Member postings for Andrew Johnston

Assuming it's not a cheapo cutter there are a couple of possible reasons:

1. Too slow a feedrate, with shallow depths of cut you can really ramp up the feedrate, search for chip thinning to see why. With a 4-flute cutter I'd be feeding at more than 500mm/min assuming the cutter is carbide and is running at 2000rpm.

2. Climb ,milling will generally give a better finish, especially in sticky materials like some low carbon steels, in particular EN3B.

Andrew

Although I never completed the governor on my Economy I changed the design of the swinging arm so that I could use a cylinder attached to each side of the arm with an internal threaded rod. If nothing else the cylinder can provide more weight than a sphere.

Andrew

I'm not surprised; it's hardly a tutorial. It's incorrect to state that the Fourier transform of a real function is even. The real part is even, but the imaginary part is odd. Of course the magnitude will be even.

Strange that they don't mention the number of samples per transform. While the number of samples can be any highly composite number, the original FFT works best when N is a power of 2. To the extent that if the data set isn't a power of 2 long, zero value samples are added to pad out the data sequence.

Andrew

Don't know if one is available for the amateur, but they exist for the professional. They're called signal or waveform analysers. They're part spectrum analyser, part oscilloscope and part network analyser. Generally they will perform Fourier transforms, measure the response of a filter or amplifier and allow waveforms to be shown over time. The latter is often called a waterfall display. Successive scans in time are displayed offset in two dimensions so you get a 'waterfall' showing how the signal changes over time.

In the past I've used them for measuring filters with arbitrary pole/zero locations, for assessing the performance of active noise systems (*) and the performance of a 30Hz radio.

Andrew

(*) Being stuck 150ft up on an electro-precipitation stack at a cement works, at midnight, trying to work out why an active noise controller isn't working doesn't rate highly in my list of fun things to do.

That's a wrong assumption, the cutting geometries are different. The ground insert generally has a better defined edge and a higher rake angle. Since it is ground the surface finish is better and thus swarf is less likely to stick to the insert. That's important when turning materials like aluminium. Moulded insert on the left, ground on the right:

Ground inserts are intended for non-ferrous materials, although of course they will work with ferrous. I don't use them for ferrous as the edges are a bit delicate.

Off topic - given your sobriquet are you a real doctor or a medic?

Andrew

Andrew

Edited By Andrew Johnston on 10/05/2020 11:40:06

Andrew

Not so much envy as astonishment! The only cleaner workshops I've seen have been in F1 and Indy car.

I'm not even going to think about showing my mess. Other than to say lathe chuck keys sit on top of the headstock and drill chuck keys are stored in the drill chucks.

Andrew

Chip load; the amount, in millimetres, a tooth cuts per revolution. To get the feedrate multiply the chip load by the number of teeth, 6 in this case.

Andrew

There's hardened steel and then there's hardened steel!

Most of my feed 'n' speed charts list hardened steel, by which they mean around 40-45Rc. I suspect this is roughly wwhere the tool holders will be. On the other hand there's fully hardened steel, in the range 60Rc to 70Rc. The silver steel test piece I turned was ~65Rc.

Andrew

For spot facing the depth of cut is small so I just set the depth stop on the quill and use that. For deeper counterbores, like those in the ellipses in the casting shown above done with a 7/8" endmill, I lock the quill and use the knee to move the work rather than the tool.

Generally I find that tools only tend to grab in brass and bronze.

Andrew

I asked my mum yesterday about VE day. She said she can't really remember what they did, if anything. She remembers listening to Chamberlain's speech in 1939, announcing we were at war, like it was yesterday, but not VE day. She lived on a poor council estate in London, so the cash and resources needed for any sort of party simply weren't available. My grandfather was in the building trade; if it rained and you couldn't work you didn't get paid. which meant the family didn't eat. So I suspect there was just world weary relief that it was over. Except of course that it wan't, as rationing got worse and continued for nearly another decade. Rather oddly, bread wasn't rationed during the was, but was afterwards.

Andrew

In theory it's not ideal, as a slot drill is ground slightly concave. But in practice it's fine. However, I'd use a different sequence of operations, especially if the surface isn't machined:

1. Spot drill, preferably carbide as it's much stiffer than HSS

2. Drill clearance hole

3. Spot face - while a slot drill is centre cutting if you're spotting only the cutting speed at the centre is zero. Best to avoid that by drilling the clearance hole first.

The above is the sequence I used recently on this rear cylinder cover casting:

Andrew

A facemill is the wrong tool for machining hardened steel. The inserts are not designed for it. I machine HSS toolbits with standard carbide endmills. The swarf should be coming off red hot. Hardened steel can be turned with ordinary CCMT inserts, but they don't like it. Much better to use CBN inserts.

