how accurate should a cabinet saw be?

[john tomljenovic]

so I bought this new saw (a pm2000) and it does weird stuff. namely when raising the blade it "heels" to the right about .020" at the far end of the blade. even when measured over the arbor we are looking at .008" of movement to the right. what gives? I pretty much determined the arbor and the pivot are out of parallel and I don't know what else can cause this.

so no problem, the powermatic people come out to see it. send me a replacement and this one does the same thing, albeit a couple thou less. Am I going nuts? or is expecting the blade to stay in the same plane by a couple thousands asking too much. really, I am not sure. the powermatic customer rep has not yet provided what tolerances I should expect, and I have asked repeatedly. what are the creekers experience with thier cabinet saws? what kind of readings should I expect?

and any other owners of a pm2000 chime in? how straight is your blade staying? a quick check is to put a machinist square to the back edge of the blade or better yet right to the riving knife. raise the blade and see how much of a gap is created. on mine you can definately see the blade move away from the square. do you see the light? or does your saw stay tight to the square?

powermatic is sending someone again to look at the replacement, actually both since I still have the first bum saw and they where incredible to hear both saws do the same thing. I am assuming there may be a bad batch of saws out there? Or I gotta sniff different glue.

anyway, I am of course really disappointed, after reading all the glowing reviews on this saw and I can't seem to relate to all that. both saws even came with the table and trunnions out of parallel and had to shim the front of table about a 1/16" just to keep blade alignment when tilting. I bought a new saw since I was getting peeved at my contractor saw going out of alignment at the slightest provocation, but at least I could get it aligned.

any info you guys can provide would be greatly appreciated.....

john

[Chuck Wintle]

In my opinion a table saw should be dead on no matter which way the blade is orientated. For this saw to be so off when new suggests something shifted during transport or they have a problem in the factory. I would check for any broken parts inside.:)

[john tomljenovic]

charles, I think you meant to say "for BOTH saws to be so off"...

anyway the tech and I looked at the first saw really close and found nothing broken. well at least without taking it apart piece by piece . and the arbor is tight, not experiencing any loose fitting parts or clunks.

[Chuck Wintle]

Sorry,
I misread your post. I did not realize they sent you a whole new saw.:o

[glenn bradley]

Hmmmm, two in a row sounds like an assembly or procedural problem at the factory. Building something that will keep an arbor aligned with a surface is not that tough when it retails around $2000. From what I've read I believe PM will make good. If not, $2000 buys a good saw from several builders.

[john tomljenovic]

I am sure the WMH people wil make good. I am just very afraid of saw #3.

has anyone else with a pm2000 encountered this problem? or any saw for that matter.

[Phil Thien]

You aren't raising the blade all the way to the top, are you? The blade movement is nonlinear at the limits.

[john tomljenovic]

nope, not hitting the stops. the blade moves to the side in a consistant manner as you raise the blade. I stop a half turn before the limit.

[Phil Thien]

had to shim the front of table about a 1/16" just to keep blade alignment when tilting.

john

Tell us more about this?

[john tomljenovic]

Tell us more about this?



if the the blade "heels" when going from 90 to 45 degrees of tilt, that demonstrates the trunnions and table are not parallel with each other. to rectify this on a cabinet saw you shim one end of the table. in my case, .055" added to the front between table and cabinet. which corrected about .020" of blade heel.

you would think a top dollar cabinet saw made today would be pretty spot on, not worse than a $500 contractor saw.

the new guy powermatic directed me to look at the new saw is not returning my calls for scheduling. looks like a call to powermatic is in order....:rolleyes:

[Tom Maple]

John,
I'm sorry to hear of this problem, not just for you but the many others with this same saw. I have a new PM2000 still in the box, and am now concerned with what I may find when I finally get it unpacked.
Let's hope WMH does a better job for you than Redmond did for Gary and his General Jointer. It should be straight forward in your case since you are dealing directly with the manufacturer.
We will be watching how this unfolds with great interest.
Best of luck,
Tom

[Jim Becker]

My understanding of cabinet saws is that (in theory) the table top should have been shimmed at the factory to insure proper tolerances on the blade/arbor to table relationship.

[Joe Jensen]

Shimming will help, but I'm not certain shimming will completely compensate. If shimming was the answer, I'd think the tech Powermatic sent out would have known how to do this. For those who haven't read about or experienced, this, the owners manual for SawStop explains it pretty well. Fundamentally, the surfaces where the trunion mounts to the moving carriage must be prefectly machined, adjustable, or you need to shim. I think shimming will only take car of 1/2 of the error if memory serves. I'll try to find the book on this topic when I return from this trip...joe

[Joe Jensen]

I googled and found this patent disclosure on a table saw heel adjuster.
Interesting...joe
http://www.patentstorm.us/patents/6502493-description.html

[john tomljenovic]

I think we are getting a little off track. my problem is not so much blade heel when tilting. That I can fix, completely with shims. although I do find it a little odd I need to use so many for this vintage of saw.

the real problem is blade heel when raising/lowering the blade. I don't see too much of an obvious fix. the arbor and it's pivot are not parallel, how do we adjust that?

although the tech that did come out to see the first tablesaw never has seen a pm2000 before mine. although he understood the problem I was having.

the serials on these saws are:

703 2000 2720
703 2000 2781

if anyone's saw falls between these numbers I would be looking at it real close.

I also wonder how many are produced in a day and if these two were made at the same time. the powermatic tech could not answer that question either.

I should also add that the powermatic and sawstop have completely defferent designs. the powermatic being more old school with a pivoting arbor. the sawstop has more in common with euro-sliders. the arbor slides on ways.

[Rob Bodenschatz]

the serials on these saws are:

703 2000 2720
703 2000 2781

if anyone's saw falls between these numbers I would be looking at it real close.


602 2000 0421 here.

This thread really has me concerned. I've had my PM2000 for a year & I'm embarrassed to say I haven't really taken a hard look at how well it's tuned. I really need to take some time & do that.

[john tomljenovic]

rob, if you get a chance to measure your saw and post the findings I would be very interested. your saw's # is pretty far from mine so I would think everything is OK with yours.

just put a dial indicator to the blade and work the blades elevation and see what reading you get. you can even try it on the riving knife to get a more exaggerated reading.

