A few mods/projects for my TaoTao Atm-50 A1 By glavey (Read 25,059 times)

[glavey]

The extruder on my printer can go above 250°C. I print with PLA at about 195°C. I have given thought to nylon filament... However as the temperature of the printed material goes up, the chance of and degree of parts warping and shrinking as they cool increases.

I don't have a heated bed (platform you print on) on my printer yet, that is why I can only print in PLA so far. If I had a heated bed, I could print in ABS (the stuff legos are made from). ABS has a higher melting point than PLA, but is much more prone to warping and it emits a noxious smell when heated. PLA, by contrast, is made from corn by-products so it almost smells like cream corn when I am printing with it.

On to other news; I had another successful test with the arduino and the stepper motor. I recorded a video this time. I got the motor to move about the same distance and speed as a potentiometer I was turning by hand. This might not seem like much for a week's work, but I'm darn proud of it and myself.

Now, I have to take what I learned from doing that and make the stepper motor into a tach like I did with the servo.

[scooter2]

Apr 4, 2015 10:52:51 GMT -6 glavey said:

The extruder on my printer can go above 250°C. I print with PLA at about 195°C. I have given thought to nylon filament... However as the temperature of the printed material goes up, the chance of and degree of parts warping and shrinking as they cool increases.

I don't have a heated bed (platform you print on) on my printer yet, that is why I can only print in PLA so far. If I had a heated bed, I could print in ABS (the stuff legos are made from). ABS has a higher melting point than PLA, but is much more prone to warping and it emits a noxious smell when heated. PLA, by contrast, is made from corn by-products so it almost smells like cream corn when I am printing with it.

On to other news; I had another successful test with the arduino and the stepper motor. I recorded a video this time. I got the motor to move about the same distance and speed as a potentiometer I was turning by hand. This might not seem like much for a week's work, but I'm darn proud of it and myself.

Now, I have to take what I learned from doing that and make the stepper motor into a tach like I did with the servo.

Great job, glavey! In case anyone is wondering, ABS melts at 221F 105C according to a source online, so it would be a no go for an oil gear. I see heated beds on ebay for $9.00 and up. 

Back to the tach. How are you getting your tach signal and how does it come out of where you get it, as a voltage?

[glavey]

At the moment, I'm using a hall effect sensor and a small electric motor which has a printed flywheel with magnets in it simulating an ignition signal. The hall effect sensor I have is very similar to this one. Normally, when there isn't a magnet near the sensor, there is an output of +5 volts (or what ever voltage you feed into the sensor, between 3 and 5.5 volts). Now there are two outputs on the sensor; analog out and digital out. The analog out will drop in voltage linearly as a magnet gets closer to the sensor. Once the magnet is close enough, the output will drop all the way to 0 volts or ground. The digital out has a threshold the voltage must drop below before going straight from +5 volts to ground, no linear transition.

I'm using the digital out right now, if I use the analog out, the arduino starts taking incorrect readings intermittently.

When I get to the point where I need to test this with the scooter, I plan on tapping off of the primary side of the ignition coil, like the tach I took apart did. I can duplicate, up to a point, the circuitry it used to lower the voltage to create a nice clean signal that the arduino can safely use.

[scooter2]

Apr 4, 2015 16:56:06 GMT -6 glavey said:

At the moment, I'm using a hall effect sensor and a small electric motor which has a printed flywheel with magnets in it simulating an ignition signal. The hall effect sensor I have is very similar to this one. Normally, when there isn't a magnet near the sensor, there is an output of +5 volts (or what ever voltage you feed into the sensor, between 3 and 5.5 volts). Now there are two outputs on the sensor; analog out and digital out. The analog out will drop in voltage linearly as a magnet gets closer to the sensor. Once the magnet is close enough, the output will drop all the way to 0 volts or ground. The digital out has a threshold the voltage must drop below before going straight from +5 volts to ground, no linear transition.

I'm using the digital out right now, if I use the analog out, the arduino starts taking incorrect readings intermittently.

