Member postings for Andrew Johnston

Must be transgender as it's a Victor.

Andrew

Both have their uses. It's all to do with the size of the gullet. With a fine tooth saw a large depth of cut is a problem as the gullet may get full of swarf, which will cause the saw to jam. For a shallow depth of cut (like slotting a screw head) a fine tooth saw is fine, and can be feed at a higher rate as it has more teeth. For a deep depth of cut a coarse tooth saw has a much larger gullet which will be less prone to jamming.

Andrew

The teeth on this counterbore were filed by hand, but they still did a precise job on the part shown:

In the holes can just be seen a hollow spigot that will hold a wing valve. The spigot is counterbored so that it stands proud from the outer shell.

Andrew

ER collets were originally developed for tool holding. Tool shanks are normally hard. So when the collet is closed they do not 'give'. So if using a larger than needed collet the contact area becomes a line rather than a surface. Collet, and tool, runout and other parameters are only guaranteed if the collet is used at it's maximum value.

Andrew

None of my milling machines are bolted down. Doing so may help damp vibration, but won't stop it. A better idea would be to cure the vibration in the first place. The first thing I'd look at is cutting conditions - quality of cutter and it's sharpness and cutting conditions such as spindle speed, feedrate and depth and width of cut.

Andrew

Bottom right to top left equals conventional milling.

Always lock the quill when it's not being used.

If the cutter span in the collet then it wasn't tight enough. But given it started cutting again I suspect it's a combination of a loose cutter and unlocked quill. Pretty tight is a bit vague, I'd suggest around 100Nm, ER collets need to be tightened rather more than you might think.

Andrew

They all seem to be fine tooth saws, so only really suitable for shallow cutting, such as slotting screw heads. For deeper slots you need coarse tooth saws. Here's a picture of several saws (all coarse tooth apart from bottom right) for comparison:

Ideally for deep cuts with a coarse tooth saw the saw needs to hollow ground to reduce rubbing in the slot.

Andrew

All cutters run eccentric to some degree. Slitting saws probably just highlight the effect due to diameter and shallow cuts. Eccentricity will be due to tolerance build up from grinding spindle runout, accuracy of the centre hole diameter, grinding errors and runout on the teeth, and spindle runout and cutter fit when using the slitting saw. I'd be surprised if a grinding wheel wore significantly while grinding the teeth on one saw. Modern CNC grinders have means of detecting wear and dressing as required anyway.

Here's a picture of a coarse tooth slitting saw:

The gullets are quite rough, almost certainly milled rather than ground. When grinding the teeth only a short (~20thou) primary clearance on the tip of the tooth is formed. The OP specifies a fine pitch slitting saw. These are intended for shallow cutting only, such as slotting screw heads. The shallow cut and reduced feedrate will make any eccentricity more obvious. For the coarse slitting saw shown I'd be happy taking a 0.5" depth of cut and 4 thou/tooth chip load. That reduces the effect of any eccentricity. Incidentally the slitting saw is hollow ground. About half an inch from the periphery the width is 10 thou smaller.

Andrew

Good grief, this all seems way over-complicated. It should be possible to make a working spotface cutter using a file to cut the teeth. The fundamental question is why didn't the orginal spotfacing with a slotdrill work as expected? If it's a problem with the setup, milling machine or technique it needs to be bottomed out rather than covered up.

Andrew

Some of the statements by SoD need elaboration. The PWM frequency used in VFDs is usually a few kilohertz. The reason that the signals radiate at much higher frequencies is because the edges of the waveform are fast, probably a microsecond or faster, fall and rise times, not because of the fundamental frequency.

All AC currents will radiate an electromagnetic wave, even at 50Hz. However, there are two things we can say about radiation at 50Hz. First the wavelength is long (6000km) so even the power distribution network is electrically short, and electrically short antennas are very inefficient. Second, in the near field (below about half a wavelength) the magnetic field dominates. So in practice interference from 50Hz is by magnetic induction.

Copper or aluminium shielding will not stop a magnetic field. Instead you need a high permeability material such as mu metal.