[Bob Feeser]

I have a Powermatic 66 5hp, pre-2000 release. I could write a book on the saw, both it's merits, and detractions. If you want to read more than you may want to, go to the following link at Funums. http://www.funums.com/showthread.php?t=116

I also did some digging at Amazon.com, on the PM2000. One of the reviewers stated the following, which may be part of the problem:confused:


Reviewer at Amazon>>>The boring and threading on the splitter mount was done off center of 90 degrees (making the blade guard and splitter useless, and a poor pulley/arbor fit out of the box, resulted in a wobble and related rattle and vibration that just would not go away. I spent hours trying to reset the setscrew, lock it with Loc-Tite, and moving the pulley slightly from side to side, to no avail, only to find out from Powermatic, that, "Yes, were having problems with some of those. Powermatic sent me a new arbor and pulley and bearing set, (different than original, so perhaps they knew FOR SOME TIME they had a problem with the other type of arbor, which would not secure well to the motor shaft, despite my repeated tries to pack it with various shims of paper plastic and steel to eliminate the annoying wobble.)

Back to Me>>>My comments on PM's accuracy of build on the 66 models. To be honest I was surprised at PM for having made a machine with such tight tolerances, but at the time, a total disregard for assembly procedures. It was as if the employees were pissed at the company, and were using their union as an excuse to create a less than stellar machine. Due to the fact that you could re-assemble the machine, by reaming the holes for the table top mounts, in order to get the blade true to the table, and too many others to mention here, but finally after redoing it, the saw is flawless. I have not used a dial indicator to measure the blade at full up, but a starret combination square, measures it without any slight separation. I checked the Starret square against a Bosch Digital protractor, which measures in tenths of a degree, equal to 3600 dots in a circle, not matter how small the circle. 360 degrees X 10 for tenths, and it wasn't off by even a tenth of a degree. Then when using the Starret square if you can see even a sliver of light, or for more exacting measurements, if you can slide a feeler gauge, which measures in pre-determined thousandth increments, in between your square and the blade, anywhere along its surface, then you know it is on or off by thousandths.
I hope the problem resolves properly for you. It seems that of the first 15 reviews I read at Amazon, none of them mentioned the same problem. Maybe others farther back did. It would be worth a look.

I just joined this site, and am excited to be here. Great site with helpful members.

[Phil Thien]

I think we are getting a little off track. my problem is not so much blade heel when tilting. That I can fix, completely with shims. although I do find it a little odd I need to use so many for this vintage of saw.

the real problem is blade heel when raising/lowering the blade. I don't see too much of an obvious fix. the arbor and it's pivot are not parallel, how do we adjust that?


I just brought it up in case both issues were related. Just in case you created one problem by fixing another.

So, I thought modern cabinet saws were more or less immune to blade heel resulting from changing the arbor angle? I thought that the balance of the motor was such that it provided equal forces on the front and rear trunnions regardless of angle.

I get blade heel on a contractor saw, where the motor is hanging out the back and changing the angle causes it to torque the support bars (and trunnions) unevenly. But nothing near those forces should be occurring inside your cabinet saw.

So if I'm right (and I'm not positive I am and I'm open to an explanation if I am wrong), I can't see where you'd see that kind of heel resulting from changing the arbor angle.

Possible problems as I see it: (1) That saw has something terribly wrong on the inside. Like something is assembled or cast incorrectly, or bent. This sorta makes sense because you have two matching saws with similar inconsistencies. (2) I guess it is possible you are confusing .01" with .001" (one-hundredth vs. one-thousandth of an inch). I had one friend complain about the fact that his blade/slot alignment was off by two-hundredths, when in fact it was .002" (two-thousandths--which is perfectly fine). But this doesn't seem to make sense because from reading your description you do seem to understand the difference. (3) You somehow are measuring it incorrectly. But again, who can argue with placing a square next to the blade and then raising it. You see the blade move away from the square. Hard to screw that up! Plus, going back to #2, if you can see it substantially move, probably moving in the hundredths range and not the thousandths.

So #1 seems the most likely. Sorry to bother you with all that. Let us know how it turns out. At least you get a free bump in case someone else knows something.

[David DeCristoforo]

You should not have to mess with shims or any other "fixes" or "workarounds" when dealing with a brand new saw that you shelled out two grand for. Some slight alignment of the blade/slot/fence would not be out of order but beyond that, I don't think so. Having said that, I have never been a big fan of PM, esp. since the mid 80's when, IMMHO, their quality went into the tank. As I see it, you have two options. One: Return the machine, get your money back and try a saw from one of the "other" major brand mfgrs. Two: Keep "trying another one" until you "finally get a good one". Actually, there is only one "good" option.

[john tomljenovic]

thanks for everyone contributing to the discussion. It will probably be a week or so till I get a tech to see the saw, although I will try to accelerate that as quick as possible! I actually have stuff to build! I am still curious if any one else with a pm2000 or any other saw has done this. I am certain there are 59 other saws having this problem.

I just never personally owned a cabinet saw before, and really don't know what to totally expect. But from messing with my contractor saw on a continual basis I at least learned the dynamics of the tilting arbor saw. and figured a cabinet saw would end the constant fiddling. I understand whats going on with these saws I just wasn't sure if I am asking too much. a lot of people on these forums and others brags about the blade aligned to .001". I just wanted that too, getting frustrated on my old saw with the burning, teeth marks, and blade wobble even.

My girlfriend always tells me my standards are always too high, so after two saws doing the same thing I am figuring I am nuts. I thought I could get a straight and stable blade for two grand, if I spend ten can I get that perfectly aligned blade? and stays that way?

I chose the pm2000 because it had a riving knife. I find that very important to me, I don't know of too many other options beside the sawstop, which I totally considered but was more than I wanted to spend for a saw I would eventually replace with a euro style format machine.
there is also that new grizzly, but when I checked one out at their bellingham location I was not totally impressed. I will give WMH at least one more chance to get me a saw that works. Or I can give up my euro trash dreams of a riving knife and "settle" for a pm66 or something.:cool:

[terry hansen]

I received my PM2000 in May of this year, SN 612 2000 2181. Haven't noticed problem but will check tonite and post results. Whew - checked and I have no "heel" that I can measure with a square and feeler guage. Good luck

[Bob Feeser]

Since PM sent out a rep, and then they replaced the saw, that is an indicator that you were not measuring wrong. You know first hand what the shipping charges are on a 500 pound plus machine. Nothwithstanding PM has to redo, rebox, resell, etc. the one you sent back. So onto the idea of just how accurate a cabinet saw should be.

