03-21-2019, 09:19 AM
I finally have the pH circuit ready and will be shipping them this weekend, thanks to everyone who ordered.
I'm quite happy with how it turned out, it's very accurate according to the buffers and runs reliably. I can do a 3 point calibration using 6.86, 4.01 and 9.18, after calibration I get those numbers down to the 1000th's. One of the calibrations all 3 were absolutely perfect as in 6.860, etc... always within +/- .003.
Now that I've been playing a lot with pH I didn't realize how stable the readings are in 1000th's so the next update, coming in a few days, will have it displayed as 8.035 but I'm going to leave it to 100th's for rules and alerts.
I still have a bit of work to do on the library but that should be ready before the circuits arrive, this is only needed if you want to run it without Robo-Tank.
So with all the good news there is one minor issue that surprised me. About a week ago I happened to have the I2C address set to 99 and connected it to the controller and to my surprise it didn't work. Later I discovered there was an issue with the ATtiny85 wire library and the Due, only address 0-64 will work, 65-127 doesn't work. Using a Mega 2560, Uno or any other AVR all addresses work fine. I've spent nearly the last week trying everything I could to get all working but couldn't do it without writing my own Wire library which is currently above my pay grade and I have no time to figure it out.
In the big picture this is meaningless unless you have 64 devices connected to your project and they all use the lower end of the addresses. I'm not using any of these address on the controller so it's still possible to connect 64 circuits which I'll obviously never be supporting anyways.
Well you all should be getting an email in a few days confirming it was shipped.
I'm quite happy with how it turned out, it's very accurate according to the buffers and runs reliably. I can do a 3 point calibration using 6.86, 4.01 and 9.18, after calibration I get those numbers down to the 1000th's. One of the calibrations all 3 were absolutely perfect as in 6.860, etc... always within +/- .003.
Now that I've been playing a lot with pH I didn't realize how stable the readings are in 1000th's so the next update, coming in a few days, will have it displayed as 8.035 but I'm going to leave it to 100th's for rules and alerts.
I still have a bit of work to do on the library but that should be ready before the circuits arrive, this is only needed if you want to run it without Robo-Tank.
So with all the good news there is one minor issue that surprised me. About a week ago I happened to have the I2C address set to 99 and connected it to the controller and to my surprise it didn't work. Later I discovered there was an issue with the ATtiny85 wire library and the Due, only address 0-64 will work, 65-127 doesn't work. Using a Mega 2560, Uno or any other AVR all addresses work fine. I've spent nearly the last week trying everything I could to get all working but couldn't do it without writing my own Wire library which is currently above my pay grade and I have no time to figure it out.
In the big picture this is meaningless unless you have 64 devices connected to your project and they all use the lower end of the addresses. I'm not using any of these address on the controller so it's still possible to connect 64 circuits which I'll obviously never be supporting anyways.
Well you all should be getting an email in a few days confirming it was shipped.
