cutting 8x4 sheets

[jeffinfrance]

hi all,

as you may have guessed from my last post, i'm trying to cut up a load of 2.5m x 1.5m oak veneered blockboard.

each board (except two) has a unique cutting plan.

my process was going to be:

1 lay board lengthways with leading short edge butted up against crosscut fence and cut off rough edge.

2 set parallel fence for first width, slide clean cut edge onto parallel fence and cut.

3 repeat for second width.

4 take remaining panel by cut edge, flip over and but against crosscut fence again. set stop on crosscut fence, maneouvre piece and make final cut.

in theory, the quickest and easiest way to do it.

for the first panel, i only needed to repeat step 2, 3 times and they all came out perfect (after i paralleled my fence).

second was different widths so all steps used and not one piece is spot on. granted only a gnat's c*ck more than a millimetre, but irritating none the less. this stuff is expensive as well as big and heavy, so i'd rather get this right. can anyone see any flaws in my plan? keen to learn!

by the way, i did a test run of the tounge and grooved lipping idea on the shelves from the panel that went well. planed the lipping to exactly the thickness of the board, ran lipping and shelves through the spindle (30 minutes), hint of glue, tap with a mallet.......perfect!

some success then.

all the best,

jeff

 

[moz]

Hi Jeff,

If accuracy is important, I cut the panels a little over size to begin with. That way I'm not struggling with a load of heavy material sagging all over the place. I'm also not using the waste cut of the previous panel as a reference. Also you mentioned only three cuts and you need to cut all four to get a truly square panel, unless you're lucky. This is how I do it then.

First, cut all pieces slightly over size. Then, for each panel set the rip fence to width and the crosscut stop to length.

1. Take a dust cut using the slider on one of the longest lengths. This is the reference/face edge.
2. Cut to width with the rip fence, running the face edge along the fence.
3. Rotate the workpiece placing the face edge against the cross cut fence, retract the stop and trim one end.
4. Cartwheel the piece, keeping the face edge against the cross cut fence, place the previously cut edge against the stop and make the final cut.

Obviously, it isn't the only way but, whichever method you use, I think it's important to always reference from the same edge to avoid multiplying errors. You could, of course, use a second cross cut stop (or reposition the one supplied) and do all the cuts from the slider but, for a panel which is oblong is shape, it's more difficult to reference the short edge against the cross cut fence. I have the parallel fence for cutting long, narrow pieces such as bookcase sides which are difficult to reference from the cross cut fence and yet can be unwieldy to cut using the rip fence. But, far be it from me to suggest giving Felder any more of your money just yet

Hope that's of some use.

John

Edit: Oh, and another thing. I try to make all cuts of any particular dimension without altering the fence/stops although it can be difficult to organise sometimes.

 

[jeffinfrance]

hi john

thanks for the reply, i just really wanted to do it all long cuts in one go (which i would be able to if the rip fence and crosscut fence were perpendicular). having the table and fence in different directions makes this really hard. it would also give me larger offcuts to turn into plinths and cornices.

the machine is capable of the accuracy, no doubt.

the parallel guide to which you refer......is that the optional table extension which goes on the side of the slider with a measure rule stop that's hundreds of euros?

i'm thinking that might be a decent aquisition.

jeff

 

[moz]

Hi Jeff,

There's nothing wrong with that but I don't think the toe-out is a factor here. Even if it were, say the toe out is 0.05mm, that will be about 0.4mm over the full 2440mm, not the 1mm+ you described. I find the most difficult thing is that you're lucky if you manage a cut 90deg to the surface consistently if you have most of an 8' x 4' sheet sagging down at the far corner. Then, you don't have an accurate reference edge to start with. Other than that, for the reasons I've said, it's important to pay attention to keeping the work tight up to the rip fence rather than relying on the cross cut fence. It all depends on what it is. It's not always necessary to go the nth degree but, if something needs to be bob on, I always break the sheet down roughly even though it takes a little more time.

Yes, the additional fence is referred to, rather confusingly, as the "Rip fence for panel sliding tables" at £288.51 + VAT. I suppose I don't use it that often but it's brilliant for what you're describing though it takes a bit of setting up to start with. Goes up to 1000mm which is plenty.