Ground planes are a whole different ball game! When laying out PCBs, especially a mix of low noise analogue and switchmode power supplies one needs to 'visualise' how and where the currents are going to flow to prevent interference. Way back multiple ground planes for different areas where in vogue. But the modern view is to use one solid ground plane with possibly small local planes for noisy circuits like switchmode supplies and be careful how high current tracks are run. The return current in the ground plane will try and follow the track as closely as possible.

Just because the OP says he doesn't have a problem doesn't mean it isn't there, and is possibly affecting somebody else. It isn't clear if he's actually measured the radiation?

Andrew

I'm posting in the aircraft thread rather than 'What Did You Do Today' to avoid some smartypants pointing out that it's not model engineering.

Last Friday, much to everyone's surprise, the government announced that private flying could start again, provided 'social distancing' was mantained. In effect that means solo flying in private, but not club, aircraft. Like all clubs my local gliding club has been caught on the hop. We started flying again this weekend, private gliders only, flown solo and aerotow only. They've introduced a traffic light system, green, amber or red. Fortunately I'm in the green category, allowed to do what I like based on experience and currency. Since you can't fly with an instructor and maintain 2m distance a lot of rules and regulations have been temporarily binned. I actually last flew a glider in March, two days before the lockdown was imposed.

A number of aircraft airworthiness reviews have been on hold as I haven't been able to go to the club to do the required physical survey. Today I've done three surveys, including one of my gliders which we also rigged. Which was challenging while maintaining 2m separation. The next couple of days will be spent ploughing through paperwork for Airworthiness Review Certificate renewals. This is just the start. A lot of people have been unable to do the annual inspections due to lockdown. So I expect there is now going to be a flood of inspections and ARC renewals to do. I might even fit in some gliding and power flying! I've almost finished the regulator design for my traction engines, so I'd like to finish those too!

Andrew

Having acquired a manual vertical/horizontal rotary table many years ago I rushed out and bought a 3-jaw chuck for it. But I've never fitted it and in retrospect have never missed it. My rotary table has a parallel centre hole, so it's easy to make mandrels to align parts. I've never understood the 'advantage' of having the centre hole be a Morse taper.

To be fair my dividing head has an indexing plate behind the chuck so it's easier to use that, as a sort of spin indexer, for machining squares and hexagons.

The rotary table for the CNC mill has a commercial 5C collet chuck on a home made backplate, as most parts needing a 4-axis CNC mill can be mounted on arbors.

Andrew

When I fitted a VFD for the high speed spindle on my CNC mill I used 300/500V rated 3-core plus screen cable from Farnell. It makes no difference if the motor is star or delta, you only need the 3 phases connected. I connected the screen to earth at the VFD but left it unconnected at the motor. The motor casing is earthed via the CNC mill.

A sweep with a spectrum analyser didn't show any extraneous RF, just the WiFi at 2.4GHz.

Andrew

Shouldn't be a problem. Although I didn't completely part off the item shown below due to the overhang and tailstock support I had no problem cutting the groove:

The OD is 3-3/8" and material is cast iron.

Andrew

I'd agree that the angle bracket is flexing; you only need it to deflect half of the diameter error. As to the error with the cross slide locked, that'll probably be down to something else. May tool or work deflection? Or the lock isn't that effective for the loads being applied?

Andrew

Some advice:

Recycle all your unknown metal and buy known material from a reputable source - otherwise you're just wasting your, and our, time.

Andrew

The obvious assumption is that the stop is not rigid enough, but pictures will help. For comparison I'd expect to hit a diameter to a thou or less just using the same dial reading on my cross slide.

Andrew

Those figures make more sense. So the transformer only takes one phase from the overhead as per Stuart's diagram, not two as originally stated?

Andrew

What Stuart says make sense. But the OP thinks he has something different, as he talks about the transformer taking two phases from the overhead? The voltage readings he gives for phase to neutral and phase to phase on the outpur don't seem to make sense for 2 phases of a 3-phase system or for the configuration elucidated by Stuart. May be it's a one off special?

Andrew

I didn't explain myself very well, although the inference that I meant not as hard is rather a leap of faith.

What I meant was that brine seems to be less prone to creating a blanket of steam around the work leading to less hardness. So less violent in the sense of less vigorous boiling of the liquid.

Andrew