Don't feel bad wanting to get accurate measurements to within a thousandth. If someone pounces on you for that, ask them, "Just how off is acceptable" What a lot of people are thinking is here you are beating yourself up to get a measurement that is not neccessary. I can understand that, but what if you were using less effort, than the sloppy methods, and getting consistent accuracy to within a thousandth every single time effortlessly. Let me explain.

I use an Incra Table Saw fence, on the PM 66, and I break out the dial indicator when setting it up, and get consistent cuts that are within a thousandth every time. Oddly enough the fence that comes with the Powermatic, I use on my Dad's vintage 1950's saw, and am able to get close to the same level of exactness, due to the smart way they designed the fence.
http://inlinethumb57.webshots.com/3896/2457906590100733997S600x600Q85.jpg
The reason why you can get that level of exactness out of the saw is that the accu fence (that came with the PM saw) has a hair line set up, that is unique. They put a line up indicator that is somewhat clear on the top, looking down through it, to the black line underneath. For some reason, being able to look down at it, you can see even the slightest edge, and I mean slightest edge sticking out on either side. So if the black line on the rule running along the bottom is out of line, even the slightest, you can easily see it. Once it is perfectly centered, not even a wisp of a line shows on either side.
When I saw the major competitors line set up, it came with 2 black lines, and it made it difficult to decipher the fine edge between the lower black line, and the upper one. The rule and the indicator were both black, which made it more difficult to tell.
Now that everyone wants to start yelling at me for even stating that a thousandths is important, let me tell you about the Incra Table Saw fence system.
http://incra.biz/assets/TS-III.gif
What that picture does not show you is the locking teeth in the clamping system where the red label is shown above. On the other side of that clamp is a row of teeth, that come to a sharp point every 1/32 of an inch. So once you reach a measuring line, using one of the rules, you clamp it down, forcing into perfect alignment when the teeth mesh with one another.
http://incra.biz/assets/rack.jpg
Then their is a fine adjusting knob that allows you to dial it in to a perfect (within a thousandths of accuracy).
http://incra.biz/assets/mic_knob.JPG
So you cut and tweak, cut and tweak until it is set just right. Once you dial it in, you can get accurate cuts to within a thousandth of an inch every time. (If the rest of the saw is set up right) I cut, then take out a dial caliper and measure, and that is what I get. It might be off by a thousandth. Interesting to note that Forest Woodworker II blade manufactuer advises you to set your fence at 1 or 2 thousandths wide at the exit end of the blade, so the blade teeth don't shave your cuts. So a thousandth isn't anything to be ashamed of, it is something to demand on every cut.

Let me ramble on about the Incra fence for a moment. Because of the row of teeth that you see in the pic above, you not only have incredible accuracy, guaranteed to within a thousandth, you also have perfect repeatability. When you make a cut at lets' say 5 1/8" then go make other cuts, then you realize you need a perfect mate to the original 5 1/8" cuts, when you set it at 5 1/8" setting again, and it falls into those teeth when you clamp it down, you are guaranteed to get perfect repeatability every time. This is nuts.
Another cool feature of the micro adjusting knob is that if you need to tweak a custom cut ever so slightly to make a fit, you can dial it in. Move it a thousandth or two, or 40, make another pass, and you have the right cut. The knob has lines in one thousandth increments. Once you are done with the custom setting, you move the knob back to dead on zero to maintain your accuracy.
This post is getting long, if you want to see more about the Incra Table saw fence, you can see it at Incra.com They also have a measuring rule that... Oh well, check out their stuff, lots of very exacting measuring tools. That is why they call it "Incra" mental. I call it Incra mania. It's ok to be exact. No fussing, fighthing, or settling, just perfect cuts every time.

[Jim Summers]

I got my pm2000 this year. Mine was out of parallel, but after an hour or so with a dial indicator and a rubber mallet That was within tolerances. The problem I had initially was I was trying to dial it in with the blade raised up all the way against the stop. I learned from fellow contributors here to not max out the blade. No issues with the blade being 90 degrees to the table during height adjustments though.

I am finishing up one project and I will do a tuneup before the next. I'll let you know how that goes and I'll get my serial number also.

Good luck!

[john tomljenovic]

just an update, I got hold of the tech that I worked with earlier and we are scheduled for monday. we both have yet to hear from the powermatic people. But I am still keen on what others have measured with there saws. that is...does the blade stay parallel and equidistant with the miter slot whether you are lowered or raised. thats all.

the problem is not with the fence. although I did have to replace the curser and tape. the stock items were too coarse with the hash marks.for my taste. I don't think I need an incra to stay parallel to the blade. My eyes are good enough and I am not confused with lines on a tape measure.

the blade also stays perfectly at 90 degrees to the table the whole time. parallel with the miter slot is another story, well actually the main one. I hate limit stops on any table saw and avoid them, I even back off or remove the stops for blade angle so I don't get near them when setting 90/45 degrees . I use a square and test cuts instead. if you watched your dial indicator spin wild when you even get close to those wretched limit stops you would see their uselessness too.

[Bob Feeser]

You mentioned> "I don't think I need an incra to stay parallel to the blade. My eyes are good enough and I am not confused with lines on a tape measure.
the blade also stays perfectly at 90 degrees to the table the whole time. parallel with the miter slot is another story, well actually the main one. I hate limit stops on any table saw and avoid them, I even back off or remove the stops for blade angle so I don't get near them when setting 90/45 degrees . I use a square and test cuts instead. if you watched your dial indicator spin wild when you even get close to those wretched limit stops you would see their uselessness too."

