By continuing to use this site, you agree to our use of cookies. Find out more

Member postings for Dave S

Here is a list of all the postings Dave S has made in our forums. Click on a thread name to jump to the thread.

Thread: Part breakout /fixturing
03/07/2020 21:54:29

Evening all,

Now my CNC has potential to work nicely I'm thinking about making things

I will be making some small watch parts, mostly 2d in general shape. I think I can effectively cut these out of sheet stock, but what's the best way to fixture it?

Additionally what do you do to prevent the part breaking out and jamming/snapping the end mill.

For general size the would be parts that require 1mm or smaller endmilll to get into internal radiuses, and typically 1mm or so thick.

I occurs to me that I could use (say) 1.5 mm gage plate,surface the top by 0.25, cut around the part leaving 0.25mm bottom stock. Then fill the "gutter" with shellac and flip over, probably on my surface grinder to remove the 0.25 left.

Any better techniques?


Thread: Measuring Gears (including pressure angle)
03/07/2020 16:52:43

Another method, assuming involute teeth:

Take some plasticine or other softish clay like material, create a small block on a flat surface and then roll the gear over it.

The matching rack form is then created and the pressure angle can be measured directly with a protractor from the sides of the flanks.


Thread: 2mm endmill help
01/07/2020 13:52:54

I stuffed the endmill into the collet further - as I could do now its ER16. This spindle is loads quieter, both cutting and idling.

I'll run the job from scratch once the bearings are sorted, but I am also interested - that should allow me to tune which is where I started this whole thing.


01/07/2020 13:03:59

Good news!
My CNC failed to snap my last 2mm endmill last night
The Job started:

and then it progressed some

As it was all looking good I even went and sat in the garden with a beer, keeping an ear out for the noise.
I was back to watch the pocketing process, as the rest of the cuts were small width, and this was the full width, after all the other milling.

took just under 2h20.

End result of the roughing pass:

Somethings gone wrong here, the pocket is supposed to be a curve:

Turns out the recycled bearing on the X Axis has collapsed/ jammed.

Must have happened pretty much at the end, as the other shapes look OK.

That's why I took the stock off - cant run the finish on it until the leadscrew is sorted, which means stripping the table out...

Still, massive progress, and the new spindle seems to have made a difference.
I think its now in the 'tune' stage, rather than the throw things at it stage

Thankyou all


29/06/2020 15:30:33

Many years ago I needed to do some very small precision drilling, so I bought one of ARCs ER16 high speed spindles.

That project got done and as part of workshop tidy the spindle got packed away for “later”.

I don’t think they still sell them, but similar spindles can be found out of China.

it runs of an invertor, 750w and 4-24k rpm IIRC

Bit overkill for this mill I suspect, but already in the workshop and should solve the runout issue.

Seemed worth a try for the time investment


29/06/2020 12:57:16

I've had a busy weekend in the workshop:

I didn't manage to run the CNC , but I did manage to re-spindle it:

Its all assembled, and the runout on this spindle is around 2 tenths with the same end mill as the proxxon one measured 2-3 thou. Speed should be more stable (and upto 24K rpm), and with the ER16 collets I can choke down to reduce the stick out.
I still need to check/fine tune the tram, and ensure the stepper is ok with the additional weight. I did a quick check before making the parts by cable tying this spindle to the existing head, and it drove that OK, so I think it should be fine.
Hopefully Ill be able to report success later in the week.
I'll also look at getting some different cutters if this moves ok - with ER16 I'm no longer limited to the 3.2mm max diameter, so the cutter selection should be wider.
There are pictures of the build on my twitter feed : @SolutionsByDave if anyone is interested.
26/06/2020 20:15:16

I have no idea how to drive a CNC, the little I did was many years ago. So I guessed a tool path, based on the few things I remembered being a typical clearing path as was.
I’ve never “run the numbers” as such on my manual mill, it’s built like a brick outhouse and being manual you can “feel” how the cut is going. It’s a while since I snapped a mill, iirc the last one was a 12mm rougher that I took an overly optimistic DOC into 316.

Im having a “family night” but hopefully some garage time at the weekend will allow me to try some more.

Ive already learnt loads and the video of a successful 2mm cut is encouragement.

If I’m being overly optimistic on my machines capacity maybe I need to rethink my longer term plan, but for now it’s the only CNC I’ve got.


26/06/2020 13:07:39

I sort of expected to break a few things on the learning path, but I was hoping it wouldn't all be at the start....

I clocked the runout on the shank of a cutter held in the collet where it exits the spindle. As I did this with 2 different sized collets I assumed it was the spindle, but maybe all my collets are a bit 'wonky'

Ill clock the actual nose tonight.

Does your spindle feel 'steppy' when turned by had - way stiffer than it fells like it should be? I assume this is cogging on the spindle motor, so if another one does then Ill ignore that.

Might also try a different lump of steel. My other mill has not trouble with this stuff, but then is full sized industrial and I rarely use less than 6mm dia cutters on it - its top speed is about 4500 RPM.


