The lathe is a simple enough tool to understand, but requires much practice to truly master. During the turning process, it’s often necessary to inspect the workpiece. This generally necessitates stopping the lathe, waiting for everything to spin down, and then starting again. This can be a major time sink when added up across the full scope of a project. However, the magic of strobes can help.
The basics of [Darcy]’s project will be familiar to any hacker who has worked with rotating machinery before. The rotational speed of the lathe is measured, in this case using a reed switch and a magnet. This signal is fed to a microcontroller, which controls the strobing of an LED lamp. By synchronizing the flashes to the speed of the lathe, it’s possible to view the workpiece as if it were standing still. By adjusting the offset of the flashes to the position of the lathe, it’s also possible to rotate this view to see the entire workpiece – all while the lathe remains spinning.
Further photos and videos are available in the Reddit thread. [Darcy] reports that despite his best efforts, he couldn’t quite find a business case for producing the hardware commercially, but the idea was too useful to leave languishing in a notebook. We’d love to hear your ideas on how this could improve turning projects, so be sure to let us know in the comments. If you’re just getting started with turning, it might be worth cutting a test bar to make sure your rig is up to snuff.
That sounds like an awesome tool but one which might lead to anyone prone to being a klutz or a spazz around mechanical things having a momentary brainfart and, falling for the illusion of a stationary workpiece, reaching out to touch it and *thwack* touching a corner spinning by ending up with a pulverized finger tip.
I’ve seen someone almost make this mistake with a timing light working on an old car, so it’s not a new hazard so much as a new way for the klutzes of the world to hurt their hands.
First I thought nonsense, but then I clearly imagined a situation where you show it to someone, he notices an imperfection and points to it with a finger.
I don’t think a regular wood lathe would produce a pulverized finger tip, but it may make one willing to lower the head dangerously close to the spinning bit. Hair etc, really dangerous.
I wondered about the safety of this too. I’d thought about doing exactly the same thing on my metal lathe, so I could check surface finish etc, but was also worried about the safety aspect.
One approach that may work would be to enable the strobe with momentary switches and put one at the headstock and one at the tailstock (or at least separated by more than a hand-span), so you’re required to put both hands away from the work to enable the strobe. It wouldn’t help as much when there’s more than one person about, but feels like an improvement
I’d probably add a sliding plexiglass handguard there and only allow the strobes when the handguard is in place.
That would make it too tedious to be useful.
Rather, use push-button-only operation, so the moment you take your thumb off the strobe stops. That way you need to actively fudge it up by pushing the button while you’re grabbing the “stationary” workpiece.
But the fundamental problem is that this is a neat invention, but ultimately trying to find a real use case is hard. You can only look at the thing – and that doesn’t tell you very much. If you want to e.g. take calipers to the work, you usually have to stop it for real.
Once the strobes have the object appear stationary, you can use computer vision to perform non-contact measurements. Sounds like a neat extension…
Wolfgang, just use a bright constant light and a short shutter time on a normal camera and use the now very nice image to do your measurements with no need to try and sync the strobe to the part and the camera shutter to the strobe, not to mention the rolling shutter effects that could happen.
Really this is only good for making it look stationary to a human.
Cynyr,
Care to speculate how much light you need to yield a short enough exposure to get an un-blurred image? (hint: more than you’re likely to want to be near)
Why didn’t you just suggest a xenon flash? At short range even cheap flashes get down to 50 microseconds — sufficient to “freeze” a few hundred rpm.
Company I work for issued a safety notice because a worker who “knew what they were doing” having worked the same machine for 5 years put their hand around the safety guard to unjam it due to wanting to meet a production target and score an efficiency award.
Against policy I might add.
The machine chomped their hand and they needed surgery.
The solution ?
More elaborate guards and two dead mans switches instead of one.
Sadly I cannot share the before and after pictures, but suffice to say the before pictures made it clear you’d be a dumb f*** to put your hand in there.
The after upgrade pictures were to the point of taking the piss, like loading a shotgun with safety guards and blasting away for a good hour or so.
