Shopbot never cut flat

[ClayM325]

I have a problem that I cannot seem to figure out, as it doesnt make any sense. My machine has been doing this since i first purchased it new, just never mattered that much because of the way I use it, but I am wanting to do some 3d stuff so i need to get it taken care of.

Machine is a PRS standard (96x48), purchased late 2015 new. It cuts downhill as it goes down the y axis. But what I dont understand is, I can resurface the spoilboard perfectly flat, then switch in a different bit and zero near somewhere close to the end lets say x12 y12. If i run the spindle and move it down toward the end, it gradually starts cutting into the board more and more to im guessing .020" or so, maybe less. Any ideas? I have checked and rechecked level of the table as well as tightness of bolts and the hardwood surface under the vacuum plenum.

[pro70z28]

Does the Z axis drift? I don't have a clue if that could be the problem, just a thought. Maybe mark the Z axis rails to see if that axis actually moves down as the gantry travels? May not be the problem, but at least you could cross it off as a cause. I had an issue where the Z axis, while doing a series of up & down moves during a job, would go down slightly less every time it cycled up and down. After several moves it would no longer even touch the substrate. I thought I had a bad stepper, turned out to be a bad USB cable.

[Chuck Keysor]

Hello Clay. I had/have a somewhat similar problem. Though I bought my PRT alpha used.

I can resurface the table, then install a V-bit, and zero it at any point of the table, and move the bit around the table, and find that the tip of the bit does not remain the same distance from the table top...…. I posted on this a long time ago, and talked to tech support. I got all kinds of helpful suggestions from the forum, and tried out everything anyone could come up with. Someone suggested finally that my Z bushing may be worn, and upon removal, it was in fact worn.

And I started to re-align my Z axis after putting in the new Z bushing (I replaced all bushings while I was at it). But I never got back to it because I got caught up in making a big 3D model, that turned out to have big problems...… so I am not firing on any cylinders, so to speak.

Maybe by parking my machine, I am leaving it where I HOPE it will finally work after I finish the re-alignment.... better false hope than no hope....:rolleyes: Sorry to be of no help, other than to point out odd things like what you have going on, may not be easy to fix.

Good luck, and unlike me, press on! Chuck

[Brady Watson]

Clay - How is the sacrificial (top layer) board attached to your machine? Is it glued to the grid/plenum (if so what glue) or is it screwed down?

-B

[Red F]

It may be your surfacing pass and not your second cut.

Could a dull surfacing bit cause push itself up as it cut, or push the table material down?

When I'm problem solving, I try not to take anything for granted and start at the beginning. I'm just making a suggestion, not speaking from a much experience.

[ClayM325]

Does the Z axis drift? I don't have a clue if that could be the problem, just a thought. Maybe mark the Z axis rails to see if that axis actually moves down as the gantry travels? May not be the problem, but at least you could cross it off as a cause. I had an issue where the Z axis, while doing a series of up & down moves during a job, would go down slightly less every time it cycled up and down. After several moves it would no longer even touch the substrate. I thought I had a bad stepper, turned out to be a bad USB cable.

how/where would i mark them and look for the drift? like compare the gantry to the rail?

[ClayM325]

Hello Clay. I had/have a somewhat similar problem. Though I bought my PRT alpha used.

I can resurface the table, then install a V-bit, and zero it at any point of the table, and move the bit around the table, and find that the tip of the bit does not remain the same distance from the table top...…. I posted on this a long time ago, and talked to tech support. I got all kinds of helpful suggestions from the forum, and tried out everything anyone could come up with. Someone suggested finally that my Z bushing may be worn, and upon removal, it was in fact worn.

And I started to re-align my Z axis after putting in the new Z bushing (I replaced all bushings while I was at it). But I never got back to it because I got caught up in making a big 3D model, that turned out to have big problems...… so I am not firing on any cylinders, so to speak.

Maybe by parking my machine, I am leaving it where I HOPE it will finally work after I finish the re-alignment.... better false hope than no hope....:rolleyes: Sorry to be of no help, other than to point out odd things like what you have going on, may not be easy to fix.

Good luck, and unlike me, press on! Chuck

what and where is the Z bushing? doubtful since it was doing it brand new but worth a look!


Clay - How is the sacrificial (top layer) board attached to your machine? Is it glued to the grid/plenum (if so what glue) or is it screwed down?

-B

It is glued using Titebond 3, applied with a roller.


It may be your surfacing pass and not your second cut.

Could a dull surfacing bit cause push itself up as it cut, or push the table material down?

When I'm problem solving, I try not to take anything for granted and start at the beginning. I'm just making a suggestion, not speaking from a much experience.