John I see what you are saying. I don't think you need an Incra either. I mentioned the great results with the stock accu fence, which is a super industrial grade fence. It is not an eye thing, for lining up the fence on the ruler, it is a mechanical design thing, which is the reason why it is so great. I was more commenting on the competitors fence, than I was trying to criticize the accu. As a matter of fact, I rave about it. Pertaining to the setting of the blade as to how it measures to the miter slot, I know that is critical. When having to ream out the cabinet mounting holes, so I could even get the blade paralell with the top, I measured off both the front of the blade, then was careful to spin the blade around to the rear setting to make sure that I had the same reference off of the blade, and checked the setting again. I did so because it is possible, as you know, to get a perfect distance on the front, and then the rear part of the blade could be way off. What you are mentioning is that you are getting a different reading when the blade is fully extended. I will have to double check that on mine.
Pertaining to the stops, I will check that as well. But the way I use my stops is a matter of touch. Moreso for the perfect 90 or 45 than anything else. I don't have the stops set at a grinding halt, but rather a softer touch. I would explain it as arriving at the point of gentle resistance, then tightening the setting knob so it doesn't get any play. I tried it that way, and was able to get perfect angles every time. I played with that setting for quite a while to get it just right.
By the way, you are dealing with a totally first class company, when it comes to backing the saw. They went out of their way to correct a few things in accessory parts on mine, without a whimper. So even though I was initially disappointed in the assembly, they more than made for it in follow through to get me happy.
I hope my input is beneficial.

[john tomljenovic]

I received my PM2000 in May of this year, SN 612 2000 2181. Haven't noticed problem but will check tonite and post results. Whew - checked and I have no "heel" that I can measure with a square and feeler guage. Good luck

thanks terry, this is the kind of info I am looking for. I almost missed your updated post with the giant incra ad and all.

I am not certain but I believe the first three digits to the serial is the date code. If anyone can correct me on that please chime in. at least I know what to ask for in a serial number before the third saw arrives, because I am not accepting any saw who's serial # starts with 703. And anyone else that is buying a pm2000 we know to avoid any saws that start with the evil 703.

people with pm2000, feel free to post findings about your saw and serial. thanks..

john

[Jeff Raymond]

Me told fashioned.

Hold square to blade; adjust fence to blade.

Cut two pieces.

If they match up for glue job.

Good saw.

[glenn bradley]

"so no problem, the powermatic people come out to see it. send me a replacement and this one does the same thing, albeit a couple thou less. Am I going nuts? or is expecting the blade to stay in the same plane by a couple thousands asking too much."

John,

Did the PM tech bring a plate with him as opposed to using the blade as the reference? My WWII is true at the edge but measurably inconsistent across the surface. Just a thought so you don't chase tail.

I've had the experience of putting quite an effort into setting up a "bad" jointer in feed table and just couldn't get it right. Felt that Homer Simpson 'Doh!' sensation when I realized my reference edge was bad.

I hope PM gets you to a satisfied place with this. None of us enjoys having to go through this stuff when we've spent our hard earned cash.

[D-Alan Grogg]

FWIW, I just checked my 3 month old PM2000, SN 70120002491. I didn't have any noticeable "heel" when raising the blade (referencing off the blade).

I'm not sure it would be desirable to reference off the riving knife because it may not be machined to the same spec's as the rest of the assembly and it isn't what is cutting the wood.

[john tomljenovic]

[quote=
John,

Did the PM tech bring a plate with him as opposed to using the blade as the reference? My WWII is true at the edge but measurably inconsistent across the surface. Just a thought so you don't chase tail.

[/quote]


No, he didn't need it. the problem was pretty obvious to the naked eye.

but I do use one of these in seting up a saw.

http://www.mastergage.com/display_product.asp?id=4

and I don't rely on the plate in it's entirety either. marking a point on the plate to make a reading off the same spot. although that is overkill and noticed little difference measuring that way or straight across the plate. just a way to eliminate any variables.

but thats all moot anyway since when raising the arbor it moves (to the right in my case) more than any errors found with the blade or riving knife. If I use a square and my eyeball, I see the same amount of movement whether I refrence off the blade, flat plate, or riving knife, the same amount of gap, (well more at the riving knife) but we still have all three tails pointing in the same direction.

I bring up the riving knife and use to measure in this specific situation because it is the farthest from the pivot. any errors will be greatest there. sure that punched little piece of metal is not the flattest. but it's flat enough for a quick and dirty test with a square. if, when raising the "riving knife" the blade stays close to the square you are most likely fine. (we are all testing both sides of the blade/knife right?) IF it moves away, well, then you need to investigate further. preferably with a nice machined flat plate. but then again, you will be able to reproduce my problem with any of the aforementioned items.

I realized I need to put this in picture book form. will get the camera out tomorrow.

[Phil Thien]

I realized I need to put this in picture book form. will get the camera out tomorrow.

Oooh! Pictures!

[Joe Jensen]

I think we are getting a little off track. my problem is not so much blade heel when tilting. That I can fix, completely with shims. although I do find it a little odd I need to use so many for this vintage of saw.

the real problem is blade heel when raising/lowering the blade. I don't see too much of an obvious fix. the arbor and it's pivot are not parallel, how do we adjust that?

although the tech that did come out to see the first tablesaw never has seen a pm2000 before mine. although he understood the problem I was having.

the serials on these saws are:

703 2000 2720
703 2000 2781

if anyone's saw falls between these numbers I would be looking at it real close.

I also wonder how many are produced in a day and if these two were made at the same time. the powermatic tech could not answer that question either.

I should also add that the powermatic and sawstop have completely defferent designs. the powermatic being more old school with a pivoting arbor. the sawstop has more in common with euro-sliders. the arbor slides on ways.

John, this is what the SawStop has an adjustment for and based on my understanding the geometries, I don't see how this can be fixed with shims.
Here is a quote from their user's manual.




Aligning the Blade to the Tilt Axis:


When making bevel cuts, the blade is tilted on an axis that runs along the surface of the table between the front and rear trunnion brackets. See Fig. 55. If the blade is not parallel to this axis, the blade will move out of parallel with the miter slots when it is tilted away from 0º (this assumes the table has been aligned to be parallel to the blade at 0º tilt angle). The SawStop cabinet saw is the only major cabinet saw that allows you to fine tune the parallelism of the blade to the tilt axis.

[john tomljenovic]

John, this is what the SawStop has an adjustment for and based on my understanding the geometries, I don't see how this can be fixed with shims.
Here is a quote from their user's manual.




