Combating bandsaw drift

[andersonec]

Oh for a bandsaw with a table like this one to combat drift..
http://www.youtube.com/watch?v=IAFMEg0H ... re=related

 

[condeesteso]

watched the first half - a rotating table, wow. Strikes me as gross over-engineering to resolve a simple issue, but if the fence track is fixed (non-adjustable) then I suppose so.
Normal practice I think to do the freehand rip (as in video) then true the fence to the ripped stock. On my Record BS500 this is very easy as the mount bar is held by 2 bolts and you can loosen one to adjust the bar angle to table. Not so easy on some with extruded ali tracks for the fence, but worth modifying as necessary to enable fence angle adjustment.
One golden and certain rule I think is that fence set square to table will never be right and you will always get drift. So one way or another being able to adjust fence angle a few degrees is vital. Another way (hit n miss though) is to adjust the blade on the wheels, generally running it more forward as I cannot recall ever having to adjust the fence clockwise (into blade) - it always needs to be a little away from the blade in my experience.
Another thing to note is the centre of the blade for setting on the wheels is the centre minus the teeth, so a 1" blade may have its tension centreline about 3/8ths from the back.
Basically I reckon bandsaw designers need to do 2 things as a matter of course - make the fence adjustable as above, and always make the guides tool-free adjustment. My BS needs 2 different allen keys... ridiculous!

Sorry if any of this is eggs for grannies! It looks like a nice bandsaw though.

 

[RogerP]

It's easy enough to account for blade drift during rip cuts by adjusting the fence but the big advantage of a rotating table is that the sliding crosscut square will also be accurate.

 

[andersonec]

watched the first half - a rotating table, wow. Strikes me as gross over-engineering to resolve a simple issue, but if the fence track is fixed (non-adjustable) then I suppose so.
I cannot recall ever having to adjust the fence clockwise (into blade) - it always needs to be a little away from the blade in my experience.

I think if the engineers can come up with answer like this to solve an ever recurring problem then it must be worth doing. Mine also has never had to have been adjusted clockwise.
If the table were adjustable then this would keep the entire table / fence / runner square with the line of cut which would increase the usefulness of the machine tenfold.

 

[Steve Maskery]

One golden and certain rule I think is that fence set square to table will never be right and you will always get drift. .

Absolutely not! It is quite possible to have the fence square to the table AND be drift-free. The whole point of eliminating drift is to achieve this situation. Compensating for drift, which is what you do if you simply twist the fence to match the drift, is always a second-best solution, even though it is widespread practice and even recommended by some writers.
It is MUCH better to eliminate drift and still cut straight and true and it is not actually difficult to achieve this. I have yet to find a bandsaw that I've not been able to fettle like this, although I do accept that others claim that theirs is not de-driftable.
S

 

[woodaxed]

I dont have a problem with drift, that machine looks a complicated way of correcting something simple and the more complicated a machine is the more to go wrong with it

 

[tool613]

On the My 1950's Wadkin 30" DR band saw that i rebuilt the table can be adjusted for drift b/c the the trunnon is not cast into the C frame like most saws but bolted down to a milled surface. There is play in the bolts to rotate the table. Really a real good bit of engineering

You will note the the fence uses the table as an index and that it tilts too with a sliding rip plate but there is no adjustment in the fence for drift.
An old idea?

jack
EM

 

[andersonec]

It's easy enough to account for blade drift during rip cuts by adjusting the fence but the big advantage of a rotating table is that the sliding crosscut square will also be accurate.

Absolutely!!! saves all that messing about every time you change a blade, if it takes you seconds to align the table (and not the fence) then it's got to be a huge benefit, I agree with what Steve says and after spending many minutes fettling tension and running placement have managed to get the cut parallel to the table but it has taken some time and some teeth gnashing.

I dont have a problem with drift, that machine looks a complicated way of correcting something simple and the more complicated a machine is the more to go wrong with it

I see nothing more complicated on a table that turns than a table which tilts and they nearly all do that.

