I’m late to the party on this project, but those wemo boards are awesome. I think I’ll be ordering a few.
If you want to try this helper library I made, let me know. I found it useful, but AFAIK, no one has tried it.
So, went on vacation, bringing my “workshop” laptop with me to code/check code on it, and once arrived on site, the laptop refused to get up. (Fan error). So now it’s at the repair shop, and I can’t work on this side of this project until it’s repaired (should be one or two weeks more) 
Will try and work on something else in the meantime, just cut the first good plate of my MP3DP V2 (and for once, I got good cutting results ! \o/ more on that later.)
So the 660 is a DC motor with a rectifier? How would all of this be different if we just yanked out the rectifier and ran the spindle with DC? Isn’t DC less tricky to control? Wonder what sort of DC voltage/current would be necessary and how best to supply it? Just pondering. At any rate, you guys have blown my mind several times in this thread. I enjoy hanging with people smarter than me, but damn I felt dumb when I read most of this!
Pfffff, I have no idea. I just guess and plug and until it works. I feel like it is group learning. There are a ton of page views but few comments. I feel like I am having my back watched. No one says anything unless I really do something stupid and then they will chime in…or hit a milestone and they want to try it.
Zero point crossing, snubber circuits…I only know what I found on wikipedia.
The AC board I am using should be much cheaper and more safe than high voltage DC power supply.
600W is no joke. I don’t know what voltage these are designed to run at (it could be anything if there’s a transformer in the box somewhere). Something like this might work (it can do 900W, supposedly):
You still need to have the power available, do you’d need to replace the power supply with one that’s compatible or come up with something that would reuse the original in a “safe” way. In this case, the power supply is the rectifier, transformer, and filters.
Interesting idea. I’m not sure it would end up easier it cheaper than the solution Ryan is building.
Another point of confusion is that there are brushless DC motors. They are easier to control (well, harder to power, but easier to control). The 660 is a brushed motor. I think many of the no name chinese spindles are brushless.
I debated suggesting taking apart the DW660 to get to the DC voltage directly. I’m sure it would work, but it would not provide the more general solution for other machines. So, I started the thought exercise of what it would take to design and build a single phase variable frequency drive with the capability to adjust voltage and frequency - looks like the cost of a commercial one is >$100. I’d like to think it would not be much more than a nano (or ESP32), a display, a rheostat, some switches,resistors, capacitors, inductors, diodes and transistors. Well, maybe it wouldn’t be that easy, but it could be fun. I think I’d have to get (or build) an oscilloscope first just to satisfy my curiousity on the waveform at the various stages - it would be neat to show how square waves can produce sine waves using pulse width modulation and a filter.
What’s stopped me from doing anything yet is:
- As Jeffeb3 said, 600W is no joke and the circuit would be using 120VAC rectified to be the DC rails. So, how do you protect the people from the electrons?
- Component selection to keep the cost down, yet meet desired goals for current and voltage. I'd like to think parts could be bought for under $80, with most of the cost going to the transistors, processor, and display.
- What other things need to be added to make sure you don't break anything, such as how do protect against voltage spikes when you turn disconnect the motor when it is running?
Just a random thought here, I have no idea if it is completely irrelevant or not. Treadmills drive DC motors with variable speed PWM controllers. Could one of those controllers be used to drive a DC motor in a router? Most of those DC treadmill motors are in the 1HP range, which I believe would equate to about 750 watts.
I’ll have to take mine apart and look. I didn’t think it was a DC motor. I would think it would have circuitry for the soft-start but I’m pretty sure the motor is AC.
Pretty sure it is just a switch and brushes unless they have something really hidden in there.
[attachment file=“IMG_20180503_120446~2.jpg”]
I’ve been viewing this thread a lot. It’s interesting stuff. Is it not possible to strap a today encoder to the side of the router? Or maybe mount the shaft directly to the top of the fan? They seem to be pretty well sealed units which would eliminate any dust getting in. I guess price could be an issue again though
I think the fan is possible, I am just concerned with regular dust build up on the “optics”. I am going to put a hand full of sensors and resistors in the shop so anyone who wants to check it out can without buying a bunch of them. Nano’s and AC dimmers are ordered but not available yet.
Placement does seem to be fairly distance sensitive to get a really strong signal but a week signal is really easy from any distance and angle. It comes down to how well the rising and falling part of the arduino sketch picks it up.
How about a cap like this over the end of the spindle
[attachment file=59388]
[attachment file=59387]
Tried printing this one but I think I need a better set of calipers to get the size correct.
Yeah that would work, but that is a bearing so liquid glue is a no go, and I am not sure how hot that shaft gets. Todd sent over some reflective material to try as well.
I believe that it would not need glue. Looks like I can get a nice snog fit around the molded case that hold the bearing.
Shaft heat should not be a problem, the only thing going on the shaft would be the reflective material.
I am trying not to make any changes to the case so I can use my tree year extended warranty if it burns up.
[attachment file=59413]
Oh, I understand, to hold the sensor. Actually that is better than what I was thinking about, Nice.
Perfect. I was messing with orientation as well, you mount would make that so much easier. I use the larger resistor for the LED to dim it a bit, seems to give a better signal bit try them both.
Attached is the simple sketch I use to test, with the plotter on.
ReadAnalogVoltage.zip (848 Bytes)
Parts for those of you that might want to get in on the experiment.
Still can’t get out there, but it never hurts to plan ahead. Order placed. 