26/06/2020 12:47:46

Yes, as suggested by Jason - engaging more of the flute length in the cut, but with shallowed stepover. It felt like that was a sensible thing to try - it spreads the wear over more of the cutter - rather than relying on just the tips.

Sort of like vertical slab milling (if you squint a bit)


26/06/2020 11:27:59

Sad news, another 2mm endmill has been sacrificed to the cause of getting my CNC to work.

I used Jason's suggested toolpath - although I did re-generate it first as I moved the work coordinates.
However the basics were as he suggested: 0.2mm engagement, much greater DoC.

It all started promisingly:

And it made a couple of the passes all round to the back side before I was called to eat:

During the meal I was keeping a ear out, and when it all went silent I returned, mid meal(!) to discover an ex-endmill

So I paused the machine, and went and finished my steak and had a beer.
Seem the most productive thing to do at that point...

On the plus side the swarf looks more 'swarfy' than last time:

I put the part under the microscope, this is where it snapped from the side:

You can see the not quite left the metal cut that was in progress.
From the top there is no obvious extra width of cut:

I managed to find the end of the endmill from where it had been thrown behind the lathe.
It shows a single tooth has more wear that the other 2, and one of the other 2 appears to be poorly formed in the first place, or possibly chipped - its very hard to tell.

For comparison here is the last unused 2mm, not snapped one:

The edges on that however look distinctly rubbish compared to the 1mm Kyocera mill

After a serious dose of thinking about it I measured the runout on the tool - using the as yet unsnapped 2mm which has a long shank - ideal for an indicator to run on. approx 0.076 mm (a little over 3 thou). With a toolpath supposed to take a side cut of 0.2mm this is quite a significant change in cut load per rev I think.

I also checked with a 3mm tool, in case it was the collet - same result, so I guess that's the natural spindle runout.
I think I need less runout than that to successfully use tiny mills - certainly if I put in a 1mm end mill and the runout is nearly 10% diameter that sounds bad.

Should I formulate a new spindle plan at this point?


25/06/2020 17:59:30

I did some looking at my mills. The 2mm is about 10mm flute length, the real killer I think is the shank length:

Fresh one, last nights snapped one and the proxxon collet. The spindle bore is solid about 1/2 mm from the collet end, so you can see in the next photo how much extra stick out there was because of shank lenght:

Thats about 8mm extra, nearly doubling the stickout.

I also bought 3mm and 1mm, and they also have super long shanks compared to collet length.

Ive chopped a 2mm down (I bought 3 2mm as its the size I'll mostly be snapping using) so I can give that a go with revised tool paths tonight (SWMBO permitting)


25/06/2020 17:51:34

My conversion was done when I had little money, so its scavenged bits from printers and hacked together mounts. The handles are in the mill box, shoudl I ever be mad enough to want to use it manually - I really like my TOS, even for tiny things.

Overview shot, please excuse last nights swarf and spanners...

The X and Y ends have been replaced with alloy, which have pockets for the bearings to sit in. They are held there only be the lead screw tension. I had to turn a couple of adaptors to fit the inside bearing bore - where the leadscrew is thinner for the handle mount.

For the Z - because the lead screw is just 'hanging' I made a new top from bike crank arm with a pocket for back to back magneto bearings:

The top plate is held on with the original screws, which are a bit long at the front, but at the back they also hold the stepper mount piece - scavenged from an old dot matrix printier IIRC.

IT all looks a bit heath robinson, but the motion part seems to work ok. I built the stepper control box and PSU - this was before simple availability from China.

If you want any more details info just ask.


25/06/2020 13:41:08

I've not got to backlash yet - first I need a part...

However my plan is to try and remove it. The screws into the plastic lead nuts are pretty much backlash free, and I have the leadscrew 'stretched' between bearings, with back to back magneto bearings on the Z.

Ill post some pictures of the setup later if its of interest. A friend of mine had the same setup before he sold his mill and moved to an X1 based conversion - IIRC his was pretty much bang on, one of the reasons I picked up the MF70 (spindle speed was the other).

Originally I was running parallel port and tried Linux CNC and Mach 3. This was many years ago - somewhere around 2010 I think. So the tooling was not very simple, and I wasn't interested enough in battling computers after a full on day of real time control software writing.

Now with Fusion and its CAM processor the tooling is much nicer. Just hope that the CAM module remains available for hobbyist use, I'm quite against an annual subscription just to play in the garage - If it was a cost of making money its easier to swallow, but that's a whole other set of things to discuss

I have had a RepRap since around 2012 - and its tooling once sorted is good enough, with the USB interface working with most computers. That and the lack of parallel port machines that were reliable in my garage prompted the move to a USB interface, which lead me to GRBL in a round about way.