Andy

 

[condeesteso]

Excellent then if it is possible to easily rotate the table / fence assembly a little - Jack's Wadkin.
I accept the point re mitre fence... forgot as I never ever use mine.
Regarding ripping, I just want to check some basics: If the back edge of the blade is not central in the kerf , drift will result, especially if the blade is touching one side even slightly.
Bandsaw tables tend to be square but could just as easily be round, hexagonal etc. Just like router tables that are always rectangular, but fence orientation is totally irrelevant on them.
To eliminate drift, the line through the blade cross-section needs to be dead parallel to the fence.
That line depends on a number of things and is hard to control accurately (I imagine things like where exactly the blade rides the wheels, how the blade stretches under tension (not evenly across), guide types and their set-up maybe... and perhaps other factors).
So as I understand it, if I really want the rip fence square to the table (I see no reason why unless you use a mitre fence regularly), I must get the blade section exactly at 90 degrees to the fence.
I do find it easier to get the fence square to the blade and ignore the table.
Apart from the mitre fence issue (and I find them of little use), what am I missing?

 

[Mr Ed]

One golden and certain rule I think is that fence set square to table will never be right and you will always get drift. .

Absolutely not! It is quite possible to have the fence square to the table AND be drift-free. The whole point of eliminating drift is to achieve this situation. Compensating for drift, which is what you do if you simply twist the fence to match the drift, is always a second-best solution, even though it is widespread practice and even recommended by some writers.
It is MUCH better to eliminate drift and still cut straight and true and it is not actually difficult to achieve this. I have yet to find a bandsaw that I've not been able to fettle like this, although I do accept that others claim that theirs is not de-driftable.
S

I agree Steve. I can honestly say I have no drift on my current saw (Axminster AP5300) with table, blade and fence all square to each other. I used to suffer massive drift on my old Scheppach Basato 3. I think the stiffness of the frame of the saw and it's ability to fully tension a blade is a key determining factor.

Ed

 

[Steve Maskery]

Ah yes, indeed. If the machine won't tension the blade then you'll never get it right.
S

 

[condeesteso]

I have a bit of an engineering fascination for these things, and I'm not getting answers or explanations.
I checked yesterday, my fence is currently set 0.6 degree anti-clock (best I can measure, but well under a degree).
I reckon there are too many variables to simply say 'I have no drift' - really? From one blade to the next, never?
And re tensioning the blade - what is the right tension and how is it measured (given the devices on machines can't measure it accurately and many older ones have no tension indicator anyway)? I've heard many times that the machine must be able to tension the blade correctly - BUT if the machine can tension the blade, the blade has to be tensioning the machine, and tension means movement. Steel under tension that does not move at all will fail (fact)... it may be very small but the frame has to move.
I'm not asking how to fettle drift out Steve (I'd need to buy the DVD for that) but anyone care to explain why ignoring the table orientation and setting fence parallel to blade (I mean a degree max say) is wrong?

p.s. the wheels have hard rubber tyres - what happens to the profile of those when you correctly tension the blade?

 

[RogerBoyle]

IMHO
There is no right or wrong way as long as you are getting the results that you are happy with
What works for one will not work for another etc etc.

this is where i stand.

It used to drive me up the wall that i would get blade drift I tried everything to eliminate so that i could keep everything square to each other
With my Saw at the time ( an Elu ) i did get it so that everything was square and relative to each other It used to take me 2 or 3 hours to do so but as soon as i changed a blade I had to redo it again..and again ...and again
Depending on what i do I can change blades as much as 2 or 3 times in a week or may go weeks with the same blade
(Perhaps the accuracy in the blade lengths wasn't as good then as it is now LOL)

I then tried just adjusting the fence to eliminate it took me less than 5 min so for me its a no brainer
All my jigs get referenced off the fence and as long as that is correct then the jigs will be

I have carried on this vein with the Scheppach but I may just try it once to see if its still as much hassle as it used to be
(Some how i think it will be LOL)

If you do a search on U tube every video out there only shows you adjusting the fence not one of them says this is how to totally eliminate it
Roger

 

[andersonec]

) but anyone care to explain why ignoring the table orientation and setting fence parallel to blade (I mean a degree max say) is wrong?

I'm not saying it's wrong, as we all, well most of us, set the fence square to the cut because the table cannot be rotated. If you had drift, which most of us do, and the table could be rotated so that it was square to the cut you could now use your mitre gauge for all those mitre cuts / square cuts which were not possible with the table out of square. I guess the reason you don't use your mitre gauge is for that simple fact, it is not square to the line of cut.

Andy