I just can't seem to square up sheet goods

UKworkshop.co.uk

Help Support UKworkshop.co.uk:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Andre

New member
Joined
8 Jul 2020
Messages
3
Reaction score
2
Location
Guildford
Hi,
As you can guess I am new to woodworking :)

I am trying to make a square cut jig as shown by Peter Millard using 6mm MDF as the base. The issue I am having is not even related to this alone, many times when I try and cut sheet goods things seem to always be off by anything from 0.5 - 1.5mm.

After watching more videos from people like Gosforth Handyman, I have now learned that you can't always trust the factory edges.

I am using an Erbauer track saw (the screwfix one) and the Evolution tracks (ST2800).

This is what happened to me recently and I just can't figure out where I am going wrong on this.

I take a straight edge (Aluminium level) and look for the straightest (no gaps) edge on the MDF. Then I took my speed square (which I have done the check for squareness with a pencil) as well as an engineer's square and line up my track (which I also checked for straightness) to cut the left hand side off. Then I check the corner between the reference edge and this new cut. The engineer and speed square has no gaps and is flush.

Next I repeat this process on the right hand side edge, thus still using the first reference edge I then line up the track with speed square and double check with the engineer square. Forgot to mention, I clamp down the track and check probably 10 times. I.e. last check I make is just before I turn the saw on (so I know I didn't knock it while moving the saw around etc.). I make this cut.
Check this new corner and reference edge and yup it looks square, no gaps and is flush.

So in theory by now I have a straight bottom edge and 90degree corner on the bottom left and bottom right edges of my board.

Next I go to cut the top edge. Line up the track so it is square with the left hand side edge as well as the right hand side edge. Tripple check, make the cut.

And now the issue is. I check the corner between the left hand edge and the new top edge, 90degrees. I check the corner of the right hand edge and the new top edge, it looks like it is an obtuse angle.
I checked all the other corners, they look square apart from this one.

Unfortunately I don't have an MFT / bench dogs etc. that I can use.

Any thoughts or tutorials I can follow?

Thank you very much.
 
Don't use the factory edge as a reference, cut a few mill off it first so it is straight.

Now cut the opposite edge parallel to this by measuring.

Cut one end square then mark the other end, check the diagonal measurements (they should be the same) before you make the final cut.
 
What Doug said above is your solution. Only thing I’d add is how sharp the saw is and how you hold and cut needs to be the same each time, pushing on the saw from a sideways direction will throw it off.

If you make the second cut referenced off the previous ones you are likely to get a cumulative error. If you are 0.5 degrees out on each one the second cut will be 1.0 degrees out, the third 1.5 degrees and so on.
How big are the boards? 1.0mm over say 1000 mm maybe acceptable (width of a pencil line!) depending on what you are building, but you wouldn’t want it on a 200mm wide fine box lid.

Stick with it you’ll improve.
 
How much longer are these cuts than you speed square?

You might have checked your speed square for squareness, but if the cut is 3 times the length of the speedsquare, then any potential error you missed when checking it for squareness is multiplied, and then accumilated over several cuts.
 
I tend to buy high quality baltic birch boards from a reputable supplier and have had very few issues. Even so, I habitually measure the diagonals on every new board. This tells me straight away if I am dealing with a parallelogram with unequal angles or not. I do that first. If it is off I know I need to square the board up, or at least the part of it I want to use.

I do not find long levels especially good as a straight edge. They almost always have a radius on the corners so it is hard to run a pencil along cleanly. It is better to use a long builders square if you are struggling. These can be pretty cheap. Tape measures are tricky as people muddle the flexible zero point system where the tab at the end of the tape moves (it is supposed to).

A frequent cause of people making errors with squaring up boards etc, is bad measuring, often using a mix of tapes and rules. For accurate marking out on boards I religiously use high quality steel rules: I have three at different lengths, all by the same maker and all checked that 300mm and 600mm is the same on rules compared. Good steel rules are pretty cheap.

I do not obsess with clamping the track down when sawing. I use a Mafell track saw and have never used yours, but if I place a track on a board (I work on a big table the size of a big sheet) then it stays there fine without clamps. But - as said above, it is essential to let the saw do the work, not push it hard in any direction.

The other thing that I find makes a big difference to accurate clean cutting, is to use extraction on the saw rather than the dust bags they come with. When blades or teeth start to clog, they wander.
 
Once you are confident you have one straight edge I suggest you use a knife or a 0.5mm pencil for your marks.
Don't trust an 8" ? 10" ? speedsquare to accurately project a right angle across a 4 feet wide board.
Use the classic 3-4-5 triangle, all measured with your best tape measure pulled tight and counted forwards from the 1", 100mm, 12" or whatever mark, not from the unreliable witness that is the hook.
3' across, 4' along and 5' hypotenuse makes a right angle triangle big enough to be trustworthy.
Cut one out of something stable and keep it for future use if you plan to do a lot of this...
 
Is there any flex in the sheet when you cut?

I had issues with accuracy due to not having material completely flat.
 
