Thanks for the feedback, and your continued interest in this project!!
I am currently running the same firmware, v1.2, so that is not the problem! When the firmware starts, it first blinks the LED, then beeps the beeper, then initializes the GLCD display. It sounds/looks like all that is happening successfully. The next thing that happens is a routine called 'InitTimeDisplay' is run, where the outside border on the GLCD is drawn, and the time separator colons are drawn. The very next thing that occurs is a call to 'Read_DS3231SN_Time()', followed by the display of the time data. My guess is that the firmware is getting hung up in this routine. The buttons are 'polled', not interrupt driven, so if they are unresponsive, the code is not executing the main loop where time/temperature is continuously read, and then displayed.
It's not a battery issue, as I took the battery out of my clock, and restarted the clock. The RTC continues to function without the battery, it just starts at 00:00:00 and 01/01/00, for the time and date.
My guess is that the problem lies with the DS3231, or some of its support components. I would double-check to be sure that each leg of the RTC is soldered to the board. I would also check the PIC pics for SDA and SCL. Finally, I would check for the presence of R2, and R3, the I2C pullup resistors. They should be about 3K.
Oh, one other thing. Sorry about the mongrel ICSP pinout! I'm not sure where I got that pinout from, but I've been using it for years on all my designs, and I don't dare change it now!
Lots of reasons. I did one pre-beta prototype by hand, and the whole board took about an hour to solder except for the 0.5mm connector which took about a day to get right, and three connectors. Eventually, this is going to be a product, so I'm going to need to find a decent assembly house then anyway. Besides, my expertise is design and coding, not soldering, and I like to stick to what I'm good at.....
I'm looking for suggestions for an assembly house that will do "prototype/beta quantities" (about 25 pieces) at a "reasonable" price. This is for an all surface mount project (about 30 components total) that is actually quite easy to hand solder except for a fiddly (0.5mm pitch) GSM modem connector. I'd be happy to just send Gerbers and a BOM, or I can provide the boards and parts. At this point, price is the critical driver. Location is not terribly critical either. I'm in the US, so that works, but Asia and Europe are also fine.
I've emailed Seeed and iTead several times, but neither has replied to my email, so I gather they are not interested.
Here is a photo of the completed GPS Display Time Standard I intend to use for setting a variety of clock projects I've made, or am working on. This GPS module, the GlobalSat EM-406, has excellent sensitivity as I've been able to lock onto the necessary satellites inside my home without any difficulty!
I'm currently working on a couple of clock projects, and I wanted to have a reasonably accurate time standard to set them, so I put together this quick project, a GPS Display/Time Standard.
[attachment=0]
This board uses a GlobalSat EM-406 GPS engine, a PIC 16F689 MCU, and an 8x2 LCD display. The board is intended to be powered externally for stand-alone operation, or by the attached clock. Time data is output to the clock via a 4 pin MTA connector. The firmware will display the number of "locked" satellites on power up, and then revert to time display mode as soon as a valid position/time is received. Time display can be in 12H or 24H format, and a programmable UTC offset can be set to display "local" time if desired.
This project is not intended for ultimate time setting accuracy, but should deliver sub-second (< 1 second) accuracy, which is good enough for my needs.
I'm sorry for the lack of updates on this project. I changed the PIC on this board to an 18F2550 shortly after my last post, but then ran out of time to get the board checked and fabricated before leaving on an extended trip to Asia (I'm in Seoul right now). I'll be back home at the end of the month, and I'm looking forward to rolling up my sleeves, getting back to the bench, and finishing a bunch of stalled projects. This one is at the top of the list!
I use X-10 a lot in my home automation system. I've been using it for years, and frankly it's not the most reliable system in the world. It's primary benefit is that it uses existing wiring in your home. If you can live with an occasional light that doesn't come on once in a while, or one that doesn't turn off, it's great!!
The tricky part of the X-10 system is the actual interface to the AC mains. Lethal voltages exist, so I don't recommend[/u] messing around here. Fortunately, the folks at X-10 make a module called the TW-523 Powerline Interface, which handles all the interfacing to the powerline. This module is a X-10 to logic interface that plugs into a wall outlet, and has a modular telephone connector for I/O. The connector has GND, Rx (from the powerline), Tx (to the powerline), and Zero (a zero crossing detect output). It's relatively straightforward to encode and decode X-10 formatted data with a microcontroller once you've got these nice, handy signals!
