robertbu

This is a very cool project! I'd like to suggest some alternate pathways for the electronics. Using a transistor and base resistor you can get rid of your NE555 circuit and drive the PWM directly from the Arduino. Using a real time clock module, you can get rid of the clock kit and drive the clock directly through the Arduino also. Driving the clock through the Arduino would give you the ability to have your model track the earth/sun in real time. Adding a display for month/day (driven from the Arduino) would allow you to display the month/day/time represented by the model when you manually change the rotations of the two stepper motors. Note most 28BYJ-48 stepper motors are geared so that 2037.886 steps equals one rotation.

Your project is done, your circuit works, and this is a well written instructable. So I'm *not* suggesting you make any changes. My suggestions are only for people to consider who want to follow your instructable. Using a transistor and the Arduino's PWM, you can still use 12V and not draw the power for the LEDs from the Arduino voltage regulator. If you bought your motor from Aliexpress, then the 2037.886 steps per revolution is very likely, but I doubt it will make any difference since your small model does not require precise positioning. A bigger source of problems is the crystal in the arduino itself (which is not precise). You may be losing/gaining hours per month on your timings. But even that is not an issue with a small model. If it were me, I would have powered everyt...

Your project is done, your circuit works, and this is a well written instructable. So I'm *not* suggesting you make any changes. My suggestions are only for people to consider who want to follow your instructable. Using a transistor and the Arduino's PWM, you can still use 12V and not draw the power for the LEDs from the Arduino voltage regulator. If you bought your motor from Aliexpress, then the 2037.886 steps per revolution is very likely, but I doubt it will make any difference since your small model does not require precise positioning. A bigger source of problems is the crystal in the arduino itself (which is not precise). You may be losing/gaining hours per month on your timings. But even that is not an issue with a small model. If it were me, I would have powered everything from a 1A 5V USB charger. This gets rid of your 7805 circuit.

He is running the circuit on 12V DC, so the circuit is already DC. Note with a few changes, many of the components can be removed and the whole thing can be driven at 5V using a 1A USB wall brick.

This would be an excellent entry for the "Colors of the Rainbow" contest for the "Judges Prize for the best overall use of all of the colors of the rainbow."

Thank you very much for providing this resource! I've beencollecting ideas for a future ROV build for some time, and there are a numberof things in your build that are new to me.+Rather than holes in the PVC pipe, I'm considering fillingthe pipe with Silicone. The Siliconedensity is close to the density of water.+It is unclear how you ran your wires into the drychamber. If you ran the sheathing in aswell as the wires, you can have problems. If the sheathing around the Cat6 gets nicked, water can run into yourdry chamber in the sheath. +In some builds, the tether is given a slightly positive buoyancy. The whole tether is placed in a light meshbag. Small floats are added to the baguntil reaching a slightly positive buoyancy. The floats are then attached along the tether. This way the...

Thank you very much for providing this resource! I've beencollecting ideas for a future ROV build for some time, and there are a numberof things in your build that are new to me.+Rather than holes in the PVC pipe, I'm considering fillingthe pipe with Silicone. The Siliconedensity is close to the density of water.+It is unclear how you ran your wires into the drychamber. If you ran the sheathing in aswell as the wires, you can have problems. If the sheathing around the Cat6 gets nicked, water can run into yourdry chamber in the sheath. +In some builds, the tether is given a slightly positive buoyancy. The whole tether is placed in a light meshbag. Small floats are added to the baguntil reaching a slightly positive buoyancy. The floats are then attached along the tether. This way the tether does pool on thelake floor where it is more likely to be caught by something.+My ROV will be for saltwater, so I cannot get away with a wetbattery.

"Why not just let it flood with water?" - I may be over thinking things, but I figure stuff in the water will get into the pipes...silt, small aquatic creatures, salt, etc. The light mesh bag is only used during the construction phase. The tether is put in the bag and placed in the water, float are added to the bag over time until the right buoyancy is reached. It is a simple way to figure out how many floats to add. As for the floats, I plan to 3D print some tapered, hollowish forms that clip or thread onto the tether...the shape of long lead sinkers. I'd fill the 3D shapes with foam. Interesting though on the battery test.

