Board shifts in machine part II

[Dale]

This is an on going problem I have had for several years and I am once again trying to figure it out, my first post on this problem was on 8/10/2014. The problem I am having is that eventually I end up with an 1/8th inch gap between the board and the left side of the squaring plate.
The board is perfectly square and flat (I have the same problem when I use the Universal Carrier Board) as was discussed in my post from 2014 I have checked the following:

head pressure - 85lbs,
the head is level within .001 on a 14" wide board (which I think is good?)
checked the bolts for tightness on the vertical posts
the out feed tables are adjusted properly
I took the belts off and made sure there was no sawdust build up in the springs

I have now discovered that the board stays in contact with the squaring plate as the machine runs the board all the way to the left during the measuring process but when the board starts moving back to the right as soon as it contacts the right pressure roller the board will shift a little away from the squaring plate on the lefthand side then as the board travels it's length again I end up with the gap on the left end of the squaring plate and the right end still in contact. Does anyone have any suggestion?

[fwharris]

Are you using sandpaper or rubber belts? Is sandpaper and chance the belt end is rolled up on the key pad end?

[Dale]

No rubber belts, they look good

[DianMayfield]

Any chance there is gunk built up in the roller ends, causing uneven pressure?

[Dale]

Had not thought of that, will check it out tonight.

[bergerud]

Sounds to me like the squaring plate is not adjusted to be square.

[DianMayfield]

Dan, my sliding plate rocks from front to back, maybe as much as an 1/8in. Do you know if this is normal? Don't mean to hijack, but I keep forgetting to ask! It sometimes allows a board to drift a bit. I can usually fix it by making sure that my outfeed rollers are square to the machine.

[fwharris]

Dan, my sliding plate rocks from front to back, maybe as much as an 1/8in. Do you know if this is normal? Don't mean to hijack, but I keep forgetting to ask! It sometimes allows a board to drift a bit. I can usually fix it by making sure that my outfeed rollers are square to the machine.

Dian,

The sliding plate should have some play in it and it should never be placed tight up against your board. If it is up tight it could cause the board to jam and strip out the X gears.

[DianMayfield]

Thanks Floyd! That's one more worry down. I do stick a business card in between, helps with board tracking errors. :)

[Dale]

Thanks for all the inputs, I have checked the squaring plate several times and it is dead on. I have not had a chance to check the compression rollers for gunk build up yet.

[bergerud]

I do not know how you checked the square of the squaring plate, but the behavior you describe is exactly what happens when the squaring plate is not square with the belts.

[Dale]

I've never checked the belts being square to the plate, as I did not know there is an adjustment to the belts. I've always checked the plate being square to the bit using a framing square aligning the bit at the keyboard side and then at the far side. Will check this out.

[bergerud]

There is no adjustment to the belts unless a positioning tap is broken off on the base. Square with respect to the belts, however, is what matters.

[Dale]

OK, not sure if this is a good check or not but put a 3" wide straight edge against the the squaring plate. The ribs on the left belt are parallel to the straight edge, the ribs on the right belt are cocked off about one rib's worth. Loosened the four mount bolts and there is no play to straighten it out.

[bergerud]

How about measuring the squaring plate with the framing square relative to a small board placed vertically between the belts?

[Dale]

Checking it that way both belts are square to the squaring plate. Any more ideas?

[fwharris]

Load the board up tight against the squaring plate and put in a sharp pointed bit (1/16" carving, 60/90 V bit) in the chuck. Lower the head down so the bit just touches the board enough to leave a mark. Manually move the board through the machine while insuring that it keeps contact with the squaring plate. After making the length of the board with the bit measure the mark from the edge to see if it is equal the full length of the board.

[Dale]

Did that and the line is parellel to the edge of the whole length of a 3' board.

[bergerud]

What happens if you place the board away from the squaring plate? For example, place the board with a 1/8" gap between the board edge and the squaring plate. Now watch what happens to the gap as the board is measured.

[Dale]

Tried that procedure using 1/8" shims to set the distance from the squaring plate to the board and then removing them before starting the measuring process and I kept getting a tracking error because the board would run off the encoder. Reduced the distance to 1/16" inch and when the machine stopped after measuring the lenght I had 1/16" gap on one side and 3/16" on the other

[bergerud]

(I assume the board was back in about the same place after the measurement.) So, I guess that it must either be the belts or the rollers above??

