JonMackey

Re setting fuses: Correct, not in SD mode. In USB pass through mode fuses can be set. In pass through mode the loader acts like a generic ISP. lockbytes and fusebytes for almost all AVR boards are the same. I don't even think avrdude supplies these values because they aren't really needed.At some point I plan to add support for setting fuses and for installing bootloaders. The Mac OS utility app can easily be modified to get the fuse values and the bootloader hex file associated with any sketch. I would just need to figure out how to expose this feature in the loader UI. It's not a priority for me. The loader does what I need it to do.

I think this is getting a little too detailed for the comments section. Please send me a private message if you want to discuss this further. Thanks.

The objective-C application GUI would have to be rewritten for Windows OS. Most of the classes that parse the configuration and elf files use C++ Standard Library and would work on Windows with minimal changes, if any. The sources in the GUI that instantiate these classes are written in objective-C and these would have to be rewritten. It's not a small amount of work. It's been about 5 years since I've written a Windows application, and I no longer have a machine with an up to date Windows OS.Assuming you were to use my board, you could extract and create the configuration information manually. It's only a key/value text file (unix line endings.)byte_count - optional, an unsigned integer that is the size of the binary. This only effects the progress indication so you could enter a d…

The objective-C application GUI would have to be rewritten for Windows OS. Most of the classes that parse the configuration and elf files use C++ Standard Library and would work on Windows with minimal changes, if any. The sources in the GUI that instantiate these classes are written in objective-C and these would have to be rewritten. It's not a small amount of work. It's been about 5 years since I've written a Windows application, and I no longer have a machine with an up to date Windows OS.Assuming you were to use my board, you could extract and create the configuration information manually. It's only a key/value text file (unix line endings.)byte_count - optional, an unsigned integer that is the size of the binary. This only effects the progress indication so you could enter a dummy value or not provide it.desc - required, name that appears in the loader to identify the target. This is from the boards.txt config file of the target board, "build.mcu". It's only for display, so it could be anything.f_cpu - required, the frequency of the target. This is from the boards.txt file.flash.page_size - required, an unsigned short representing the flash page size of the target. This is extracted from avrdude.conf.signature - required, an unsigned integer representing the 3 byte signature of the target. You can get this from the documentation for the mcu or from avrdude.conf.upload.speed - required, an unsigned integer of the baud rate for the target bootloader. This is extracted from boards.txt. If the target doesn't have a bootloader and/or you want to load using the ICSP connector, set this value to zero.There are other key/values but the above are the most critical.Example: for an ATmega644 running at 8MHzbyte_count=18772desc=ATmega644Pf_cpu=8000000flash.page_size=256signature=0x1e960astk500_devcode=0x82upload.speed=57600To get the actual binary hex file, the Arduino IDE has an "Export Compiled Binary" menu command. The binary and the config file must have the same prefix. Example, DCF2Tool.ino.hex DCF2Tool.ino.txt

Sorry, no. I only work on the Mac.

I'm glad you like it. The spring is hard to find and they aren't inexpensive. I found that the spring in the shampoo bottle I use is the same diameter as the one I specified, but it's a bit long so I cut it in half.

The maximum number of data points depends on the size of the EEPROM. I'm currently using a 32KB chip. This chip also stores the hike locations so there's only 31KB available for the logs. For each hike, the log header + trailer takes 60 bytes and each data point is 6 bytes (temp and pressure.) If no other log from a previous hike is being stored, the maximum number of data points would be (31744-60)/6 = 5280. The sample period is currently set at ~4 seconds. When the log is full the last entry is constantly overwritten till the hike ends.

There was a possibility that the BMP280 remote would also be used in a different project that needed greater range. Other factors: I also had several RFM69CW on hand, and I already worked with it on other projects. The sleep current is good enough, around 0.1uA. Something is chewing up battery on all of the boards and I suspect the DC-DC converter. Maybe I don't have it implemented correctly. The BMP280 remote lasts about a month on a charge, but it should be longer based on the specs. I've gone through all of the MCU pins to make sure they aren't back feeding during sleep (pulled up when they should be pulled down, etc..) I don't have a meter that can accurately measure uA.The RTC on the gateway is a just a clock crystal. It's accurate down to about 10F and then it drifts a bit. …

There was a possibility that the BMP280 remote would also be used in a different project that needed greater range. Other factors: I also had several RFM69CW on hand, and I already worked with it on other projects. The sleep current is good enough, around 0.1uA. Something is chewing up battery on all of the boards and I suspect the DC-DC converter. Maybe I don't have it implemented correctly. The BMP280 remote lasts about a month on a charge, but it should be longer based on the specs. I've gone through all of the MCU pins to make sure they aren't back feeding during sleep (pulled up when they should be pulled down, etc..) I don't have a meter that can accurately measure uA.The RTC on the gateway is a just a clock crystal. It's accurate down to about 10F and then it drifts a bit. I'm changing the design to use a temperature compensated crystal oscillator (the ubiquitous DS3231.)Re ton of work: most of the time went into the font classes and the app that generates the compressed bitmaps.