When I get to the point where I need to test this with the scooter, I plan on tapping off of the primary side of the ignition coil, like the tach I took apart did. I can duplicate, up to a point, the circuitry it used to lower the voltage to create a nice clean signal that the arduino can safely use.

Good deal. Then you can map the stepper position to the output voltage. That will be the easy part. I did that with a winch servo and a compass sensor so the servo always points north. If you wanted to go digital maybe you could have the arduino count pulses and translate that into specific positions on the dial, but I'd think using the analog input would be easier. Either way, even if it's not linear, you can make the stepper go anywhere you want for a given voltage or pulse count.

[scooter2]

Apr 4, 2015 16:56:06 GMT -6 glavey said:

At the moment, I'm using a hall effect sensor and a small electric motor which has a printed flywheel with magnets in it simulating an ignition signal. The hall effect sensor I have is very similar to this one. Normally, when there isn't a magnet near the sensor, there is an output of +5 volts (or what ever voltage you feed into the sensor, between 3 and 5.5 volts). Now there are two outputs on the sensor; analog out and digital out. The analog out will drop in voltage linearly as a magnet gets closer to the sensor. Once the magnet is close enough, the output will drop all the way to 0 volts or ground. The digital out has a threshold the voltage must drop below before going straight from +5 volts to ground, no linear transition.

I'm using the digital out right now, if I use the analog out, the arduino starts taking incorrect readings intermittently.

When I get to the point where I need to test this with the scooter, I plan on tapping off of the primary side of the ignition coil, like the tach I took apart did. I can duplicate, up to a point, the circuitry it used to lower the voltage to create a nice clean signal that the arduino can safely use.

Double post. BTW I have been reading up on 3D printers. That 2beigh3 looks promising and can do nylon, but it isn't cheap and you have to build it yourself. 

[glavey]

I have the sketch on the arduino counting the microseconds it takes per revolution and then calculating the rpm based off of that. At first I tried counting the revolutions in a given time period, but that limited the tacho's low-speed calculating ability and the resolution of the calculations.

I briefly looked at that printer, and it looks a bit... unnecessary. It seems to be more of a CNC machine converted to printer than a purpose built printer. If you are into CNC as well, them by all means, go for it.

If you are just starting out I would recommend one of the less expensive models from PrintrBot or a RepRap Prusa, Huxley, or MendelMax.

[scooter2]

Apr 4, 2015 21:58:39 GMT -6 glavey said:

I have the sketch on the arduino counting the microseconds it takes per revolution and then calculating the rpm based off of that. At first I tried counting the revolutions in a given time period, but that limited the tacho's low-speed calculating ability and the resolution of the calculations.

I briefly looked at that printer, and it looks a bit... unnecessary. It seems to be more of a CNC machine converted to printer than a purpose built printer. If you are into CNC as well, them by all means, go for it.

If you are just starting out I would recommend one of the less expensive models from PrintrBot or a RepRap Prusa, Huxley, or MendelMax.

I like cncs and a machine that can do both would be excellent, even if the cnc side is weak at best and I think it has a larger than average print area. I found that one because I was looking to see which models could do nylon, but right now I'm just in research mode. The reprap looks good for people who want to build and modify their printers and the price is good too. I'd love to have an off the shelf one but part of me wants to learn how it all goes together and how to modify it if needed. I checked out the Printrbot too. I'll check out the Huxley and the MendelMax next.

The cheap tach I posted a picture of uses a wire wrapped around the plug wire a few turns. I wonder what kind of readings you get from that induction? I'm guessing you get a variance of voltage that might be used with an arduino if it's not too small.

[glavey]

As long as the extruder can go up to about 250°C, you can print with nylon. I don't think you even need a heated bed. I remember one person in the printrbot forums who discovered that putting canvas over the printbed instead of the usual kapton or blue painters tape helped the nylon stay put during printing.

I've also thought about using an inductive lead around the spark plug wire for the rpm signal, but it appears that getting the right voltage and current requires a lot of math regarding the length and diameter of the wire, how many loops around the spark plug wire, in which direction to loop, where along the wire to have the loops... - not my strong suit.

I have an oscilloscope so I can probe different potential signal sources and see which one is best suited to the job.

