Forum home made wall saw

[jfc]

I am sure we can do this between us . The frame work is easy , its just some timber to support the board with a level base . the saw is just a normal circular saw and the guide can either be a guide rail or just a straight bit of timber to run the saw along . Myself i'm thinking of using the clamp and guide because you can lock it on from one end and ... well i already own one .
The problem as far as i can see is making the guide rail what ever it is run level so you dont need to keep checking both ends every time you move it .
If we actually start to get somewhere with this idea then i will build it to see if it works .

[wizer]

I think a track guided system would be better. I see your point about the one point locking, but with a clamp/guide you'd have to allow for the saw offset for each cut. Not very convenient. Tho, having used the EZ System a bit now, it's a shame they done do a system like this that is 'dovetailed' onto the track. Hence preventing it from coming off the track. You have thos shopnote article about this, don't you?

[engineerone]

since i guess you need it to be about 9 feet long, the problem is
ensuring that the rise and fall can stay level over that time, and is synchronised.

seems to me the logic is to use some form of hydraulics tomove the guide up and down, much easier to make work together i think.

however, over time, even hydraulics loss pressure, so you need to have a kind of positive locking device, which i would suggest is dowel pins through holes. to get the small incremental movement you could use a snail cam which rotated. that way you only need holes at say 1 inch centres. you would also need to have a level guide on the cross rail to ensure that it stays accurate.

i guess you could start out by using a pair of acrows, and mark the adjustment nut to ensure that you turn each the same amount. would not be quick, but give you an idea of how to develop it.

all the american one seem to have large balance weights, and i am not sure you need them if you use hydraulics.

paul

[andy]

engineerone said:

ensuring that the rise and fall can stay level over that time, and is synchronised......

seems to me the logic is to use some form of hydraulics tomove the guide up and down, much easier to make work together i think........

paul

servo motors and encoders

[jfc]

Paul , you have just added £1000 to the build yes the wall saws you can buy use this but they nicked the idea from a simple wall saw somewhere .

[andy]

jfc said:

Paul , you have just added £1000 to the build yes the wall saws you can buy use this but they nicked the idea from a simple wall saw somewhere .

And now you introduce a budget constraint

Just when we was getting inventive ;D

So a Fanuc control is out of the question

[mailee]

I have the plans for one of these but it is only in DXF format. Looks impressive although I have never decided to build it. Not enough space in my shop for one.

[jfc]

Not enough space If you have enough space to cut an 8x4 sheet you have enough for 3-4 wall saws

[mailee]

Yes but you need a wall for it and I haven't an empty one!

[sainty]

Often thought about this problem myself. When we were at school in CDT (Technical Drawing) the drawing boards had a parallel guide that run on a wire and pulley system. I'm not sure if this could be scaled up or not?

This is the sort of thing that I mean

www.drawingboards.uk.com/priory_drawing_boards.htm

I dont know if it would lose accuracy on that scale? Speaking of accuracy, what sort of tolerance would you be expecting?

[engineerone]

jason you're as usual not looking at things logically.

second hand stuff will not add that much money ;D

sadly don't think the drawing board analogy would work cause of the weight of the cables i think.

as a really cheap skate way i initially thought about a couple of motor cycle sprockets and some chain. but then you need a long shaft which adds more money.

go back to the old days and nick the workings of a steam powered factory with an overshaft ;D

paul

[9fingers]

How about something like this?
homepage.ntlworld.com/bob.minchin/Visio-panel%20saw1.pdf

I've done a very quick sketch but the idea is that a square base is fitted to your circular saw which can fit into the carriage in two positions one with the blade vertical and able to slide vertically and the other with the blade horizontal and fixed in the carriage.
The carriage has wheels to run between two guide rails, two at the bottom spread as wide as practical for stability and a single one at the top to guide the top of the carriage. For vertical cuts the carriage is locked somehow and for horzontal cuts it moves from the parked position on the right. Hope this is clear.