Its possible, I need a new bit anyway so we will see, but like I was sayin its done this since day one basically

[Brady Watson]

Clay,
Get yourself a dial indicator. They are like $25-30 with a magnetic base from any number of places. Stick the indicator base on the YZ car and indicate the spoilboard. See if you can measure the deviation reliably in the areas you describe.

I would also put the indicator on the YZ car and measure along the bottom edge of the gantry beam and makes sure it isn't going up & down. I would also indicate/measure the length of your table and see if there are any dips etc.

There is no other tool more useful than a dial indicator for times when you can't find where the problem is.

Keep in mind that when your bleeder board starts to get thin, it can warp internally and pop up in the middle - plus other weird things. I'd say once it gets below 5/16" thick, it's game over. Titebond as good as it is can still fail on MDF if there is a significant amount of movement from atmospheric conditions. The glue itself doesn't fail, the MDF does - it will literally pull away and leave jagged break points.

-B

[dlcw]

I have a problem that I cannot seem to figure out, as it doesnt make any sense. My machine has been doing this since i first purchased it new, just never mattered that much because of the way I use it, but I am wanting to do some 3d stuff so i need to get it taken care of.

Machine is a PRS standard (96x48), purchased late 2015 new. It cuts downhill as it goes down the y axis. But what I dont understand is, I can resurface the spoilboard perfectly flat, then switch in a different bit and zero near somewhere close to the end lets say x12 y12. If i run the spindle and move it down toward the end, it gradually starts cutting into the board more and more to im guessing .020" or so, maybe less. Any ideas? I have checked and rechecked level of the table as well as tightness of bolts and the hardwood surface under the vacuum plenum.

Do you have large humidity swings in your area over the course of the day? (your info doesn't indicate where you are) This can cause the spoil board to swell and shrink. Gary Campbell described to me some of the issues he dealt with when his shop was in FL.

[jerry_stanek]

Are you using vac hold down

[ClayM325]

Clay,
Get yourself a dial indicator. They are like $25-30 with a magnetic base from any number of places. Stick the indicator base on the YZ car and indicate the spoilboard. See if you can measure the deviation reliably in the areas you describe.

I would also put the indicator on the YZ car and measure along the bottom edge of the gantry beam and makes sure it isn't going up & down. I would also indicate/measure the length of your table and see if there are any dips etc.

There is no other tool more useful than a dial indicator for times when you can't find where the problem is.

Keep in mind that when your bleeder board starts to get thin, it can warp internally and pop up in the middle - plus other weird things. I'd say once it gets below 5/16" thick, it's game over. Titebond as good as it is can still fail on MDF if there is a significant amount of movement from atmospheric conditions. The glue itself doesn't fail, the MDF does - it will literally pull away and leave jagged break points.

-B

Awesome I wil try that thank you. I have an indicator I used to measure case runout on ammo casings, im sure I can rig that up somehow



Do you have large humidity swings in your area over the course of the day? (your info doesn't indicate where you are) This can cause the spoil board to swell and shrink. Gary Campbell described to me some of the issues he dealt with when his shop was in FL.

I am in Raleigh, NC. We have large humidity swings but not quickly. This seems to happen immediately. I could resurface then immediately cut and it will do this.


Are you using vac hold down

Yes, I have 2 vacuubes, each piped to 2 quadrants of the table (4 total quadrants)

[Chuck Keysor]

http://www.talkshopbot.com/forum/attachment.php?attachmentid=31650&stc=1http://www.talkshopbot.com/forum/attachment.php?attachmentid=31651&stc=1http://www.talkshopbot.com/forum/attachment.php?attachmentid=31652&stc=1
Here are the images for my PRT. I have never seen a PRS in person but it must have some clear similarities...…. Chuck

[guitarwes]

I have the same problem, hadn't figured it out yet. I have a 2006 PRT Alpha. My bit runs deeper into the freshly surfaced spoilboard the futher you go down the X axis.
I contributed it to a very worn thin spoilboard and we put a new spoilboard on there and does the same thing. Very aggravating when you have Vcarvings on an entire sheet of material, the carvings on the X0 end are perfect but the X96 end are really deep.

[Brady Watson]

I have the same problem, hadn't figured it out yet. I have a 2006 PRT Alpha. My bit runs deeper into the freshly surfaced spoilboard the futher you go down the X axis.
I contributed it to a very worn thin spoilboard and we put a new spoilboard on there and does the same thing. Very aggravating when you have Vcarvings on an entire sheet of material, the carvings on the X0 end are perfect but the X96 end are really deep.

Do you have a car pump under the table at the 96 end?

-B

[bill.young]

When you bring it back to the starting point does it come back to the right height, right at the surface? That should tell you if your z-axis is drifting down or if something wonky is happening with your table or spoilboard?

