On 19/08/2020 14:38, T i m wrote:
On Wed, 19 Aug 2020 13:56:47 +0100, John Rumm
wrote:

snip

Eh? It's *the* tool for that job, if not injection molding, especially
if you have to also make / setup jigs etc?

I was referring to the speed aspect - and for the level of accuracy
required. If you can tolerate a couple of thou, then you can sand to that.

Sure, but I still question the concentricity (within the general
tolerance etc). If you are rotating something by hand the rate at
which (and to what depth) can vary as a function of how fast you
present the work to the cutter. So, even if you rotate it several
times there may be bits that may have been 'overcut' as you first
presented them that are then lower than the surroundings.

I can't quite visualise how you would do that - if the pinion distance
from the "cutter" is fixed and rigidly supported, then you can't cut any
closer than that distance - same as with something spinning on a lathe
and the cutter held on the tool post.

Sure a lathe will be more accurate, but if you ain't got one, and the
sanding method is "good enough", then not much to worry about.

Not a
problem of making a woodwork washer but might if concentricity could
be relevant (like guide rollers that subsequently 'stick' in the same
places and then gain further flats). I'm not saying your solution
doesn't work, it obviously would / does, just that I wouldn't use it
against turning or 3D printing (if either were available and the
materials suitable etc).

Jigs can be *really* basic if only doing a one off job. I made posher
versions because I need to true up a number of bandsaw cut cogs.

Oh, I'm sure for certain scales and styles of work they can be more
than adequate.

I just stick the stock in the 3jaw, face up the end, centre drill,
drill ID, slide out to live centre, do OD, part off the lengths ...
done and very true / concentric?

Indeed, I am familiar with a lathe :-)

Just checking. ;-)

snip

And you accurately control the od how?

You are rotating the work on an accurate spindle (i.e. the drill bit
that made the clearance hole though it)

That will often have an element of slack on it and so wouldn't
typically be used as a mandrel on any similar lathe based job.

IME, if you use the shank of the drill for your mandrel, its close
enough in wood.

(if worried and you have a set of tap drills, drill your holes 0.1mm
undersized)


at a fixed distance from the
sander. Once spun through 360 degrees, you will have a very accurate
circle concentric with the pivot.

'Very accurate' ... for that level of equipment / use. eg, I'm not
sure NASA would use it for *anything*. ;-)

Perhaps for a guide bush on one of their bandsaws? :-)


(the non posh version - bit of wood and a clamp or two)

Here using a Forstner bit as a pinion:

http://wiki.diyfaq.org.uk/index.php/...andingRing.jpg

Neat. I can see how it could work but question how good it might be
for even nylon, compared with a lathe ... and especially for anything
longer (tube V washer etc)?

ISTR it was only 10mm required - should not be a problem.

Knowing how even leaning on a toolpost can make a difference, and
certainly any backlash on pretty well any of it, again I can't see
that other than for an approximation of 'true'.

Well that's true of any engineering. You use a three jaw chuck, or make
a measurement with a caliper etc, or even make a cut on a hot workpiece
those will also be approximations. The important question is will if be
fit for purpose when done?


Not that Pete's guide rollers necessitated any super high tolerance
but if you have that to hand ...

I can't remember the exact setup but I was wondering if you couldn't
find some suitable ID/OD bearings to fill the gap and let it run on
them?

Yup that was my first suggestion - its actually what my bandsaw uses
anyway.

My metal cutting horizontal bandsaw certainly does but then they are
twisting the blade though 45 degrees (and managing them back).[1]

I think on my small bench bandsaw (I believe a Burgess as per the
OP's) the 'guides' are simply there to stop the blade wandering
sideways (too far) in use. They don't hold the blade tight enough to
stop it twisting a bit if you rotate the work too fast etc.

Yup, guides not actually touching the blade is generally best practice -
they should only contact under cutting loads.


Cheers, T i m

[1] It was the 6" Clarke one from Machine Mart:

https://www.machinemart.co.uk/p/cbs4...l-cutting-ban/

Bought when I was making the 3 off 6' 6" steel doors and fame to
replace the up-and-over that came with my sectional workshop.

It was doing really well till one of the blade management bearings
failed and from inspection I found they were all 'cheap and nasty'.
Replaced with SKF or similar and they have lasted 10x longer already.
;-)

Yeah, MM stuff does seem to vary in quality from just about ok to not as
good!



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/