I assume your battery is 12V. Do you have a general idea of the run time of your ROV?

Silicone has roughly the same density as water. If the pipes were filled with silicone, then you would not have to deal with bleeding the air going down and drying out the rig at the end of a session.

In rough water, the solar panels will be bounced around and continuously splashed by (potentially salty) water. Any concerns? Did you do anything special with the panels to mitigate this potential issue?

In rough water, the solar panels will be bounced around and continuously splashed by (potentially salty) water. Any concerns? Did you do anything special with the panels to mitigate this potential issue?

The RDX pin is only rated for 3.3v. Using 5.5v can be harmful for the HC-06. I always add a voltage divider to bring the voltage down to 3.3v for this one pin.

Can the receiver be bound to any transmitter, or do I have to purchase an F949 plane to get the transmitter?

I'm a newbie with respect to Arduino programming, but from my reading, there are couple of changes to your code you should consider:1) I believe any variable that changes value inside an interrupt service routine should be volatile, so randNum, randNum2 and randNum3 should be defined as volatile.2) There are issues with longer interrupt service routines including lost time on the millis() call. A better practice would be to just set a flag in the service routine and then call Blink() in the loop() function based on the flag. With this structure, only the flag would need to be volatile.

Creative solution! Thanks for sharing.

Any thought about over-voltage or polarity protection? Also your readings will be relative to theinput voltage which is not necessarily 5.0v. You can get around thisproblem (with some loss of precision) by using Arduino’s built-in voltagereference

The wrapping guns and bits (as well as hand wrapping tools) are available on ebay, but they are $$$. I use empty ink cylinders from ballpoint pens as well as brass tubing to make my wrapping tools.

Nice, compact idea. You could "double" your light for "free" by using two leds and dropping your resistor to 120 ohms. It would require a slight change in your enclosure design.

Cool looking ROV. Iassume you have not yet had it in the water. Some thoughts looking at your design:+ You mention a wireless camera. Wireless signals do not travel well inwater. If your signal is above 1 GHz (asI suspect with the Google Glasses clone), then the signal will only travel afew inches into the water. You may needto either add wire to take the wireless signal to the wireless camera, or gowith a wired camera and bring wires to the surface.+ Based on the photographs, you will need to cut holes in thecowling in front of the propellers to allow water to flow in. + It appears that you are bringing wires in a sheath directlyinto the sub. When bringing in wires ina sheath, precautions must be taken to avoid water getting to the electronics. If the sheath is nicked (a high pr...

Cool looking ROV. Iassume you have not yet had it in the water. Some thoughts looking at your design:+ You mention a wireless camera. Wireless signals do not travel well inwater. If your signal is above 1 GHz (asI suspect with the Google Glasses clone), then the signal will only travel afew inches into the water. You may needto either add wire to take the wireless signal to the wireless camera, or gowith a wired camera and bring wires to the surface.+ Based on the photographs, you will need to cut holes in thecowling in front of the propellers to allow water to flow in. + It appears that you are bringing wires in a sheath directlyinto the sub. When bringing in wires ina sheath, precautions must be taken to avoid water getting to the electronics. If the sheath is nicked (a high probabilityin a rough marine environment), water will travel inside the sheath and intothe electronic boxes. Solutions can befound on the net to avoid this problem.+ You likely will need to make the Cat5 cable buoyancy neutral to avoid it pulling on the sub.+ Inexpensive bilge pump motors usually draw around 3A with theoriginal impeller and likely are designed to handle up to 14.8V (the chargingcurrent of a lead acid battery). Withthe impeller replaced by a propeller the current draw of the bilge motors jumpsto 6A or more, and LiPO 4S batteries will be 16.8V when fully charged. Not sure if any of this will impact you.

Cool looking ROV. Iassume you have not yet had it in the water. Some thoughts looking at your design:+ You mention a wireless camera. Wireless signals do not travel well inwater. If your signal is above 1 GHz (asI suspect with the Google Glasses clone), then the signal will only travel afew inches into the water. You may needto either add wire to take the wireless signal to the wireless camera, or gowith a wired camera and bring wires to the surface.+ Based on the photographs, you will need to cut holes inthe cowling in front of the propellers to allow water to flow in. + It appears that you are bringing wires in a sheathdirectly into the sub. When bringing inwires in a sheath, precautions must be taken to avoid water getting to the electronics. If the sheath is nicked (a high pr...