Aligning the Blade to the Tilt Axis:
When making bevel cuts, the blade is tilted on an axis that runs along the surface of the table between the front and rear trunnion brackets. See Fig. 55. If the blade is not parallel to this axis, the blade will move out of parallel with the miter slots when it is tilted away from 0º (this assumes the table has been aligned to be parallel to the blade at 0º tilt angle). The SawStop cabinet saw is the only major cabinet saw that allows you to fine tune the parallelism of the blade to the tilt axis.


sorry joe, please stop relating the saw stop to the average cabinet saw. the internal designs are completely different. they even attach to the table/trunnion/cabinet differently.

you can correct the "blade heel when tilting" phenomenon by either manipulating blades relationship to the tilt axis (how saw stop does it) or manipulate the tables relationship to the tilt axis (how you do it on the average american cabinet saw). although in the american CS case that would be more compensatory then a true alignment, either way gets you to the same destination. BUT, that is if we are simply limiting ourselves to just the blade's alignment. If your table is not perfectly parallel to the tilt axis you get the same result. Saw stop avoids this by attaching the trunnions to the table. with an american CS You have a sheet metal cabinet getting in between the trunnions and table. introducing possible errors with table to tilt axis relationship. So thats where shimming comes in.

looking though the manual though, pages 50-52 describe the very problem I am having and how their saw can adjust for it. wish mine did....

[john tomljenovic]

her are some crude pictures. I start with the square butted to the blade then raise. the resultant gap is created. And yes, we are not hitting the stops. If any one sees their saw do this I would like to know.

[Greg Crawford]

John,

What a great post. You probably have more people looking at the inner workings of their table saws than any advertisment ever could.

No, you're not being too picky. You paid big bucks for what is supposed to be one of the finest saws in it's class. It should be right.

When my Unisaw arrived, I dug out the micrometer. To my delight, it was within .0015" of being parallel to the the miter slot. I also checked the fence alignment with a mic. I can get the fence parallel, but the fence (Biesemeyer) is not flat. It's livable, but it still bugs me. I have a TS Aligner, and used it to also check the arbor and arbor flange. After lightly cleaning the arbor flange with a fine file, I get about .001-.002" runout at the teeth of my WWII blade.

As for the heeling, it seems to be a safety issue if the rear of the blade gets closer to the fence as it's raised. This would tend to bind the piece and increase the chance of kickback.

I looked at my Unisaw trunions, and can't see what could cause that problem in this design. If the blade isn't parallel, it looks to me like it should be the same throughout the height adjustment. Even if the rod that the blade assembly (don't know all the technical terms) pivots on is bent, the angle should stay the same. All I can figure out is that the rod and bushings aren't true to one another, and the rod is fixed in one of the castings and as it rotates, it moves crooked in one of the bushings of the other casting (that is if the PM uses the same basic mechanism). No matter what the problem is, PM should make this right. They should also make sure you have the proper equipment to get your present job done.

After this post and the one about the General jointer, I'll be looking very hard at all my future machine purchases.

Please keep us "posted".

Greg

[Chuck Wintle]

her are some crude pictures. I start with the square butted to the blade then raise. the resultant gap is created. And yes, we are not hitting the stops. If any one sees their saw do this I would like to know.

so the picture on the left is with the blade lower and on the right with the blade moved to a higher position?

[john tomljenovic]

so the picture on the left is with the blade lower and on the right with the blade moved to a higher position?



yes, that is whats happening

[Bob Feeser]

Saw stop avoids this by attaching the trunnions to the table. with an american CS You have a sheet metal cabinet getting in between the trunnions and table. introducing possible errors with table to tilt axis relationship. So thats where shimming comes in.
....

John you bring up a good point, in that with the trunion attached to the table as opposed to the base, you can get a truer line up. On the PM66, touting the heaviest trunion in its class, they attach it to the base, which is a thicker grade of steel than you normally will find in "sheet metal". When I bought my PM66, it was in frustration after measuring across the top with a straightedge on my old 50's series Craftsman saw. The weight of the trunion, and in this case motor too, were sufficient to turn the saw table into a sway back mule. The reason why they attach it to the base instead is to avoid that very problem. Over time, a heavy cast iron trunion, attached to the table will warp it. Could you imagine placing a heavy cast iron piece, non stop on the table surface for lot of years, and not have it affect the saw table surface. My old Sears saw, with a cast iron table was so far swayed, that when cutting a kerf out of an 8' long board, it would kerf it to the proper depth at the beginning of the board, and the end of the board, but in the middle, where it would straddle the front and back of the table, it was actually off by 1/8". I was so frustrated by blowing fuses, and all of the other problems associated with a 1hp saw, when I discovered the sway back table, I went in the other room and ordered the PM66. (I always wanted one anyhow)
The Powermatic 66 is set up at the factory to be perfectly accurate to the top. Their are threaded adjustment rods, with allen wrenchable openings at the end of each rod, with a securing nut to lock in the setting. I never had to use a shim on anything. I am not making PM out to be God, because I had to reassemble some things when I got it, but once redone, it is totally accurate to within a thousandth on every plane. I can put a penny on the table top, turn on the 5HP 100 pound plus motor, instantly it is up to full speed, let it run for a minute, while the penny stands, then shut it off, let it come to a complete stop, and to my observers amazement, in a full standing position, blow a quick unexaggerated puff of wind from my lungs toward the penny, and it immediately falls over. It is a very dramatic demonstration, but emphasizes the advantage of having the heaviest trunion in its class; no vibration whatsoever.
By the way I saw the SawStop on the new This Old House, and was blown away by how effective it is. Not even the skin on a hot dog was affected, from just lying one on the top of the board. Phew.
Maybe Saw Stop put in some extra bracing or something to compensate for the aging sway problem.

[Joe Jensen]

sorry joe, please stop relating the saw stop to the average cabinet saw. the internal designs are completely different. they even attach to the table/trunnion/cabinet differently.

you can correct the "blade heel when tilting" phenomenon by either manipulating blades relationship to the tilt axis (how saw stop does it) or manipulate the tables relationship to the tilt axis (how you do it on the average american cabinet saw). although in the american CS case that would be more compensatory then a true alignment, either way gets you to the same destination. BUT, that is if we are simply limiting ourselves to just the blade's alignment. If your table is not perfectly parallel to the tilt axis you get the same result. Saw stop avoids this by attaching the trunnions to the table. with an american CS You have a sheet metal cabinet getting in between the trunnions and table. introducing possible errors with table to tilt axis relationship. So thats where shimming comes in.

looking though the manual though, pages 50-52 describe the very problem I am having and how their saw can adjust for it. wish mine did....