John

 

[JonnyD]

I do most of my parallel cuts on the sliding table. you dont need a fancy fence. I use an old fence from a scheppach sliding table which i have drilled out so that it can be attached by the 2 t slots in the table using a couple of bolts but you could easily use a lump of 3x2 drilled out and attached to the table in the t slots. I cant remember if the felder only has one slot if so you can use a g clamp as well to stop the fence or bit of 3x2 twisting.

To set the fence just cut a piece of scrap using the saws crosscut fence and then use the scrap to set a stop block on the other fence by putting the scrap against the fence and holding it on the sawblade edge while you clamp on a stop. A l shaped block of wood that hangs over the fence is ideal for this and you can just clamp it on and it will stay parallel.

You have top be a bit carefull that you dont knock it too hard by mistake but will give excellent results for very little expenditure.

cheers

jon

 

[ondablade]

Hi Jeff. I'm struggling to make sense of what's happening in your situation.

Where is the 1mm+ error coming from? Is it a problem of loss of squareness, or that the sides are parallel, but that a dimension isn't accurate? Or inconsistency from one piece to the next?

Pardon me if i'm presuming too much, but a couple of things come to mind in addition to what John has said. (but i'm no expert - so please correct me if i'm wrong guys)

I'm presuming also that when you say 'parallel fence' you mean the rip fence - the one parallel to and to the RHS of the blade.

1. It doesn't in principle matter if the rip fence is not perfectly accurately parallel with the line of the blade - because in the end the blade will quite happily cut with some toe in or toe out - although having the rear of the blade closer to the fence does create a risk of kick back, and will result in the upcoming teeth marking the cut edge. The point though is that it's the rip fence that determines the line of the cut.

2. It's important if you want to get a straight rip that the workpiece is held accurately against the rip fence, that the fence itself is straight, and that the fence is not deflecting under pressure due to not being stiff enough. Failure to get a straight reference edge means you can't hope to get accuracy thereafter.

3. Another (and dare i say it more 'Felder') way of achieving the mythical straight rip is off the slider - using parallel bars and clamps (lots of photos on FOG, but it's expensive kit), or a toe bracket and clamp to locate your workpiece.

Presuming your slider is long enough to handle your work pieces this avoids the tricky problem of keeping the work accurately aligned with the rip fence at the start of a cut when the overlap is short - this problem gets much worse with a big panel that's hard to control. Given that the 'rip' capacity on the slider is limited you might make a rough cut first off the rip fence, and then revert to the slider to get the straight reference edge.

4. The cross cut fence must however be set accurately square to the line of travel of the slider to get a square cut/corner, but again the blade does not have to be perfectly parallel. This also presumes a straight reference edge resting against the fence, and no dust or chips to get in the way.

In this case where the fence is not perfectly square, and depending on which way up/off which edge you orient the workpiece for the second cross cut you can end up doubling the error (get two sides each diverging from 90deg by the amount of the angular error), and this especially on a larger panel can become quite a large dimensional error. (which is why i think John is saying that some cutting sequences are better than others)

5. Another possible issue relates to calibration of scales on the fences. I'm not sure how it's supposed to be done, but generally i find that a trial cut is advisable to get the setting spot on. Scales and tapes are almost by definition not zeroed to the blade (kerfs vary a bit, plus any wobble will throw the measurement out), and are hard to read very accurately. As well as bearing on a good reference surface the stop needs to be tight and slop free too, and some i've seen are not so good.

6. Finally it makes sense once you get a setting right to use it for all of the cuts needed at that dimension - apart from wasting time repeat set ups increase the risk of problems.

Again please pardon me if i'm presuming too much, but i think you're going to have to sit down and step by step work your way through what you are doing and the result each step is producing to figure out what's going down. There's no easy way out, in that while correct cutting sequences minimise problems, it's necessary to understand the 'why' of these too.

I'll be into the same space (sorting workable technique) quite soon when i get my Hammer panel saw up and running - so bear in mind that what i have to say about technique on sliders is based on what i've read.