Check those rollers above. Try the experiment with minimal head pressure?

[Dale]

I've visually checked the rollers and blew them out with air, both rollers have play in them length wise but not side to side so I would think that is OK. Guess it's possible for a bearing to be bad under pressure. Do you think it would be worth it to pull the rollers out? Not sure how much of a job that is but I could try it.
I will try lower head pressures first.
Thanks for the help!

[mtylerfl]

Perhaps the head is not Level, and causes the pressure to be "lopsided" across the board width? Worth checking if you haven't already.

If you want to try the minimal head pressure without disassembling anything, just lower the head slowly with some downward hand pressure as the head goes down onto the board. Stop turning the crank before it clicks.

[fwharris]

Could it also be possible that the head is not level right to left as you are looking at the key pad? Just a WFG!

[Dale]

Head level has been checked before - a .001 difference between keyboard side and other side.
Just tried using lower pressure,used just enough to get both roller switches made - no difference to problem.
Removed both pressure rollers and everything looked good, no sawdust build up or any "gunk". Found out they use bushings not bearings.
Tomorrow I will check level left to right. Not really sure how to do that, compare distance between belt and compression roller?

[fwharris]

Not really sure how to do that, compare distance between belt and compression roller?


I would do it the same way you check the level across the width of the machine. Measure from the flat area of the corner post to the top of the head.

[Dale]

That makes sense, I'll let you know tomorrow what i find.

[bergerud]

I think it must have something to do with the belts. Are you sure all of the alignment tabs on the base are all there? If they are and the belt assemblies are aligned, maybe you have to look at the assemblies themselves. Could, for example, one of the rollers be sliding back and forth? There are some set screws which hold the position of the the belt rollers.

[mtylerfl]

Dan's suggestion is a good one to check. I was about to suggest inspecting the x axis gears for funk or damage as a last resort, since you seem to have already checked just about everything else. :):)

[Dale]

Michael. I had the belt assemblies off yesterday and I checked all the x gears.

Dan, I think it has something to due with the belts also. I pulled them off and removed the belts mainly to see if the springs were weak or sticky. I did not check for any loose screws. Where did you find that diagram? Sure could have used it yesterday.

Thanks everyone for coming up with all the ideas. I am determined to find the problem this time. Like I said I have been living with this for over three years. The machine carves fine it is just that all projects are cocked off at an angle sometimes it is not noticeable and sometimes it is. When it is I then have to mess around with a table saw to get all four side parallel to the carved region.

[bergerud]

The diagram is from the rubber belt installation instructions.

Maybe you have a defective belt of some kind. If you take the belts apart again and do not find the problem, try flipping the belts over. If it is a belt, maybe the problem will reverse direction.

[Dale]

Checked the head level left to right and it is good.
Flipped the belts starting with the one that the ribs were not parallel to the squaring plate and after I re-installed it checked the ribs again and they are parallel. Ran a board through it and it is staying against the squaring plate even after running it back and forth several times, so am fairly confident I have it fixed! Now just not sure if there is something wrong with the belt or were the springs not applying even pressure. Guess i will wait to see if it comes back and if it does I'll try new belts.

Thank you everyone for your help.

[mtylerfl]

That's great news! Glad you got it licked.

[Digitalwoodshop]

No internet for 2 weeks and see all the fun I missed.... I was having the same problem but with sand paper belts... Remember I rip my belts... it was getting to be about 8 inches wide... Replaced with new sand paper belts and the walking problem stopped... As in walking away from the key board side.

I like the idea of flipping the belts....

The 1 to 4 inch storm turned out to be 20 inches and with and wind trees came down.... Took the house wires and shop wires down... no power for 7 days. Fun fun...

AL

[mtylerfl]

Wow! Glad you're back and "electrified" again, Al!

Did you live off bear meat or rattlesnake? :rolleyes:

[DianMayfield]

Al, glad to see you are back up!! I had fun once, it didn't look anything like that :)

I have a new sandpaper set of belts that came with my used A machine. Could you use them? I converted to rubber. Your address is no longer in my PMs, so you will need to send again.

[Digitalwoodshop]

Would love them.... Sending PM