John, this is why I keep posting this. I know that shimming can help, but my understanding is that shimming just shifts the problem, and when shimmed, you are just averaging the error between 90 alignment and 45 alignment. If this were me, I'd insist on a new saw, or at least make certain that the post shimmed saw is aligned at both extremes...joe

[Joe Jensen]

John you bring up a good point, in that with the trunion attached to the table as opposed to the base, you can get a truer line up. On the PM66, touting the heaviest trunion in its class, they attach it to the base, which is a thicker grade of steel than you normally will find in "sheet metal". When I bought my PM66, it was in frustration after measuring across the top with a straightedge on my old 50's series Craftsman saw. The weight of the trunion, and in this case motor too, were sufficient to turn the saw table into a sway back mule. The reason why they attach it to the base instead is to avoid that very problem. Over time, a heavy cast iron trunion, attached to the table will warp it. Could you imagine placing a heavy cast iron piece, non stop on the table surface for lot of years, and not have it affect the saw table surface. My old Sears saw, with a cast iron table was so far swayed, that when cutting a kerf out of an 8' long board, it would kerf it to the proper depth at the beginning of the board, and the end of the board, but in the middle, where it would straddle the front and back of the table, it was actually off by 1/8". I was so frustrated by blowing fuses, and all of the other problems associated with a 1hp saw, when I discovered the sway back table, I went in the other room and ordered the PM66. (I always wanted one anyhow)
The Powermatic 66 is set up at the factory to be perfectly accurate to the top. Their are threaded adjustment rods, with allen wrenchable openings at the end of each rod, with a securing nut to lock in the setting. I never had to use a shim on anything. I am not making PM out to be God, because I had to reassemble some things when I got it, but once redone, it is totally accurate to within a thousandth on every plane. I can put a penny on the table top, turn on the 5HP 100 pound plus motor, instantly it is up to full speed, let it run for a minute, while the penny stands, then shut it off, let it come to a complete stop, and to my observers amazement, in a full standing position, blow a quick unexaggerated puff of wind from my lungs toward the penny, and it immediately falls over. It is a very dramatic demonstration, but emphasizes the advantage of having the heaviest trunion in its class; no vibration whatsoever.
By the way I saw the SawStop on the new This Old House, and was blown away by how effective it is. Not even the skin on a hot dog was affected, from just lying one on the top of the board. Phew.
Maybe Saw Stop put in some extra bracing or something to compensate for the aging sway problem.

BTW, the SawStop trunion attaches to the base not the top. I've had the top off of mine, and the store where I bought keeps one with the top off on display, and the trunion clearly mounts to the base...joe

[Jim Summers]

I finally remembered to get my serial number. It is:

612 2000 2075

No problems noticed. But will be doing a tuneup very soon. I'll holler back if I discover anything like you have been experiencing.

Take it easy.

[Tom Henderson2]

John-

When do you expect to have the PM guys come out again? I've kind of lost track of where this stands.

I've been shopping for a used cabinet saw, and came close to a used Unisaw a couple weeks ago but the deal fell through. I was giving some serious thought to just popping for a PM2000 since I like the built-in wheels and riving knife. But reading this thread has given me pause.

So by all means keep us informed about how this plays out.

-Tom H.
Ventura, CA

[john tomljenovic]

The powermatic tech is scheduled to come out monday morning. interested myself what will occur..

[Jim Summers]

Hi John,

While I was out messing around in the shop tonight I did some testing. I have a Grolz square and held it up against the blade and then raised and lowered it. No visible light came through or anything odd. The miter slot to blade parallel test is still set at .001 that I got it to about two-three months ago.

Sounds like PM has a bad batch. I am sure they will make good on it though. They were very accomodating on shipping me extension tables until I got a flat one.

Good Luck!

HTH

[john tomljenovic]

thanks Jim, I also feel there is a bad batch of pm2000s out there. feedback like yours helps bring that to light. at least we can get that info out there and let others not have to go though this drama I have been dealt. I wonder if powermatic is even aware they have a bunch of bad saws. I am going to ask them and see what they know.

[Noah Katz]

John,

I thought you mentioned it in one of your first posts that you thought the blade rotation axis is not parallel to the arbor pivot axis.

If that's the case, no amount of shimming anywhere will solve the problem; the angle between the blade and the table will change continuously with blade height.

I'll see what the blade does on mine with the Wixey digital gauge tonight and report back.

"You aren't raising the blade all the way to the top, are you? The blade movement is nonlinear at the limits."

That's news to me, and disturbing at that.

"Because of the row of teeth that you see in the pic above, you not only have incredible accuracy, guaranteed to within a thousandth, you also have perfect repeatability."

Too bad they don't make a Bies type fence with just the indexing rack and fine adjuster; the amount of real estate the Incra fence takes up is ridiculous.

Anyone know if Incra sells parts?

[john tomljenovic]

my WMH technician came by monday and confirmed my assertions about the saw(s). of course powermatic will send me another one, but even my tech is now in discussions with PM's quality control people since I will not be the only one to have this problem. And they will have to measure the replacement before going out of course.

third time is a charm right?

noah, you are correct that no amount of shimming will fully correct my problem. But the arbor/tilt axis relationship or even arbor/pivot relationship if askew, will show some symptoms similar to a table/tilt axis problem, and in that case of the table being out of whack (which is the usual case with an american style cabinet saw) shimming is the corrective action.

[Noah Katz]

Whew, checked mine, no angle change at all from full down to full up.

[john tomljenovic]

noah, on my saw(s) the blade angle stays consistent when raising/lowering blade. It is the blades alignment to the miter slot that changes. with the back of the blade drifting to the right around .020" as you raise.

[Bob Feeser]

John,

Too bad they don't make a Bies type fence with just the indexing rack and fine adjuster; the amount of real estate the Incra fence takes up is ridiculous.

Anyone know if Incra sells parts?