On the other hand it's predecessor was a Robland combo with a very cranky sliding table and fence set ups - which more than once forced me to sit down with a pencil and paper and paper cut outs to figure out what was going on dimensionally, and then to figure out a cutting sequence to get around the issues...

 

[jeffinfrance]

ian,

yes, the rip fence.

i'm obviously going to have to have a rethink on this. i can understand why the rip fence toes out, and possibly also why the slider does the same, but with a decent, sharp and good condition blade, is it necessary? it would seem from the poll, most people here would prefer a parallel fence.

using the rip fence for long accurate cuts on roughly sized blanks i have found almost impossible. there is very little table infront of the blade and the technique relies on a stable infeed. using the slider to move the piece into the blade is much more stable (and the whole point of having a slider i thought) but with the slider and fence going in different directions, with a heavy panel, also not easy.

with the slider mounted support and parallel stop, this would make final cuts perfect using just the slider. only really usefull if it is a quick set up which it should be once the measure scales are properly referenced from the blade. it also gives the panel support to prevent sagging. however, this method can only be used for final sizing as it only has a width capacity of 1000mm, so useless for the first cuts.

the blanks i cut yesterday can all be recut using this method and slightly larger lippings can be put on. but i still say, that with a slider, fence and blade all running parallel, my initial method (with the addition of doing the last cut with the new parallel stop) would more than halve the time taken. am i really risking kick back using a slider with a heavy engineered panel on it. i'm not getting any tear out on the offcut side either, which is in contact with the teeth on the upstroke, so it would seem that the one benefit of having toe out cutting doesn't make any difference anyway.

i didn't really want to do a crosscut first to get a decent leading edge as i want to max the height of the units. once i had my first longitudinal cut i was goint to use that as the reference for the rip fence. but using the slider for the motion it carries the piece away slightly. also with the fence pointing away from the blade, the front edge of the fence is nearer to the the cut line of the blade. what i could try is clamping a stop to the table in line with the leading edge of the blade but i'll need a better leading edge to butt against the crosscut fence. this should be ok for the first two cuts as i'll have lots of egde to reference. the final cut i'll have to wait for the parallel stop.

none of this would be necessary if everything was parallel! when i cut the first panel, i paralleled the fence to the slider. this makes me think that the slider might not actually toe out, because i got perfect cuts. well within what i can measure anyway.

the engineer has to come to adjust the machine table anyway so i'll find out how parallel he can get it.

meanwhile, i might just have to cut oversize blanks and wait for my parallel support/stop/fence for the slider. my lamello hasn't arrived yet so i cant put them together.

if i'm having to cut blanks and then size cut, i'm not really getting much advantage having a huge panel saw.

jeff

 

[ondablade]

I'm a little lost at this stage Jeff - no doubt partly due to my own inexperience.

Toe in or toe out seem unlikely to be the cause of your problem, because as above its the line of the slider, mitre slot or the rip fence that determine the line of your cut.

Once the saw is accurately aligned (whether parallel or toed) getting a good result has got to be about technique - about getting a good straight datum side, and cutting off it in the correct order using set ups that utilise known accurate alignments so that errors are not created or multiplied.

I haven't figured it out, but this may introduce additional steps. Rough and final sizing will slow things down too, but i kind of doubt that there's many panel saws that can be set up square all round to give more or less perfect results without consideration being given to technique and working around what it's good at, and what not.

There's for example not very many saws about that you can lean heavily on the stock rip fence without causing quite a significant amount of deflection. Which straight away suggests that a light and consistent force to hold the work against it is advisable. (which probably explains quite a lot of our US friend's liking for heavy duty full length (heaven forbid!!) Biesemeyer rip fences)

Put it this way - a very technically aware friend in the high end kitchen business for many years cites one of the primary advantages of a big 3 axis CNC machine he bought (over a large, heavy duty and almost new Italian panel saw) as being it's ability to consistently cut square cabinet sides. Which saves so much time in assembly.

I'm no expert on the toe in/out topic, in that i've never had the kit or the need at my level of working so far to measure it properly - although i've had poor cutting as a result of the trailing edge of the blade dragging and causing splintering after mis-setting the sliding table on the Robland. I fixed it by trial and error realignment of the table though.