[scooter2]

Apr 5, 2015 9:15:48 GMT -6 glavey said:

As long as the extruder can go up to about 250°C, you can print with nylon. I don't think you even need a heated bed. I remember one person in the printrbot forums who discovered that putting canvas over the printbed instead of the usual kapton or blue painters tape helped the nylon stay put during printing.

I've also thought about using an inductive lead around the spark plug wire for the rpm signal, but it appears that getting the right voltage and current requires a lot of math regarding the length and diameter of the wire, how many loops around the spark plug wire, in which direction to loop, where along the wire to have the loops... - not my strong suit.

I have an oscilloscope so I can probe different potential signal sources and see which one is best suited to the job.

Have you taken readings off a piece of wire wrapped a few turns around the plug wire? I'd like to know what you get. These cheap tachs don't seem to mind whether you use four turns or more. I use five turns. Maybe they are able to read the number of pulses instead of a voltage.

Canvas is a good idea. I saw someone using pc perf board too.

[glavey]

The tiny tachs and their Chinese variants go by the pulses. I'll try wrapping some wire around the spark plug lead, starting with a thin wire and very few turns, increasing both as I go.

It turns out that once you have all of the supporting code written, changing the tachometer sketch from using a servo to using a stepper motor is quite easy. Here is a video of the end result. The motor with the flywheel isn't spinning too fast because I had to power it with a nearly dead 9v battery.

As I said in a previous post, I think it is on the previous page, I want to monitor the battery voltage, oil temperature, and my speed as well. For that, I have a 16x2 white-on-blue character display (like this one), and an 8 digit, 7-segment red led display (like this one).

[scooter2]

Apr 5, 2015 12:06:10 GMT -6 glavey said:

The tiny tachs and their Chinese variants go by the pulses. I'll try wrapping some wire around the spark plug lead, starting with a thin wire and very few turns, increasing both as I go.

It turns out that once you have all of the supporting code written, changing the tachometer sketch from using a servo to using a stepper motor is quite easy. Here is a video of the end result. The motor with the flywheel isn't spinning too fast because I had to power it with a nearly dead 9v battery.

As I said in a previous post, I think it is on the previous page, I want to monitor the battery voltage, oil temperature, and my speed as well. For that, I have a 16x2 white-on-blue character display (like this one), and an 8 digit, 7-segment red led display (like this one).

That sounds like a fun project. If you need more pins you can probably stack your Arduinos, letting one work with sensors, feeding output to another Arduino to run the display, etc. I have not tried that yet, but it seems like a great way to increase their usefulness for bigger projects. 

[glavey]

I've actually thought ahead for once and bought 2x MCP23008 (for more digital I/O), 2x MCP3008 (for more analog input), and 3x 74HC595 (for more digital outputs). With those, I could add 16 more digital I/O pins, 16 more analog input pins, and 24 more digital output pins.

I do also have more arduinos, but those are mainly for if (when) I make a mistake and release all of the magic smoke from one of them.

[scooter2]

Apr 5, 2015 15:05:12 GMT -6 glavey said:

I've actually thought ahead for once and bought 2x MCP23008 (for more digital I/O), 2x MCP3008 (for more analog input), and 3x 74HC595 (for more digital outputs). With those, I could add 16 more digital I/O pins, 16 more analog input pins, and 24 more digital output pins.

I do also have more arduinos, but those are mainly for if (when) I make a mistake and release all of the magic smoke from one of them.

Those are handy. Thanks for the tip!

[glavey]

Sorry for the lack of updates, I've been obsessing over my latest new toy, a mini laser engraver. I created a thread with an overview and review of it here.

I took the scooter out for a ride today and finally got a chance to completely open up the engine after installing a new variator last fall. Top speed is 43 mph as measured with the bicycle speedometer at 8160 rpm, measured by the tiny tach. Oil temperature got up to 78C (172F) after the top speed run and settled down to about 70C-75C (158F to 167F) while riding at 25mph. I haven't looked at the variator to see if the belt is moving all the way up the ramp, neither have I tried any different weights.

[cyborg]

Not bad for the first high speed pass