25/06/2020 12:48:08

Fusion link:

I've added another setup with smaller stepover, and DoC. It was doing 1.9mm stepover. Hopefully I can try it tonight when I get in from work

I thought the surface wasn't that bad, but if it can get much better I'll be very happy

Ill take a Dremel to the shank and choke down on the tool as much as possible. I'll also get some smaller flute length mills. Didn't really think much about that aspect when I ordered them. My other mill is 1.7 tonnes, but I don't really use less than 6mm in that.

This is one of those times I miss the clumsy bstard, with a wealth of helpful knowledge 5 mins up the road.

Still this forum has given me lots to try, and some encouragement that I should persevere - Thank you all.


25/06/2020 08:46:05

Forgot to include cutter details. They are these ones:

3 flute, don't claim to be coated, but are a dark purple colour on the flutes and not on the shank.

Hertel stuff seems reasonable quality from the lathe inserts I've had in the past.

Seems my numbers are not a long way off. This is the tool path in the CAD:

A variation on the Circle / Triangle / Square as a test piece to make sure the Mill cuts to size and make any backlash visible. Its roughly 30mmx70mm

The cutter last night failed to get through the first 0.5mm layer. It had done most of it, then snapped as it started to do the square section.

The surface finish was quite good, but there are burrs thrown up.

I cant get any less stick out - the shank is as far into the chuck as possible, and the flutes are quite long.

Heres the snapped cutter:

It snapped in the little 'notch' as it began the next pass around the square.

I did notice that it seemed to be mostly taking full width cuts - on a manual I'd usually avoid that - is that a parameter somewhere and should I change it?


24/06/2020 21:30:37

A few years ago I converted a Proxxon MF-70 to CNC.

I never really got in with it, Computers are my day job, so it got packed away and stored in a corner.

I have decided that I should join the 21st century, and also that CNC will make a few things I have planned easier than manual setups, so I have recommissioned it.

its not going well...

I am running GRBL 1.1 and using the UGS code sender. Fusion360 for CAD/CAM.

Took a bit of fiddling, but the motion control seems fine now.

I need to cut steel with it, so I have bought some (more) small carbide end mills.

I thought I’d worked out the kinks, but tonight it again snapped an endmill.

Im running about 10-15k rpm on the spindle - hard to be more precise than that with the dial.

feed rate of 50mm/min, 0.5mm DOC
I realise that too slow and I’ll basically just rub the cut, too fast and snap again.
Am I babying it too much?


Thread: Off cuts ?
16/06/2020 19:20:52

I’ve recently been buying silver steel from Coventry Grinders *Clicky*

Not really offcuts, but 2 13” lengths of 1/2” delivered was under £14.
Half of that was postage, so I normally buy a few lengths, but I needed this for a part for work.


Thread: Measuring wear on a surface grinder
16/06/2020 13:44:54

I've been practicing with my 540 now I have it running. I'm posting this here rather than a new thread as I think the outcome shows I don't have any significant wear on the machine.

I also don't have a master square, so I thought I could make one using 3 import squares - matching each one to another in round robin.
To do that all I needed was a flat flat. Seems like a grinder would be a way to make one of those...

I have assorted scrap waiting to be turned into new things.
In the pile is an old clutch pressure plate - a reasonable lump of old cast iron.
I hacked it into 3 pieces:

Then used my mill to make them 'mostly' straight:

(anvil for scale )

I also made them roughly the same height, and cut a small flat on the curved edge.

Then bunged them onto the grinder:

Taking care not to add heat I proceeded at around 3 thou DOC / 150 thou stepover. 46 K SG wheel.

Turned out nice:

And blackout flat:

All 3 match with no light, and blue against each other really evenly.
Think that shows my grinder doesn't have significant wear.

Next I need to learn how to grind square and arbitrary angles. (tips gratefully received)


Thread: J&S 540 belt keeps coming off
14/06/2020 19:29:20

I just bought a replacement belt from Andemar, had a long chat about various things 540.

The yellow belt is the only available one and is ok for all wheel heads.

Smooth side inward.

Not actually got round to replacing mine yet, as the old belt is currently working.


Thread: Milling a 45 degree 'v' slot
25/05/2020 14:35:27

When I made this one:

I nodded the head, rather than tilt it.
Retramming on my TOS is simple, if heavy. It has a pair of ground taper pins that locate square, one for tilt and one for nod. Within a thou over 12” takes as long as winding the screw back to get the pin to fit.


Magazine Locator

Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!

Find Model Engineer & Model Engineers' Workshop

Support Our Partners
Eccentric July 5 2018
walker midge
JD Metals
rapid Direct
Eccentric Engineering
Subscription Offer

Latest "For Sale" Ads
Latest "Wanted" Ads
Get In Touch!

Do you want to contact the Model Engineer and Model Engineers' Workshop team?

You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.

Click THIS LINK for full contact details.

For subscription issues please see THIS LINK.

Digital Back Issues

Social Media online

'Like' us on Facebook
Follow us on Facebook

Follow us on Twitter
 Twitter Logo

Pin us on Pinterest