The guide rails are possibly best made from angle iron with grooved wheels running on the edge of the angle. Steel angle is readily available. The carriage could be wood but I'd possibly use metal. However I'm biassed as I work in metal as well as wood and I guess we ought to be looking at a design that is nearly all wood for widest appeal to members.

Comments welcome

Bob

[Scrit]

sainty said:

Often thought about this problem myself. When we were at school in CDT (Technical Drawing) the drawing boards had a parallel guide that run on a wire and pulley system. I'm not sure if this could be scaled up or not?

It can indeed. I think the device would need to comprise 3 main parts, namely an angled chassis with a base board to hold the material at 5 to 10 degrees off vertical, a rip slide/carraige mechanism and a crosscut slide/carraige mechanism on which the saw runs. In other words something not dissimilar to the drawing done by 9fingers

I'd say the main issue isn't the framework/chassis it is the method of supporting the saw, i.e. saw carraige rigidity, although the strength/accuracy of the linear motion components required is also an issue IMHO. That biases me towards using metal strut to support the saw carraige and low-cost linear bearings designed to run in that strut, such as those sold by Hepco for use in Unistrut (electrician's steel channel section as used to support industrial cable tray):



The counterweight is a good idea as it really is a necessity to have some way to reduce the effort required to return the saw to it's home position (at the top because cross cuts always need to be made with a downwards stroke). To be truly functional, though, a wall saw does need to be capable of both cross cut and rips which is a much thornier problem.

sainty said:

Speaking of accuracy, what sort of tolerance would you be expecting?

Even industrial-quality machines aren't much more accurate that +/- 0.5mm repeatability and deviation of +/- 1.0mm over 3 metre length of cut. Even there the biggest problem appears to be the rigidity of the chassis or framework beneath the gantry

engineerone said:

Sadly don't think the drawing board analogy would work cause of the weight of the cables i think.

Well there was a lightweight CNC router called the ShopBot a few years back which used the same sort of cable drive to haul a gantry around and that did work. The breaking strain was something incredible for the size of wire, like 120kg, and we'd probably need a 20kg weight to overcome the weight of the saw and friction, so I think I'd disagree with you there, Paul. (ShopBots usd to be able to generate up to about 15kgf of sideways force cutting timber - a full-blown CNC router generates 50kgf)

I think that Bob (9fingers) pretty much has it

Scrit

[mickster]

In reference to Bob's idea, the angle is a winner on cost, but the grooved bearings are silly money. A simple linear bearing is easily constructed from skate bearings & M8 nuts/bolts.

I have built a CNC router along the lines of that found here:

buildyourcnc.com/default.aspx

Patrick shows the construction of the machine step by step in his videos, the linear bearings being constructed in video #1.

The angle is easily mounted in two parallel grooves & the linear bearing can be pre-loaded from the rear using a nut/bolt adjusting mechanism. Provided the linear bearing is long enough, there should be very minimal 'racking' between the upper and lower rails.

HTH,

Mick.

[Scrit]

The original ShopBot used the bearings from sliding wardrobe doors. The problem is that they have poor wear characteristics and generate rather a lot of run-out, hence the suggestion about using the Hepco sets. My understanding was that a bearing set like those was around £30 with six sets in total being required

Scrit

[mickster]

scrit said:

My understanding was that a bearing set like those was around £30 with six sets in total being required

The Hepco bearings, or the skate bearings?

My skate bearings came from eBay, from a user called 'cinspider'. I got 80 for under £40...enough to build another CNC machine. Each linear bearing uses 4 skate bearings, 2 linear bearings per axis. A wall saw would only need enough for the 2 axes, horizontal & vertical, so 16 bearings...

sports.search.ebay.co.uk/_Inline-Roller-Skating_W0QQsacatZ16258QQsassZcinspider

£7.99 for 16 bearings.