TLDR: Some people are just stupid.
Instead of playing the safety nanny I’ll just assume this guy has some sensible brains in him and knows what he’s doing. I’m also a big fan of the Darwin Award.
On the other hand. Some of those wood turners like to make bowls out of the underside of tree trunks, the part where all the roots stick out. If you put that part into a lathe and spin it around you see absolutely nothing but whirls and you’re just guessing where you put the pointy bits. You mostly do it by feel than by view.
simple solution: nagging mother. every incompetent needs one by there side constantly.
The idea doesn’t look to be to leave it on long term… rather to use it for bursts of tens of seconds (if that). The foot pedal (press-hold to keep the strobe active) and warning beeper are also there to remind you. I read the manual :-)
This could be quite dangerous. Touching a piece of wood or metal spinning at high speed… Ouch….
I remember my dad, who was working on large lathes his whole life, told me, that there must be at least one incandescent bulb to light the lathe. Incandescent lamps don’t blink. And the main fluorescent lights (which blinks with the mains frequency) should be connected to different phases to minimize the stroboscopic effect.
So much this. Doing wood turning under fluorescent lights will drive you bonkers – not sure what the new LED tubes would do but you can run bright LED worklights on DC if necessary.
Some blink, some don’t. Take a clear yellow LED and connect it to scope leads, then take it up to the light. The diode acts as a photocell and you should be able to see a waveform appearing.
Incandescent bulbs show a sine wave with DC offset – they do strobe a little.
Another trick I read about is to use a fluoro fitting with two tubes and two independant circuits inside, but remove the capacitor from one.
The whole leading and lagging current thing has the effect of one tube being on when the other is off so they pulse in turn and even out to a constant light.
cool!
Both good and bad (misleading) advice, Bretja. The reason you connect your lights to a different phase is not due to strobing. It’s so when you trip a circuit breaker, you don’t have spinning machinery in the dark.
Two phases from the same generator run at exactly the same frequency, so there will be no difference in the strobe effect.
What about VFD’s ?
You could certainly use a VFD to alter the freq. of the light!
My uncle, who had worked in many many engineering workshops and precision grinding establishments in his time, was very particular about the type of worklight used over his 6 feet between centres metal lathe at home, precisely to avoid stroboscopic effects and accidental injury.
The interlock memtioned above is not a silly idea, or some other flashing hazard light as a visual indicator of danger.
On a related note, i have wondered about using a stroboscopic light to help balance a wobbly ceiling fan.
hey hey! strobe to balance a ceiling fan! that’s a good idea thanks!
For the patenting aspect of it: digital strobe lights on spinning machinery is a trivial invention that doesn’t pass the novelty requirement for patents. It’s been used in various devices before, such as on turntables where strobing a light at certain speeds allows you to exactly tune the motor for the correct record speed.
It’s the use of the microcontrollers (and the way they off unique control) that make it novel… enough that patent lawyers had started getting involved
So much like my old timing light.
So, what’s the correct spark advance setting for maple?
And I wonder where I could clip on the pickup.
Maple doesn’t have a mechanical distributor.
However, assuming you are referring to the other hardwoods, the answer is 10 degrees off of top-dead-center.
But Marples has a number of distributors!
Cause Chevy didn’t make a 327 in ’55, the 327 didn’t come out till ’63. And it wasn’t offered in the Bel Air with a four-barrel carb till ’64. However, in 1964, the correct ignition timing would be four degrees before top-dead-center.
First off, this can only be relevant with word working. With metal, the only situation I see where one would stop the spindle, and not use some form of tool for measuring, would be to see if the piece “cleaned up”. Sharpie markers or dykum can tell you if it’s not clean with the spindle still kickn. Cheap, easy, SAFE
Second, this is an accident waiting to happen. Lets make a spinning danger appear still! I wouldn’t work on or near a machine if it had this “feature”. For wood working, which I have no experience, I feel it’s still to much of a risk.
That all being said, I still find the experiment awesome. But that’s where it needs to stay.
Some time ago I saw a metal CNC video on youtube, which was filmed with a strobe light.