I haven't worked out, and I don't think anyone else has yet, whether the OP's problem is a setting out problem, or one of inaccurate/ clumsy tool use (or poor tool set up).

It's not often I use sheet materials, and when I do, it's not often that they need to be accurate (they're generally contained in a frame). However, when they do need to be accurate I do two things which no-one has mentioned. Firstly, I only measure once, to set up a gauge. Either my combination square if the distance is less than a foot, or a batten and block for something longer. This removes one possible error, and means that your straightedge is certain to have been correctly located parallel to the existing edge. Secondly, I saw it by hand. For breaking down sheet materials or sawn boards roughly, a clamped straight edge and circular saw is great. But if you want something absolutely bob-on, then saw it by hand. An ordinary hardpoint crosscut handsaw is fine.
 
I think all the OPs answers are probably in this thread already, but I’ll just add that the crosscut jig I made relies on you to set the rail up square; I wouldn’t trust a speed square for that, personally, but the OP says they’ve checked it for square, so the error has to be coming from elsewhere.

When cutting off the jig, are you clamping the rails down or relying on the grippy strips underneath? I’ve had no issues with Festool and Mafell rails, but all others I’ve used have varying degrees of grip and it doesn’t take much for the rail to move a fraction, especially on a slightly dusty board.

The Evolution rails are also a little odd in that both low-friction ‘glide strips’ are between the rib and the splinterguard, so it’s possible to apply too much pressure and tip the saw over slightly when cutting.

I haven’t used the Erbauer saw, but I’m assuming you’ve snugged the saw onto the Evo rails to take out any slop? And they won’t necessarily be the identical to the Erbauer rails, so you’ll need to check this when you swap between them.

So to recap; eliminate any possible errors with the tool - snug it to the rail, keep it flat on the rail when cutting, make sure the rail isn’t wandering, make sure it’s cutting straight, make sure the material is fully supported - and then it’s simply down to measuring carefully, and good technique.

And as others have said further up the thread, it would be interesting to know what size pieces you’re working on; 1.5mm out over the height of a cabinet door would look horrible, 0.5mm out over a wardrobe door is something that can be tweaked.

Any engineer will tell you that there’s no such thing as square, just varying degrees of out of square; I refer to things as ‘square enough’ for a reason.

HTH P

PS - FWIW I’ve never had an issue with factory edges on boards either, but always worth checking.
 
I would argue that two parallel guides (where one is set from the other) is more accurate than using one that you keep moving back and forth (which just pivots the track) and takes several attempts to get right.
 
As said, lots of great advice.

A thought from me: if it's a combination of marking out, and accurate cutting which is causing the problem, then try doing all the marking out first. I suggest that trying one cut, then measuring 90 degrees, then cutting etc. will just cumulatively increase any error you may have. Mark it first, then cut the 4 lines.

E.g. on an 8x4 board, you should be able to mark out in pencil a full 1200mm square (don't use the factory edge), and KNOW it is good, by measuring the diagonals and using Pythagoras. Once you have good quality marks, then it's a question of whether the saw will cut all 4 lines accurately. Clearly, it's the corners that matter- you don't need to draw a perfect line between them.

Cutting wood square can be so much more work than Youtube suggests. For smaller non-sheet wood, a shooting board will be helpful.
 
transatlantic":1qqzxi62 said:
I would argue that two parallel guides (where one is set from the other) is more accurate than using one that you keep moving back and forth (which just pivots the track) and takes several attempts to get right.

I don't think it can be more accurate - either it is parallel or it isn't - but certainly it is probably easier and I've been meaning to make a slave gauge for some time but never got round to it. Just make sure that only one of them has a scale on it.
 
Steve Maskery":3u4tk7db said:
transatlantic":3u4tk7db said:
I would argue that two parallel guides (where one is set from the other) is more accurate than using one that you keep moving back and forth (which just pivots the track) and takes several attempts to get right.

I don't think it can be more accurate - either it is parallel or it isn't - but certainly it is probably easier and I've been meaning to make a slave gauge for some time but never got round to it. Just make sure that only one of them has a scale on it.

What I mean is, as is shown in your video, each time you adjust the track at one end, you pivot/rotate it, and the other end goes slightly out. you're chasing your tail so to speak. Where as if you had two guides, it would be more reliable as you can pull the track to butt it up against two points. No back and forth.

Of course, it does require that the guides can be clamped, else you need 4 hands.
 
Sorry for the late reply. Thank you very much to everyone here. The information in here has been immensely valuable to me and I have learned a lot. For me it came down to a number of issues including mostly user error (a bit of flex in the material, the rails not being clamped etc. etc.).

In the end I decided to buy a MFT replacement top (a company here in the uk that sells jigs and tops) and a couple of bench dogs (using Peter Millard's code) etc. This alone at least gives me a guaranteed way of measuring out straight lines and 90 degrees.

I can now also see what Peter was referring to in his videos when he says that not all MDF are equal. The 18mm Medite that the MFT top is made out is 1000x better in quality than the 18mm I bought from Wickes.
 
Back
Top