You're welcome.

Nope, no risk at all. I thought about that too. The most it can pass through to the 3v3 regulator is 12V and that is within its rating. You just don't want the motor attached before you adjust it.

Is this an academic question or are you seriously considering developing something yourself? Yes this circuit could be adapted for an LED/photodiode by removing the DC-DC module, flyback diode, and motor control MOSFET, adding a few jumpers and an extra wire to the drum. I wouldn’t recommend it though, dust clings to everything on the inside regardless of when you start monitoring (even if it’s well grounded.) One addition I am considering is adding a digital magnehelic to determine when the filter is overloaded (and send a different audio alert to my hearing protector.) When using a wide belt sander the filter loads way before the drum is full. I figure it it could be built using an inexpensive digital barometer module.

Yes, the interface powers the caliper. By design you should not install the caliper battery when using the interface otherwise you can't zero the caliper programmatically.

I'm glad you liked the design.I just added a new version of the board. In this latest version I added a usb serial interface, resettable poly fuses and LED function indicators below the function selection buttons.

Glad you liked it. I've been through the process of installing spiral pipe twice. Once when I was living in Massachusetts, and again here in New Hampshire. I was able to reuse about 90% of the pipe and fittings.

Let me know if there's anything specific you would like me to clarify.If you'd rather not discuss this publicly, send me a message.

jon@mackey.com

Happy New Year to you!As you were confused about the purpose of my Varmint Detector instructable, I was equally confused about the point of yours. I think some of the confusion is due to the translation to English. The first two times I looked at your project I didn’t make it past the first few paragraphs (like you with mine, I just lost interest.) I kept coming back to it because I’ve designed several boards that use the ATtiny84A and 85.Things I found confusing:- why you would use an ATtiny84A if you needed so many I/O pins? Why not use a different mcu such as the common ATmega328p? An ATmega328p doesn’t cost much more and if you really didn’t care about timing accuracy you could even use the internal resonator and eliminate the need for an external crystal.- how do you download co…

Happy New Year to you!As you were confused about the purpose of my Varmint Detector instructable, I was equally confused about the point of yours. I think some of the confusion is due to the translation to English. The first two times I looked at your project I didn’t make it past the first few paragraphs (like you with mine, I just lost interest.) I kept coming back to it because I’ve designed several boards that use the ATtiny84A and 85.Things I found confusing:- why you would use an ATtiny84A if you needed so many I/O pins? Why not use a different mcu such as the common ATmega328p? An ATmega328p doesn’t cost much more and if you really didn’t care about timing accuracy you could even use the internal resonator and eliminate the need for an external crystal.- how do you download code onto your board? I didn’t see something that looked like an ICSP connector.- why use an 74HC595 when you’re essentially implementing a bus expander? (or is that your point?) A bus expander such as the common I2C controlled PCF7584 would be easy to add, only taking two pins on the ATtiny. You can hang several of these off of the I2C bus and they even make a 16bit version which is handy for creating keypads. Both versions have an interrupt line so the ATtiny doesn’t have to poll to see if something has changed. (and they make DIP versions of these chips too)- I also noticed that you don’t consistently use decoupling capacitors. Is that because the PCB traces you use are so wide you don’t need them? I really like the ATtiny84A though. If you have never played with sleep and the other interrupts you should check them out. They’re quite useful for battery powered devices. For example I use pin change interrupts to implement the remote control for the Varmint Detector. The ATtiny only wakes up when someone presses a button. You can debounce the buttons in software pretty easily, no extra hardware needed (you don’t even need pull up resistors.)Off topic: How did you print the cover of your project?, it looks really professional.

I added a one line description of what this project is close to the top of the description. This is in the "Technology" category, so I left in the details in the description of what chips were used and most of how the detector reacts to motion.If you feel I still need to pare it down more, you'll need to give me some pointers.You have a happy 2019 too.