I guess that a parallel set up is probably OK as long as your blade is running true, although i suspect that for most toe in/out is a bit of an academic topic - i doubt that many have the kit to measure the several thou normally involved accurately.

One aspect of toe-in/out (which is which? ) i've not yet fully figured out is that if you set it relative to the sliding table/mitre slot for cutting off the LHS of the blade, then the only way i can see to get the same on the RHS/rip side when cutting off the rip fence is to offset the rip fence by the 0.030in or so needed.

This will leave the rip fence slightly out of square with the cross cutting fence/slider direction, and slightly differently aligned to the slider/mitre slot - which probably isn't a problem??? (anybody?)

The other issue that follows from toe in/out is that ideally you won't want to arrange cuts so that the workpiece comes off the 'wrong' side of the blade.

Anyway - have fun with the service guy. It seems to me that to get good results off a saw we each have to work to become more expert than the average service guy...

 

[jeffinfrance]

well,

i appear to have been proved right (although i'm sure that's still up for debate)

set my fence parallel to the slider (as precisely as i can using thicknesses of paper).....lo and behold, accurate first cuts without cutting oversize first.

5 minutes a panel might have been a bit ambitious as i have to lower the saw blade to maneuover each panel into place. bit more practice and i could deffo get down to 10 minutes for all components per panel cut accurately (within 2 tenths of a millimetre anyway)

might not be 3 axis cnc quick, but i'll be well happy with that.

teeth on the upcut were touching the cut piece to the right of the blade (offcut side when cutting with the slider), but not enough to cause any problems or tear out. as i want to produce finished cuts both sides of the blade this might need looking at, but to be honest i don't think felder are capable of installing the minute amount of toe out that some people here think should be necessary. i'm still awaiting the definitive answer from felder on that, but i think the saw is perpendicular to the slider as is.

my fence originally toed out from the blade. i can understand the reasons for this....kickback and upcut damage. but keeping a large panel butted gently against the rip fence and having to force the panel over the machine is a large price to pay and a clean perfect cut is very difficult to achieve. using the slider to project the panel past the blade with the minimun of effort leaves butting the datum edge gently against the fence a relatively easy task.

all i need now is the parallel fence stop teble extension, my machine tables re-aligned and i'll have a panel saw i'm actually happy with.

i still think i should have had these when the machine arrived, but hey, i dont think small children in far away lands should be making our trainers either. i must be a wierdo!

all the best,

jeff

 

[moz]

Jeff,

It strikes me that using factory edges is one big source of potential inaccuracies. You can't even rely on them being straight never mind square. I understand that you want to retain maximum length but the only way of getting a true 8' is to buy bigger boards. As usual, time, cost, quality - pick any two from three. With care I can produce panels which align with finger touch accuracy. I suppose I'm a fairly experienced machinist but I don't think I could do better than you by just sawing an 8 x 4 from the off.

John

 

[jeffinfrance]

i did think about that, but apart from being a bit rough, they are straight enough to use up against the crosscut fence to get my first edge. i can use this for the rip fence then and with the saggy corner supported on my new (when it arrives) parallel/stop/support thingy, not too shabby on the 90 degrees either.

we'll see how it goes with some repetition. its about taking time and care i guess.

glad i seem to have got past this headsratch though. my next will be the arched window i'm about to continue.

jeff

 

[ondablade]

Good going Jeff. Technique is an important part of the job ( and so is care), even on a slider. This thread here on Woodweb (a largely professional forum) discusses saw options for cutting square plywood panels: http://www.woodweb.com/knowledge_base/V ... w_and.html

The second last post at the bottom about operator skills and training rather captures the point that even with a very high quality and well set up slider like an Altendorf there's a lot of care and technique involved.

It's not going to eliminate the need for right technique, but here's an interesting aftermarket digital rip and cross cut fence with computer positioning. http://www.tigerstop.com/RIP.html It seems well regarded judging by forum comments, and available in Europe....

 

[jeffinfrance]

if i was cutting panels all day every day, i'd have that fence, but i'm quite happy with my new "parallel" one. seems i'm getting better results than everyone there!

must admit, with my new alignment, not as difficult as they make out. sure, you have to take care, but don't you always on the final tool cut.

jeff