[Scrit]

Linears - they are dust-proof and guaranteed to last for hundreds of thousands of operation cycles. I'm not sure that cheapening it would be worth while on the grounds of reliability. But then I'm in trade so for me reliability is very important

Scrit

[mickster]

Ah, sorry for the confusion, I realise now that you are referring to linear slide bearings.

The skate bearings themselves are also sealed, but the problem still exists that dust/debris can accumulate on the angle...having said that, there is nothing stopping one from placing a 'wiper' at each end, to clean the angle on each pass.

[Scrit]

You'd still need a wiper for the ShopBot-style angle iron and linear bearing because it is all but impossible to design a wall saw where dust doesn't get into the guides. Another possibility t be to go to self-cleaning UHMW-PE bearings such as the Pacific bearings (also sold by Hepco), but if the linear rails need to be load bearing then they'd need to be some size. Alternatively I'm starting to wonmder about something like scaffolding tube with roller bearings, a bit like a SawTrax:





Scrit

[9fingers]

mickster said:

Ah, sorry for the confusion, I realise now that you are referring to linear slide bearings.

The skate bearings themselves are also sealed, but the problem still exists that dust/debris can accumulate on the angle...having said that, there is nothing stopping one from placing a 'wiper' at each end, to clean the angle on each pass.

Hi Mickster, In my 'design' I've opted for wheels with V grooves running on the edge of steel angle. Although not good engineering design from a wear point of view, my aim was to create a self wiping effect to stop build up and also to allow the bottom of the vee to tolerate some sawdust before throwing the carriage out of line.

Interesting source of bearings though - could be used in the centres of the wheels in pairs

I'm working on the assumption of fairly light use even by the trade guys and possibly use for rough cutting of full sheets prior to dimensioning on the table saw. If the trade guys are cutting that much sheet material and need final dimension cutting then they possibly should be buying a commercial panel saw anyway.

Bob

[mickster]

scrit said:

Alternatively I'm starting to wonder about something like scaffolding tube with roller bearings, a bit like a SawTrax

Indeed, some of the DIY CNC guys over at cnczone.com are using the scaffolding tube idea, but on a smaller scale. This has the advantage of being able to space the bearings at 120 degrees and use a supported rail.

9fingers said:

Interesting source of bearings though - could be used in the centres of the wheels in pairs

Or, if I'm visualising your setup correctly, you could use two bearings with a spacer inbetween, to effectively create a v-roller, albeit with a very small contact area. Would be subject to high wear rate too, though.

[9fingers]

9fingers said:

Interesting source of bearings though - could be used in the centres of the wheels in pairs

Or, if I'm visualising your setup correctly, you could use two bearings with a spacer inbetween, to effectively create a v-roller, albeit with a very small contact area. Would be subject to high wear rate too, though.[/quote]

Yes i see what you mean, The bearing surface would be largely the as cast rounded over edge of each bearing which would abrade the steel track so I agree - poor wear indeed. It would also put side load on the bearing which they are not designed for.

Bob

Bob

[jimw]

If you go here:

www.shopnotesspecials.com/

tools & jigs

Click on "already own book"

Type JIG239 into code


You should get the article cut list etc etc

[9fingers]

Hi JimW

Thanks for the link to the further details of the shopnotes design. I have the whole issue containing the design as a PDF and I 'loaned' this to Jason after his first post on the topic. We concluded that something a bit simpler would do the job for most people and should be cheaper to make. I have pinched the basic ideas from the design and tried to make it as simple as possible. Since posting a link to a sketch I've made, I've not heard from Jason as to whether he wants to have a shot at making it as per his original posting.

If you follow the link in my earlier posting and browse around the directory you might just happen upon the pdf file of issue 88 with the original design.

Bob

[jfc]

My thoughts where to use the clamp and guide as i have the rail system for it . This fixes the saw to the rail and lets it run . I was also thinking of just using it as a horizontal cut as this could also do all the vertical cuts aswell .