It did internal turning, and the workpiece was cut in half lengthwise.
The strobe was used so you could see very clearly how the internal turning was done.
“First off, this can only be relevant with word working.”
Then it is apropos to have it here on WordPress!
B^)
Don’t people read the other comments before posting? Almost every comment so far is about the unsafe nature of this idea. Did it need so much repetition?
Anyway, safety emasculates Darwinian selection, and is overrated as a lifestyle.
Yet another product which is grossly over engineered.
The commercial variant of this would not have the fancy display, nor a uC, but use a 555.
Or it could use a mechanical way to rotate the Hall switch (sorry reed switch) around the axle.
Or it might have a uC. The’re about the same price as a 555 nowaday’s.
Having the strobe being triggered multiple times in a revolution would seem to be usefull though.
If you removed the display, you’d remove most of the functionality that made this useful… Many lathes don’t even have RPM indications, so straight away there’s extra value for anyone in that position.
I’ve done some wood turning, and I see absolutely no use for this. It takes very little time for the lathe to start or stop, and you don’t need to do it that often. You can tell when a surface is smooth by the feel of the tool.
If you’re working on something smooth/round… then the strobe doesn’t offer any advantages to you. If you’re making winged bowls, off-centre turning, or have lop-sized pieces like roots or drift-wood, then there are definite advantages when making fine cuts and being able to see your approach. But yeah… if you’re just making bowls, it offers no real advantage :)
I implemented a strobe for my metal lathe a few years ago; it is described on my Web site at http://www.fenichel.net/pages/Indoor_Activities/metalworking/lathe/timing%20light/timing%20light.htm. It turns out to be only rarely useful, but every once in a while it saves a lot of time.
I may recall correctly that this was a problem when fluorescent lights were first used in factories and shops, as they strobed at the same rate that AC motors turned, and there were serious accidents. Was it fixed by using phosphors with longer decay? Or was it something about ballasts and their phases? Or Both?
maaaaybe they used the 3 phase power that I’d assume they had for machine tools and put some fluorescent lights on each of the three phases? Then the time between each tube strobing (single phase = 120Hz so three equally spaced phases = 360Hz ) and the phosphor persistence (4ms?) might result in illumination that was sufficiently close to constant?
Use at least oneIncandescent bulb over the workpiece.
Modern electronic ballasts don’t have this problem at all thankfully… I have 10x 4ft tubes in my workshop, above a lathe, with no issues what-so-ever. I did a bit of research before installing as I was concerned about the same thing. The old-school magnetic ballasts usually flickered at 50/60 Hz
I’m a full-time professional woodturner. I write a monthly article for an international woodturning magazine that goes to over 60 countries worldwide. I’ve been Chairman of a national woodturning association. I’ve worked with turners from all over the world. I demonstrate at clubs all over the UK. I know and have met, literally thousands of woodturners. This is an entirely contrived problem for which you are searching for a solution. Not only does the problem not exist, but your solution presents great potential for even greater, real, and potentially injurious, problems.
There are a number of potential reasons to wish to stop a lathe rotating; because you’ve finished turning, because you’ve finished a section of turning, because you want to inspect what you’ve done so far, because you’ve heard a noise that suggests there may be a fault in the wood, because you want to go and make a coffee and take a break. All of these reasons, and any others you may imagine, require that the lathe is stopped. Actually stopped. You can not examine a workpiece if its rotation is artificially halted by the use of technology. “Seeing” is not examining. It’s a potentially dangerous and completely pointless contrivance. Just stop. Put your obvious abilities to a problem that actually exists. If you find the time it takes to stop the lathe hampers your ability to turn efficiently then you need a different hobby. Keep in mind that just because you can do something it doesn’t necessarily mean that you should.
I hope I never get to review such a product.
“I hope I never get to review such a product.”
You just did.
B^)
This is what I wanted to make when I worked in an engineering test lab. We had strobe lights but only worked manually or timed with a single pick up. I wanted to be able to turn the strobed view to see the other side of whatever was spinning.