Noah,
I agree the Incra does take up a lot of real estate. It is off to the right side of the table, that I do not use anyhow, unless I am ripping extremely large stock. My feelings on the change of position of Incras home territory is a matter of geometry. If you grab the holding end of a broom stick, the leverage at the other end is hard to control, albeit, Beis and PM's accu fence, are made out of heavy enough metal to offset that.
Be that as it may, with the Incra positioned dead center, with a wide, something like 5" bar it creates an even playing surface, and avoids any slight problems with forces at the deep end from causing any slight change in the spacing between the blade and the fence. Like I said, the monster metal in the Beis offsets any concerns for that, but what about the rest of the fence world.
Once you lock the main center post, all you do is lock the edge of the fence close to you, and it is all you need. If you like, their is a trailing end of the fence that locks on that end as well.
As far as the Real Estate, I'm not using the right side table for anything anyhow, except when I am ripping 4 x 8 sheets of ply, then I get cuts as wide as 52". To do that I have to release 4 knobs, move the assembly back to the rear of the side table, and reattach. The way they do that is ingenious, with adjustable stops, maintaining a metal to metal, absolutely flawless registration. You adjust it until you get it perfect, then just like in cutting pieces with perfect repeatability, the reset of the fence gives you perfect repeatability as well. The advantage of the Beis in that regard is that their is nothing to change. If you have the side table long enough, all you have to do is slide it to a wide setting, just like any other setting. Sweet.
Like I said, I love the accu and the Incra, and use both, but for perfect repeatability, down to a thousandth, which you have to admit is very important, nothing beats the ease of accomplishing that as the Incra does.
It is funny that you mentioned that it would be cool if Beis had a saw tooth aligning mechanism as an option. I was thinking about how to engineer that very thing. The problem is getting the fence to align perfectly, while tightening down on the saw teeth alignment mech. Since the saw teeth mechanism is not something you want to be adjusting on different planes, they would need to put the saw tooth mech on the bottom, to act as a secondary brace/alignment. The saw teeth would need to engage first, then by further pressing the lever down, you would lock the fence all the way. I coud see it being possible. The locking arm would need to be a compound lever.
I think the Beis has the fine adjusting screws pressing on the metal that rides the edge of the bar like the Accu does. (After all the accu is a knock off of the Beis, at least I think it is)

[Tom Maple]

It's good that the rep came as promised and will make good on providing a new saw. If you are able to get them to relate which serial numbers are affected (assuming they are able to narrow it down) it would be helpful to other PM2000 owners to have that information.
Thanks for keeping us up-to-date.

[john tomljenovic]

In an earlier post....


I....... although he understood the problem I was having.

the serials on these saws are:

703 2000 2720
703 2000 2781

if anyone's saw falls between these numbers I would be looking at it real close.



And the first 3 digits are the date code. so 703 would be march of 2007; a 611 would be november of 2006 and so on.

[Noah Katz]

Bob,

I'm not saying the Incra isn't great, but as you point out, Bies type fences are nearly as robust w/way less complexity.

To get the Incra functions, the Bies would have to be redesigned to make room for the toothed rack on the rail.

Actually, a 32 TPI shaft could be hung underneath, and a clamping split nut added to the head.

Adding microadjustment to that is another level of complexity.

But as has been pointed out, you can get pretty darn close with the sight gauge.

[Bob Feeser]

Bob,

I'm not saying the Incra isn't great, but as you point out, Bies type fences are nearly as robust w/way less complexity.

To get the Incra functions, the Bies would have to be redesigned to make room for the toothed rack on the rail.

Actually, a 32 TPI shaft could be hung underneath, and a clamping split nut added to the head.

Adding microadjustment to that is another level of complexity.

But as has been pointed out, you can get pretty darn close with the sight gauge.

I found a review online at http://www.sawstop.com/media/W184JOp46-p51.pdf the has a table of specs, the third column is fence deflection. As the author noted, he exerted 6 pounds of force on the back end of the fence, and measured the deflection. You can see the results. With the Incra their is absolutely no fence deflection, none at all. All you have to do is lock the center post off to the right, then lock the knob right in front of you. Also you have the option of locking the other end of the fence as well with another knob. If you do that, the fence is locked front, back, and in the middle. Net result, you are not getting .006 of deflection, or .001, but rather .000 when cutting heavy stock.

I think it is entirely possible to cut heavy stock while not exerting any force sideways on the back of a fence, but I like to be able to exert a little extra force to make sure the workpiece does not drift from the blade. It's just one less thing you have to juggle when cutting a piece.

On the other hand, with Incra you do not have the luxury of installing a jig that drapes over the fence. A lot of jigs are made that way. The problem is the Incra has the center post getting in the way. So back to the Accu or Beis for that one.

I want to say something about the Powermatic. The one I received had to be more or less reassembled. The engineering is superb, the guys at the factory (mine was made in USA) were disgruntled because they were shifting to overseas manufacturing with the rest of their line) so they took it out on the saws while assembling. You could not even get the blade to run paralell to the miter slot. I had to remove the table top, and ream out the holes/slots in the cabinet base, just to be able to get it in perfect alignment. So their is a top plate with the slot holes in it, which gets welded to the cabinet base, and they didn't take the time to line that up right. So I had to drill the slot a little wider. All told their was a list of 7 items I had to remanufacture. I spent hours getting it right, but after I did, I am getting settings that are not off by even a thousandth anywhere. If I had to do it all over again, I would get another one. Maybe it is me, because I have been dreaming of the PM66 for more years than I can remember. With the polished like glass top, and overall quality of the components, I can't imagine anything better, although I see from the Fine Woodworking reviews it has a lot of company.

I also want to say that the accu fence, the Beis imitation by PM cuts a piece so exact on my old Sears saw, that you can measure it, and it's ridiculously accurate in every dimension; net result flawless results. I love Beis, I love accu, I love Incra, I love Jet (I have their oscillating sander) I love Delta, I love them all. I wish I had a shop large enough to decide which color am I in the mood to work on today.

It's not yeah this and boo that, It's yeah this and yeah that. Trying to find the right saw or fence when considering price, accuracy etc, once you get close to the top, lets just say we are all one big happy family.

[Tom Henderson2]

John-
What is the latest your blade alignment issue? Any action from PM?

-Tom H.
Ventura, CA

[John Hedges]

I was setting up my new PM2000 last night. I tried this test first off and started to panic when I saw the same thing, until I realized that it kicked over right when it hit the limit stop. It was true as could be till that point. BTW my serial starts with 704, so I assume they fixed the problem on newer saws.

[Jeff Raymond]

Saws only cut 'straight' theoretically.

Every saw has a different drift angle, assuming it is built correctly. Once you determine that, you can compensate for the angle.

[Cliff Rohrabacher]

so I bought this new saw (a pm2000) and it does weird stuff. namely when raising the blade it "heels" to the right about .020" at the far end of the blade. even when measured over the arbor we are looking at .008" of movement to the right. what gives?

You need to send the thing back. something is broke, bent, missing, not assembled right, or all six.

[Noah Katz]

"Every saw has a different drift angle, assuming it is built correctly."

