RAS table shows flat all ways by level and 'light under the straightedge' methods.
Checking by arbor height showed a small but consistent dip in the center as compared to the 45 positions.
Took the table adjustment screw to max, then added some small SS plates as shims between base and table and took it as far as I could.
Comes out the same way.
Since the gap appeared the same for the whole travel at any given angle, I thought I could live with it.
Nope. Crosscutting starts fine, then the kerf turns to dado a few inches out.
Suggestions?
Craftsman 2.75 HP saw. Original table covered with MDF. Both layers show same gap.

lid wrote:
RAS table shows flat all ways by level and 'light under the straightedge' methods.
Checking by arbor height showed a small but consistent dip in the center as compared to the 45 positions.
Took the table adjustment screw to max, then added some small SS plates as shims between base and table and took it as far as I could.
Comes out the same way.
Since the gap appeared the same for the whole travel at any given angle, I thought I could live with it.
Nope. Crosscutting starts fine, then the kerf turns to dado a few inches out.
Suggestions?
Craftsman 2.75 HP saw. Original table covered with MDF. Both layers show same gap.

From your description I have no clue what you're really talking about.
A "small dip" in the table should make no difference at all on a
crosscut kerf---oh, unless you're bevel-cutting?

However, what "table adjustment" screw(?), and if you only have a
"small" dip that is less than measurable w/ a straight edge, then
either your straightedge isn't straight very straight at all or the
rails on which the saw is running aren't. And any adjustment "to the
max" is rarely consonant with a "small" problem.

How old/new of a Craftsman is it? Is this one of the Sears "Peak" HP
numbers? Sounds suspicious, nobody real rates anything as 2-3/4 hp, so
if it is, sounds like a light-weight saw, maybe, that quite likely
never will be very accurate. Some of the older (and particularly
larger) models were pretty good machines, but a lot of them were
nothing more than I would consider as rough cut-off machines, but
unfortunately, tended to be underpowered for that function.

Answer for a RAS in general is, if the table is flat and the column
perpendicular to the table, then accuracy is entirely dependent on the
rigidity and accuracy of the arm/yoke. Unfortunately, ime, most of the
Craftsmans of the last 20-30 years don't measure up. I understand, but
have no personal experience, that recently they've made some
improvement, but that's hearsay, not as I say, from hands-on. (I have
_old_ heavy Rockwell which is different beastie altogether...)

ZZZZZZ plop! Okay, I'll bite. I infer (probably not wise of me) that
you bought/inherited this saw. This saw has issues. You say you've got
your table flat. It doesn't need to be level, just flat and coplanar
with the carriage throughout it's travel. When you measure arbor height
to table, you get a dip in the center. Well, if the table's flat...
there are probably parts available through Sears. Tom I
think.
lid wrote:
RAS table shows flat all ways by level and 'light under the straightedge' methods.
Checking by arbor height showed a small but consistent dip in the center as compared to the 45 positions.
Took the table adjustment screw to max, then added some small SS plates as shims between base and table and took it as far as I could.
Comes out the same way.
Since the gap appeared the same for the whole travel at any given angle, I thought I could live with it.
Nope. Crosscutting starts fine, then the kerf turns to dado a few inches out.
Suggestions?
Craftsman 2.75 HP saw. Original table covered with MDF. Both layers show same gap.

wrote in message
...
RAS table shows flat all ways by level and 'light under the straightedge'
methods.
Checking by arbor height showed a small but consistent dip in the center
as compared to the 45 positions.
Took the table adjustment screw to max, then added some small SS plates as
shims between base and table and took it as far as I could.
Comes out the same way.
Since the gap appeared the same for the whole travel at any given angle, I
thought I could live with it.
Nope. Crosscutting starts fine, then the kerf turns to dado a few inches
out.
Suggestions?
Craftsman 2.75 HP saw. Original table covered with MDF. Both layers show
same gap.

First, make sure your straighedge is really straight. Do both edges check
straight against both edges of another straighedge? And is the table flat
according to both edges of your straightedge?

That done, obtain the Jon Eakes radial saw book
http://www.wired-2-shop.com/joneakes/ProductDetail.asp?ProdID=3&nPrdImageID=&CatID=1
(note--I'm in no wise connected with Jon Eakes other than that I have a
paper copy of his book and recognize its utility) and do _everything_ that
it says to do with regard to tuning a Craftsman saw. I could try to tell
you the procedure but he writes better than I do and it took him a whole
book so it's really too much information for a USENET post.

After that, if you still have a problem, you should be able to figure out
what's broken or bent.

Also, check the model number against http://www.radialarmsawrecall.com/.
If your saw qualifies you get a new guard and a new table for free.
Probably won't help your problem but it's a worthwhile upgrade IMO.

You need to recheck the height of the RAS head to the table closest to the
column and again along the arm.

Although RAS saws appeared to be designed for changing angle of the arm, and
rotation of the head, many, including my 1983 Sears model, are famous for
not returning to exactly the same spot requiring re-alingment after a single
change-return. You may find yourself in constant adjustment mode if you
want to use the RAS in all positions as the apparent design would allow.

In my case I have given up and now keep the RAS for 90 deg cross cuts for
items which are too long for where my table saw is placed. I also have a
10inch sliding compound mitre saw, but only use this for angle
mitres/bevels.

Since I do not trust changing the head angle, I do my table checking with
the blade installed in the 90 deg arm to fence/90 deg blade to table
position for cross cutting.

My table has right angle brackets to hold the table to the frame on the
outside. I use these for adjusting the table front to back tilt, which I
think is what you need to adjust. I got my table to be almost parallel to
the arm. I have less than 1/64 more depth at the column than at the outer
most position. I accept this. The arm itself can flex under load, so I can
get a slight difference on a given cut.

The dip in the centre is normally adjusted via screws in the middle of the
table. If the top was replaced, check if these are missing. The table
should have a T nut in the middle, and the screw adjusts between the T nut
(installed in the bottom of the top) and an angle bracket in the frame.

I can tune my RAS to be accurate for 90 deg cross cuts, but I do re-check
from time to time. It is amazing how the tool can drift out of alignment,
or perhaps evil gremlins come in during the night and knock the tool out of
alignment just to frustrate me.

Dave Paine.

wrote in message
...
RAS table shows flat all ways by level and 'light under the straightedge'
methods.
Checking by arbor height showed a small but consistent dip in the center
as compared to the 45 positions.
Took the table adjustment screw to max, then added some small SS plates as
shims between base and table and took it as far as I could.
Comes out the same way.
Since the gap appeared the same for the whole travel at any given angle, I
thought I could live with it.
Nope. Crosscutting starts fine, then the kerf turns to dado a few inches
out.
Suggestions?
Craftsman 2.75 HP saw. Original table covered with MDF. Both layers show
same gap.