[ClayM325]

When you bring it back to the starting point does it come back to the right height, right at the surface? That should tell you if your z-axis is drifting down or if something wonky is happening with your table or spoilboard?

yes it does

[Brady Watson]

Do you have a ̶c̶a̶r̶ vacuum pump under the table at the 96 end?


That should have read vacuum pump...They generate a lot of heat and do nasty things to your table glue up from the heat.

-B

[jTr]

While I'm not qualified to comment on the other mechanical variables, I wholeheartedly agree with Brady - Vac pump certainly seems to make a difference. New shop configuration = relocated vac box under table for space savings. I also have gone from a 208V in old location to 243V. Average temp in box was 120-130 degrees F. Now pushing 150. Table moves even more than it used to. General trend for the ply/mdf sandwich is to flex downward in the center as it heats. Again, even more pronounced now. If this is similar to your experience, you may want to investigate using a portable fan under table for extended cutting sessions.

Sidebar for others using the "black box" vac system /FWIW:
Ordered new pair of primary Lighthouse vac motors based on age of originals, sudden jump in temps and rated voltage. Though separated on the web page into 220, 230 and 240 volt ratings, Mr. Bartch says this is irrelevant, and these motors can run any voltages and up to 180-190 degrees and still be okay. Good news is system now is pulling a bit stronger, so I'm still happy with this "temporary" vac source at close to 6 years use.
( I'd like to have a real vac pump when I grow up;))

jeff

[dlcw]

I set my black box vacuum under my CNC table. To help with the heat issues, I have a box fan that blows air over the vacuum box towards a window where it is exhausted outside. It's not hot under my table at all.

[guitarwes]

That should have read vacuum pump...They generate a lot of heat and do nasty things to your table glue up from the heat.

-B

I do not run a vacuum (or car pump) at all.

[Chuck Keysor]

I don't run a vacuum either, no heat issues for sure for me.

[ClayM325]

Those little vacuubes arent bad for heat I dont think. Ive never noticed extra heat around my machine.

[ClayM325]

I have the same problem, hadn't figured it out yet. I have a 2006 PRT Alpha. My bit runs deeper into the freshly surfaced spoilboard the futher you go down the X axis.
I contributed it to a very worn thin spoilboard and we put a new spoilboard on there and does the same thing. Very aggravating when you have Vcarvings on an entire sheet of material, the carvings on the X0 end are perfect but the X96 end are really deep.

Exactly! Ever figure yours out?

[Chuck Keysor]

Hello Clay. Last fall, and a few days ago in a follow-up, I made the observation that I could not find any videos or training which teach how to accurately/properly align all of the parts of a Shopbot. It seems as though anyone having an alignment problem needs to rely upon small tips/advice from a small pool of experts. That seems to be a flaw to me, in that Shopbots are inherently unstable mechanically, and prone to being tossed out of alignment, IF they were ever even correctly aligned in the first place. Replace a pinion, bang the gantry with a hard stop or banging the bit, and you get even more out of whack. So the need for all Shopbots to be re-calibrated/realigned seems inherent with the machines. Yet, despite that obvious need to do a technically exacting procedure to a high level of accuracy, there is no training information available from Shopbot. I have looked on YouTube, and not found helpful videos. For example, there is a gentleman who demonstrates how to use a double headed dial indicator on a trammel bar. But he seems to be guessing as he goes along. And he never says how good you need to get, or how good you can hope to get. He says the process is exacting and tedious, and when he is all done, he demonstrates his results, which seem to not be to his satisfaction. As I watched, I came up with more questions then he gave answers.

I would expect my problem, and those of others on this thread with unsatisfactory results, are all at a loss for how to fix their problems because they lack professional machinist training, and are not able to find training appropriate to the needs of the typical shopbotter that wants to get good performance from their machine. To my thinking, because EVERY Shopbot has to be assembled and aligned before it can be used, that there should be really good training to explain all of the nuances needed to obtain good operation. So after all these years, I can't imagine that Botters haven't asked for this information, especially since there seems to be a regular set of questions from users with problems they can't figure out.

As noted, I posted my opinion on this last fall, and then a few days ago. I hoped that lots of people would chime in with a chorus of loud "Amen Brother". But only one person commented their agreement.

It used to be, to my understanding that you could hire someone from Shopbot to come and align your machine. I don't know what that cost, but it had to be a lot. Again, detailed video training on all aspects of alignment, plus showing the various effects on operation for various types of mis-alignment would be vital to having a happy pool of customers who had top notch operating machines which could produce top notch V and 3d carvings, without having to hire a tech to come to their machine, and then call them back every time the Bot got bumped...……. .

Sigh...………. Chuck