Cool looking ROV. Iassume you have not yet had it in the water. Some thoughts looking at your design:+ You mention a wireless camera. Wireless signals do not travel well inwater. If your signal is above 1 GHz (asI suspect with the Google Glasses clone), then the signal will only travel afew inches into the water. You may needto either add wire to take the wireless signal to the wireless camera, or gowith a wired camera and bring wires to the surface.+ Based on the photographs, you will need to cut holes inthe cowling in front of the propellers to allow water to flow in. + It appears that you are bringing wires in a sheathdirectly into the sub. When bringing inwires in a sheath, precautions must be taken to avoid water getting to the electronics. If the sheath is nicked (a high probabilityin a rough marine environment), water will travel inside the sheath and intothe electronic boxes. Solutions can befound on the net to avoid this problem.+ You likely will need to make the Cat5 cable neutral buoyancyto avoid it pulling on the sub.+ The bilge pump motors usually draw around 3A with theoriginal impeller and likely are designed to handle up to 14.8V (the chargingcurrent of a lead acid battery). Withthe impeller replaced by a propeller the current draw of the bilge motors jumpsto 6A or more, and LiPO 4S batteries will be 16.8V when fully charged. Not sure if any of this will impact you.

I know this is just a breadboard, proof-of-concept Instructable, but does the ADS1015 buy you anything? The arduino analog pins have 10-bit precision, so you can read down to around 0.05V and 0.06A. And you can increase the voltage precision by changing the voltage divider.

Your Ohm's law calculation does not take into account the resistance of the LED, nor the voltage drop of your transistors nor the voltage drop associated with the pin transistor. And each layer is only lit 1/5 of the time, so the average current is much smaller than the intermittent current. If your LEDs are bright enough, you are golden. I only wanted to point out that adding a 330 ohm resistor is not a "free" solution to solve the problem of too much current. The LEDs pay a price in terms of brightness.

He is only turning on one led per pin at a time, so he is only drawing 20ma per pin. The problem is that there is a 200ma limit on the total current pulled by all pins output pins. If the code lights an entire layer, the board is pulling 400ma..twice the recommended limit. Note that the amount of current that can be pulled from the 5v pin depends on the voltage regulator used on the board and varies between 0.5A and 1.5A. But that is power does not go through the output pins. And from my reading it is more complicated than just a 200ma overall limit. There also appears to be a 150ma limit on certain sets of pins.

The use of a 330 ohm resistor on each output line does solve the current draw problem, but at the cost of brightness. The result is running an LED at 6ma that typically draws 20ma. And there is further brightness loss due to the multiplexing.

I think you are drawing too much current through the Arduino. Assuming you are multiplexing each layer and each led draws 20ma, you are drawing 350ma through the arduino board. The max from any datasheet I've read is 200ma. Or am I missing something about your design?

This is a very cool project. Building something small enough for a keychain and useful using anArduino can be difficult, and you pulled it off. Some thoughts (partly for you and partly for anyone considering building one of these):+ I agree with your stance on security. In a perfect world, everything would havestrong passwords, but most people use weak passwords they can remember and thesame passwords on many sites. Strongpasswords coupled with your PassType is a dramatically better system. Someone would need to target a personspecifically and in the real world with PassType.+ Your joystick gives you an easy way to enter a fastpassword for entry into the PassType. You have four states and using five ‘digits’, gives you 1K of possiblepasswords. Coupled with increased tim...