Andrew

Amazing, being quoted and he hasn't even posted in this thread. That's definitely action at a distance.

Search for member JasonB.

Andrew

Bother, shouldn't have taken the time to proof read.

Edited By Andrew Johnston on 07/05/2020 17:17:25

A rule of thumb for HSS cutters in low carbon steel is 100ft/min cutting speed. For a 10mm cutter that's about 970rpm. Slightly slower is better than slightly too fast. A depth of cut of 0.5mm is small, but should be no problem. How do you know it was cutting? I'd be feeding at 100mm/min or more. If you know the pitch of the leadscrew you can work out how many times a minute you need to turn the handle.

The swirls can be produced by several issues. First as the tips wear, or if they're badly ground in the first place, they will cut inconsistently. Second, dragging swarf around with the cutter scores the work as the swarf gets trapped under the cutting edge. Third, the material matters. A 'sticky' steel like EN3 or hot rolled steel is much more prone to this problem than EN1A. The ridges between cuts imply that the cutter isn't cutting cleanly. Assuming the mill is trammed, you shouldn't be able to feel any ridge. Do you know what spec the steel is?

Fair enough on the marking out. Ahem years ago I was taught marking out and centre popping but now I only use it for sheet metalwork, where I'm cutting/drilling by hand. But of course one needs confidence before flying with no safely harness.

Andrew

A few notes:

I normally aim for a thou or less when tramming my mill, over a diameter of 200mm or so.

No idea what happens if I lean on the table after adjusting the gibs; never felt the need to measure it. My manual mill is pretty worn so adjusting the gibs is a compromise anyway. I just do it by judging a slight 'drag' when moving the table.

The manual mill is a knee mill and the column is large by hobby standards. The CNC mill has a conventional column but I've never leant on it to test. I just make parts.

The machining marks are indicative of a blunt or pooly ground cutter. Where did the cutter come from? Is the cutter HSS or carbide? Your speeds are a bit fast for HSS but slow for carbide. If there was vibration and noise something is wrong. As before a worn of poor quality cutter, or the most common early error is feeding to slowly. Rule 1: the cutter needs to cut not rub. Rule 2: cutters are consumsable items, don't try and make it last for ever by pussyfooting about on feedrates.

I wouldn't bother with cutting oil, especially if it isn't flood coolant. If it is applied in dribs and drabs it just makes a s sticky mess which is sure to cause re-cutting of the swarf, damaging the cutter and making a mess of the finish.

Machining a block to an accurate thickness on a mill is seemingly simple, but is much more difficult than it looks. Even if the mill is perfect the results are highly reliant on how the work is clamped and the quality of the machine vice, if used. If I get better than 2 thou on a 6" square I'm happy, and that's on a large mill with a industrial quality machine vice.

Andrew

PS: If I've an edge finder and DRO available I wouldn't bother marking out and centre popping. Just zero the DRO on a convenient corner, move to hole position and drill - quick 'n' easy.

Edited By Andrew Johnston on 07/05/2020 16:37:55

No need to worry about that. SoD is an amusing read, even if I don't always agree with him.

Andrew

Full form threading inserts are subtly different to normal inserts in that they don't tend to have chip breakers and have sharp edges. So you don't need to run them in quite the same way as standard turning inserts. Here's a M16 thread screwcut in EN16T:

Turning speed was 40rpm as it is a metric thread on an imperial lathe, so I had to disengage the tool manually rather than rely on the high speed threading unit. If it had been an equivalent imperial thread I'd have been cutting at 180rpm.

Andrew

I'd slightly disagree with DaveH. I wouldn't be surprised to see a smeary finish on EN3, but not EN1A for instance. It's not clear how big the test piece is. But assuming 1" diameter I'd be running at 800rpm, 4 thou/rev minimum feed and 10 thou or more width/depth of cut depending upon the nose radius of the insert.

Coolant can have an effect on surface finish dependent upon material. But it really needs to be flood coolant. Dabbing or squirting on now and again is a waste of time. Unfortunately there are a lot of people on the internet who think they are a legend in their own lifetime, and a few that know what they're doing. The problem can be sorting out the wheat from the chaff.

Andrew

As far as I'm aware the indexer handle is just a bent bit of metal with the thread on the end. The ends of the pipes on the spare pump just have small brass unions on them. I've no idea if they're standard fittings. I've never taken mine apart to have a look. The manual shows three hydraulic pipes, and there are three connections at the tank end, but only two appear at the attachment end. The pipes appear to be plastic so shouldn't be too difficult to get replacements. The manual also says the motor is dual voltage, so it may well be, just not with the standard means of swapping.

Andrew