 11:44 AM
One more thing i would try something non metal you could be getting a short inside the probe. That would give you your bumps same may apply to geekviking above. And you didn't misunderstand the probe does scan great and normally requires very little cleanup so don't give up on fixing it or getting it right cause you will love what you can do with it once you get it straighten out.

bjbethke
06-29-2010, 12:19 PM
Yes the probe hasnt shifted in the chuck, I took measurements before and after the scan because I was worried about that. This one is in a Rock Chuck with the adapter sleeve. I made sure to measure because I didnt want to torque too hard on the plastic, so I was a bit worried about it shifting.

It's weird, it just seems to kind of chatter along the top instead of the sweeping dragging motion I expected. When it isnt in contact with the piece, on the air scan parts it moves perfectly smooth. I guess I will see what the scan gives me, and also try the mirror test.

Thanks,
Craig
That chatter and movement is normal, the probe is looking for hight and depth of your project. Also you could use a plastic tip on the probe, there are several posts about that.

c6craig
06-29-2010, 01:40 PM
That's how some of my knife scans came out too. I used blur or degauss or something in the pattern editor to smooth them out.
By the way, thanks for the pattern, would you mind sharing any others you do, I've been trying to get a good wooden knife collection going... :)
Thanks again!

I just posted both this one and a Woodcraft Tanto blank over in pattern sharing depot if you want it..

Craig

geekviking
06-29-2010, 02:52 PM
Awesome! Thanks!

c6craig
06-29-2010, 03:04 PM
Well this is a little disappointing. I did a Best quality scan on a nice smooth piece of corian i had laying around just to use both the "glass" and "non metal" theories and it came out the same way. It kind of chattered across the surface which I know understand is normal. But there is so much noise on the pattern I don't see how I am going to be able to clean up anything detailed....

Thanks,
Craig

3497734978

AskBud
06-29-2010, 03:24 PM
Thanks, I will try the mirror trick tonight. I have the machine doing a 4 hour scan now on a different knife blank to see if it does the same thing. I really made sure this one was solidly attached to the sled and not moving in any way. I can pretty much already hear the thing tapping along the surface though so I expect the same result...

Truck seems to move along smoothly until it contacts the item, I even removed the flexshaft from the truck as recommended.

I do select Best quality and set the depth just off the board, using a credit card to touch down on.

Thanks,
Craig
The Probe is quite sensitive! I talked with another user a few weeks ago, where he scanned a carving he did, so he could make a pattern of just one specific portion. the probe even picked up the grain of the wood, which he could not fathom, as it felt smooth to the touch.

If there is any texture to your model, it will appear in the scan.

Since there is no metal/conductor on the floor of your scanning sled, there is no way for an electrical transfer to the probe of any sort.
My guess is that the knife has contours/textures/pits that you can not see with the naked eye or feel with your fingers. Knives are generally pounded into shape and then ground & polished. Even other items that appear flat & smooth may have pits that will appear. The probe, by design, have a feel/touch finer than the trained blind person!

You could clean the Z-rails and bearings to assure a smoother up & down movement, however, this may give you even more sensitivity. As a former Toolmaker's apprentice, I bet on pits on the knife!
AskBud

c6craig
06-29-2010, 03:37 PM
AskBud,

If tool marks were the culprit wouldn't I see some variance in the noise? It is identical on both knives and the piece of corian.

Thanks,
Craig

AskBud
06-29-2010, 03:45 PM
AskBud,

If tool marks were the culprit wouldn't I see some variance in the noise? It is identical on both knives and the piece of corian.

Thanks,
Craig
Do you, by chance, have a friend with a probe? You could do a test scan, of the knife, in his machine with both probes and do the same on your unit. This may help you trouble shoot the perceived problem. Use the same sled for all the test. Only use the 2nd sled if the first test warrants.
AskBud

c6craig
06-29-2010, 04:06 PM
Do you, by chance, have a friend with a probe? You could do a test scan, of the knife, in his machine with both probes and do the same on your unit. This may help you trouble shoot the perceived problem. Use the same sled for all the test. Only use the 2nd sled if the first test warrants.
AskBud

No such luck. The best I could do would be to try the same probe in another one of my machines, with the same sled. The only difference between them is the Rock and the CarveTight.

You think maybe the probe is bad? Maybe I can find a used probe for cheap since I dont need the software.

Craig

TerryT
06-29-2010, 04:19 PM
The probe goes up and down!! thats how it works. It goes down until it hits the surface and opens the switch. Then it goes up until the switch closes, then it goes down again until it opens the switch then it goes up until it closes, up, down, up, down. Scanning a smooth object will not give you a glass smooth finish. A nice smooth finish will require other software. Artcam is available as is Aspire.... only $2000. Otherwise you can smooth your scan in pattern editor.

mtylerfl
06-29-2010, 04:50 PM
Terry is correct...all scans will have a "texture" to some degree. That is the nature of a physical scanning probe method. Only way around that is use a laser scanner or create a model from scratch in ArtCAM, Aspire, Z-Brush, etc.