Could you elaborate?

[Bob Feeser]

so I bought this new saw (a pm2000) and it does weird stuff. namely when raising the blade it "heels" to the right about .020" at the far end of the blade.
what are the creekers experience with thier cabinet saws? what kind of readings should I expect?
does your saw stay tight to the square?

john

John,
First I want to apologize to you for stomping all over your thread, with information about the Incra fence, which doesn't have anything to do with the problem you are experiencing.
I have a PM66, that I bought in 2004. I checked the squareness of the blade to the table at different heights, including full up, tight against the stop, and it reads square. The starret square is perfectly mated to the blade. I am referencing it off of the cast iron top, not the insert.
So the results you are getting are not standard results for a PM saw. It is interesting to note that Fine Woodworking rated the PM2000 and the Saw Stop as their editors choice awards. The 66 did not make it to the top. I think that is due in part to the poor assembly of what is otherwise an incredible saw. In order to get the blade perfectly parallel to the miter gauge, I had to do things such as re-drill / elongate the openings for the top plate, they weld onto the cabinet, that has the holes, that the cast iron top is mounted through. By taking the cast iron top off, it is easy to see that they did not even bother to take the time, to even eyeball it's position before welding. I don't want to get into the controversy, but the 66 saw is the last item that PM was manufacturing in the US, and the rep told me that the assembly problems could be a result of that. The PM2000 is manufactured in Taiwan, to Powermatic specs. I have the PM 8" 60B jointer, and their hollow chisel mortiser, which were manufactured in Taiwan, and they arrived perfectly setup, true and parallel in every plane. Although I did have surface rust spots on the jointers bed, that I was able to easily sand out, and Amazon compensated me for having to do that.
I am not knocking PM equipment, just the saw that I received, and many others according to the online reviews, were not given the kind of care and respect a machine of this magnitude deserves. If I had to do it all over again, I would buy another one in a heart beat.
After my experience I am sure PM will make good on it. I did not notice a concluding post on how you got it resolved. Please update with your progress. I'm sure everyone wants to know. I hope this is beneficial, and sorry once again, for the Incra stuff.

[George Carion]

I've been killing myself over adjusting my new PM2000. I have this problem +.

When raising the blade it moves towards the fence by .010. It's enough to see with a square. Also, and this one bothers me even more, my blade will not stay parallel to the miter slot when raising and lowering.

I set the blade parallel to the miter slot with the blade near its top limit. When I lower the blade, not only is it moving away from the right miter slot, but the rear of the blade is heeling towards the fence. As near as I can measure it comes inwards towards the fence by at least .005.

My serial number is 603 something.

[Bart Leetch]

Hmmm I don't necessarily believe the saying that "They don't make them like they used to."

But I do work in a shop with a 66 year old Unisaw & it is accurate.

I sincerely hope that you get you PM problems solved.

[Yuchol Kim]

Joh, I've had a SIMILAR problem with my PM 2000 (not exactly the same but worth looking into). When raising and lowering the entire blade would move right and/or left. It took me 3 days to figure why what was going on. The arbor pivots on a big king pin (bolt) as you move up or down. The arbor is secured to the king pin with a set screw. Once the set screw comes lose, arbor can move about 1/4" to the left or right. Once I tighted that screw, my blade remained true. What I am wondering is that why on earth would they design a powerful saw like this only to have the arbor held on to the king pin with a set screw? With all the power/torque, I can't imagine that being the best design.

[Bill Wyko]

so I bought this new saw (a pm2000) and it does weird stuff. namely when raising the blade it "heels" to the right about .020" at the far end of the blade. even when measured over the arbor we are looking at .008" of movement to the right. what gives? I pretty much determined the arbor and the pivot are out of parallel and I don't know what else can cause this.

so no problem, the powermatic people come out to see it. send me a replacement and this one does the same thing, albeit a couple thou less. Am I going nuts? or is expecting the blade to stay in the same plane by a couple thousands asking too much. really, I am not sure. the powermatic customer rep has not yet provided what tolerances I should expect, and I have asked repeatedly. what are the creekers experience with thier cabinet saws? what kind of readings should I expect?

and any other owners of a pm2000 chime in? how straight is your blade staying? a quick check is to put a machinist square to the back edge of the blade or better yet right to the riving knife. raise the blade and see how much of a gap is created. on mine you can definately see the blade move away from the square. do you see the light? or does your saw stay tight to the square?

powermatic is sending someone again to look at the replacement, actually both since I still have the first bum saw and they where incredible to hear both saws do the same thing. I am assuming there may be a bad batch of saws out there? Or I gotta sniff different glue.

anyway, I am of course really disappointed, after reading all the glowing reviews on this saw and I can't seem to relate to all that. both saws even came with the table and trunnions out of parallel and had to shim the front of table about a 1/16" just to keep blade alignment when tilting. I bought a new saw since I was getting peeved at my contractor saw going out of alignment at the slightest provocation, but at least I could get it aligned.

any info you guys can provide would be greatly appreciated.....

john
I just went through this with my contractor saw and they ended up replacing it. The problem didn't occur until I tilted the blade to 45 degrees. Great customer service at PM though.

[George Carion]

I went through this thread, but haven't really seen a consensus opinion on where (at which point) blade deviation becomes a problem.

What do all of you think? is a shift (front to rear) of .002 too much when raising the blade or is it really closer to a hundredth (.010) where we should start worrying about performance/safety. I'm sitting at .005.... which is pretty small, but I'm getting burn on my cuts. Since the saw is new I'm not sure if the burn is due to the blade moving or the riving knife being misaligned.

Thanks.

[Chuck Wintle]

Just for sake of argument how parallel should a blade be to a miter slot? Is 1 to 2 thou close enough? :D

[George Carion]

I took the time to measure my blade movement carefully. The issues I'm having are very close to the measurements discussed way back at the beginning of this thread. Sounds like I got the same saw :). Measuring above the abor while raising the blade it gradually moves towards the fence by .020 while losing at least .005 of it's parallelism to the miter slot. Joy.

Called WMH about the problem and the immediately said the saw needed to be replaced. I was surprised at how ready they were to accept my explanation. When I asked if this was a common problem they tech guy at WMH said he'd only seen one other saw with this problem. Hmmmm.

Anyway, now I'm working through Amazon to get a new saw sent over and have my current saw picked up. What a hassle.