This is a very cool project. Building something small enough for a keychain and useful using anArduino can be difficult, and you pulled it off. Some thoughts (partly for you and partly for anyone considering building one of these):+ I agree with your stance on security. In a perfect world, everything would havestrong passwords, but most people use weak passwords they can remember and thesame passwords on many sites. Strongpasswords coupled with your PassType is a dramatically better system. Someone would need to target a personspecifically and in the real world with PassType.+ Your joystick gives you an easy way to enter a fastpassword for entry into the PassType. You have four states and using five ‘digits’, gives you 1K of possiblepasswords. Coupled with increased time between attempts based on number of failed attempts (stored in EEPROM), and a lockout or erasureafter some number attempts, your device would have reasonable, basicsecurity. And muscle memory would make the five ‘digits’ quick and automatic.+ If you were to create a male USB-C to male USB-A adaptorthat plugs into the USB-C port of the Arduino Pro Micro, then you can easilychoose between the two ports based on the computer you are using. Type A can be a real pain on a desktop (unlessthere is a male to female cable around like in your pictures).+ Using INPUT_PULLUP and five pins for your 5-way thumbwould eliminate the need for resistors.+ You could use the EEPROM on the Arduino Pro Micro ratherthan storing the passwords on the 24LC256. It would only support 25% of the passwords that the AT24C256 would, butthat is still more than enough for my password list.+ The AT24C256 comes in module form forthe Arduino. It would be harder to meetthe size profile, but easier to build without etching a board.+ The 24LC256/AT24C256 has three address pins, so up to 8can be used in the same system.+ With a combination of some of the above changes, aPassType could be built with just some wires and solder. No etched board or even prototype board wouldbe needed. The components would be reduced to 1) Micro Pro, 2) OLED display, and 3) thumb joystick.

Your delay(4) occurs when the segments are dark. That is why the display got brighter when you change to delay(1). Try moving the delay() below where you light the segments. Add a second delay() just after you clear the segments. Start with a value of 1 for the first delay and 4 for the second and then adjust from there.

edgary777 wrote: I'm not exactly an expert with electronics so I'm not sure if I should have used resistors, browsing some forums I read that you probably should, but I didn't and nothing has burned or exploded so far.I too am fairly new to electronics, but I've spent some time learning about and building multiplex displays. Each segment of your red seven segment display will draw it rated current at around 2.0V. You are driving them with 5.0V. What is saving you is that each segment is only on for 1/4 of the time. If you were to continuously display a single segment at 5.0V, it would burn out. You are probably safe since you are pulling "only" a little over twice the current through each lit segment 1/4 of the time, but you would need the datasheet to be sure. Some leds...

edgary777 wrote: I'm not exactly an expert with electronics so I'm not sure if I should have used resistors, browsing some forums I read that you probably should, but I didn't and nothing has burned or exploded so far.I too am fairly new to electronics, but I've spent some time learning about and building multiplex displays. Each segment of your red seven segment display will draw it rated current at around 2.0V. You are driving them with 5.0V. What is saving you is that each segment is only on for 1/4 of the time. If you were to continuously display a single segment at 5.0V, it would burn out. You are probably safe since you are pulling "only" a little over twice the current through each lit segment 1/4 of the time, but you would need the datasheet to be sure. Some leds have a lot of leeway...5x to 10x. Others not so much. From my reading, one way to reduce the impact would be to increase the frequency. The simplest way for this project would be to drop your delay in your code from 4ms to 1ms. Another way would be to introduce blank "frames". That is, you have four frames, where each frame is lit 1/4 of the time. You would insert between these frames times where no segments are displayed. Of course, the segments will dim a bit.Thanks for the instructable.

I cannot figure out how you make turns. Does the back fan rotate, or the back fins rotate, or ???

What a cool idea. Your instructable says use the 26 gage opening of the wire cutter to cut the tubing, but your picture shows 16 gage opening. I think you want 16.

Thank you for this Instructable. I ran across it just after Christmas and started 16/8 fasting the first of the year. I'm down 7+ pounds so far. Also I ran across an article on BBC news: 'Fasting diet 'regenerates diabetic pancreas'. It was mice tests using 5 days of restricted diets followed by 25 days of normal feedings. So there is a lot of research yet to be done on intermittent fasting.

Looks cool. It would be safest if the side that plugs into the wall is *not* the side that connects to the middle. I believe this is opposite to how you have it wired, but I cannot be sure. That way when the switch is up out of the block, the copper will not be hot.