The Pattern Editor has several tools to help you smooth things out (de-noise, smooth, blur). An alternative is to save your scan as a PNG graphic, edit that in a photo editor (use smooth, blur, whatever it takes to smooth the texture out), then re-save your edited PNG and import that into the PE, then save it as an MPW (your backup) and as a PTN (for use in Designer). Experiment to see which you prefer.

EDIT: Wanted to add that the more secure the object is affixed to your sled, the less "texture" you will see. Try using wood or other solid 'wedge' super-glued under the knife blade to support it fully. Things like hot glue globs and such will allow vibration, which translates to "texture". The probe (actually the z-truck) goes up and down when following the contours of an object - slight movement of the object enhances the up/down movement yielding a greater "texture". There is nothing wrong with your probe.

AskBud
06-29-2010, 04:55 PM
The probe goes up and down!! thats how it works. It goes down until it hits the surface and opens the switch. Then it goes up until the switch closes, then it goes down again until it opens the switch then it goes up until it closes, up, down, up, down. Scanning a smooth object will not give you a glass smooth finish. A nice smooth finish will require other software. Artcam is available as is Aspire.... only $2000. Otherwise you can smooth your scan in pattern editor.
I don't think this is quite true. the Probe reacts to any movement off dead center, not just the downward touch.
It's possible that the chatter of the motor, or dirt somewhere, may contribute to false readings. You may know more after you compare another machine. The "false sensitivity" issue, if there is one, may be coming from several different sources which you may discover in your side-by-side testing.

Cleaning the contacts within the probe, is sometimes done to improve the sensitivity of the Probe. In this case, I do not think you have a "Low Sensitivity" problem.

My bet is still in the pits in the model! You could test a piece of Oak, and see how it picks up the grain, as an extra test.
AskBud

Mikewiz
06-29-2010, 05:03 PM
Here is the only two scans that I've done and they are of a gun stock. They are smooth, but The probe picked up the wood grain. I didn't touch them up at all these are how they came out of the memory card. Hope this helps.
3498134980

mtylerfl
06-29-2010, 05:06 PM
Here is the only two scans that I've done and they are of a gun stock. They are smooth, but The probe picked up the wood grain. I didn't touch them up at all these are how they came out of the memory card. Hope this helps.
3498134980

Excellent example, Mike. This is a case where the object was very, very secure on the sled. Very minimal "probe texture", is the result. Thanks for sharing the photos.

c6craig
06-29-2010, 06:52 PM
Thanks to everyone for the replies. I will play a bit and see if I can get it to do what I need. A day or two away from it may help me focus back on it with a different view. I apparently did not fully understand the probe for what it is. I simply didn't realize there was going to be so much photo editing, my lack of skill in this being the reason I haven't made my own patterns so far. Not to worry, I will either get it to work out or stick with all the free and purchased patterns!

Thanks again!

Craig

Router-Jim
06-29-2010, 07:22 PM
Craig, one thing I'd like to pass on is to save your original scan from your card to a scan folder. This allows you to always have a fresh original. Now start playing with the editor and see what each of the features does. Save those under another name or don't save at all. You will be surprised how easy it gets in a short time.

TerryT
06-29-2010, 07:55 PM
Craig,
This may have already been asked, maybe I just missed it. What quality setting are you using? Sometimes you might have to play with the quality setting to get a good scan. Try the same object with both normal and best and see which works the best for that object. Sometimes "Best" doesn't necessarilly give the best results. I think the size of the object has a lot to do with it.

c6craig
06-29-2010, 09:06 PM
Craig, one thing I'd like to pass on is to save your original scan from your card to a scan folder. This allows you to always have a fresh original. Now start playing with the editor and see what each of the features does. Save those under another name or don't save at all. You will be surprised how easy it gets in a short time.

Thanks Router-Jim, I did make a new folder for rough scans in the hopes I can do something with them eventually. You are absolutely right, if you don't stay organized this stuff can get out of hand quick! :)

Thanks,
Craig

c6craig
06-29-2010, 09:07 PM
Craig,
This may have already been asked, maybe I just missed it. What quality setting are you using? Sometimes you might have to play with the quality setting to get a good scan. Try the same object with both normal and best and see which works the best for that object. Sometimes "Best" doesn't necessarilly give the best results. I think the size of the object has a lot to do with it.

Hi Terry,

I have only tried BEST setting so far. Thanks for the tip, maybe a normal mode scan will be a little more forgiving. I will give it a try!

Thanks,
Craig

TerryT
06-29-2010, 09:25 PM
And again, as someone else said.... make sure the object is fastened down well (hotmelt glue, etc) so it doesn't bounce or vibrate.

TIMCOSBY
06-30-2010, 01:15 AM
is the bottom of your sled at least 3/4" material? if its 1/4" it may have some up and down movement too.

Capt Bruce
09-08-2010, 09:09 PM
My first and only scan and my reaction to the noise was much the same as yours. Like an old dot matrix printer while it tapped its way across a formerly gilded wood surface. As shown below it added a surface texture that was not there on the original. Had to clean it up using the filters before going to the final pattern from the scan.

38554

Now that I know from that experience what to expect I won't be so surprised or worried next time. Just count on having to do a little clean-up. Save copies as you work through those steps so you can go back if necessary.