Are the Eagle schematic and board files available for the rev. 2 design? I've had one of these boards for a while, but I haven't built it yet as I could not find this documentation. I like to have this info to generate a parts list, and for parts placement during assembly.
Yeah, isn't that the truth! There is some pretty wild stuff available from the old Soviet Union, and it's fun to work with! Lately, Nixie tubes seem to have had a new resurgence in popularity with the 'retro' crowd as their prices have been climbing steadily. My favorite Nixie tube is the IN-18, a nice, big Nixie, but the price of those bad boys is now up to about $60 USD per tube!!
I have just about completed the first version of my open source Nixie Clock circuit, and I'll be posting the schematic in the next couple of days. In the meantime, I thought I'd show off the contents of a package I received from the Ukraine just before New Years containing my Nixie Tubes, neon bulbs (for colon separators), and high voltage driver IC's. With these items in hand, the pace of this project will surely speed up now!!
[attachment=0]
The Nixie Tubes are Russian IN-14 tubes that stand about 2" (50mm) tall. These tubes have a nice rubber base, and long solder leads for attachment to the PCB. The neon bulbs will be used as the 'colon separators" between groups of digit to create the familiar '12:59:59' clock display format. The K155ID1 high voltage driver IC's are BDC-to-Decimal display drivers with open collector outputs capable of withstanding the 180VDC required to fire the Nixie tubes. These IC's will make driving the tubes infinitely easier!
Thanks for the suggestions! I will definitely take a look at some of the more modern USB-capable PICs! I've used the 18F2550 in a number of projects already, so it's my "go to" device for USB applications. Doesn't mean that there aren't other, better, choices though.....
I'm planning on using the MAX1771 High Efficiency DC Step-Up controller in my design to generate the high voltage required for the Nixie tubes. I want the "sub-circuits" of this design to be as portable as possible, so that one is not locked into a specific processor. I think the dedicated HV power supply meets that goal without incurring a significant additional expense!
I also wanted to mention that I'm leaning toward a fully through-hole design for my Nixie Clock in order to allow as many people as possible to participate!
Several weeks ago I won one of the three Cadsoft Eagle "Hobbyist" licenses given away by DP! Thanks to DP, and especially to Cadsoft, for their support and generosity!
I won the license based on my desire to design an Open Source Nixie Clock, and I plan to share that design here in the DP "Project Logs" forum! I've been working on this design for a couple of months, but was forced to put it on the shelf due to the board size limitations imposed by the "free" version of Eagle. With the free version, my "tube board" design would only allow space for 3 of the 6 tubes I intended to use!
Over the next week or so I'll be posting the preliminary schematics, parts lists and board files for the design. I'll also be discussing the overall structure and design of the firmware. My intention is to make an initial prototype using the 'DorkbotPDX' PCB service, and then have the final boards make at Seeed. I'll probably get extra boards, so anyone that needs one will not have to order their own, but can get one from me. The firmware for this project will be written in 'C' using the CCS 'C' compiler, although the code should be fairly transportable to other compilers. I do plan to leverage a lot of the hardware design, and firmware from my recent GLCD clock project covered here earlier.
My design will be based on the 'IN-14' Nixie tube. These tubes are an excellent size for clocks, and can be readily purchased from sources on eBay for about $7 USD per tube. The design will feature six of these tubes, and two neon "colon" digit separators. The design will feature separate 'clock' and 'tube' boards, so other tubes can easily be supported!
Nixie Clock Design Features
(6) IN-14 Nixie Tubes 18F2550 Microcontroller with USB +12V DC Input (Wall Wart) Maxim DS-3234 RTC with TCXO Battery Backup Time Setting via USB or On-board Switches Power Down Mode (Increase Tube Longevity) Anti-Cathode Poisoning Mode (Increase Tube Longevity)
I've designed a hardwood base for Nixie Clock I've got in mind, and I've selected an acrylic display case to go over it. Once complete, this clock should have a very 'professional' look and should be worthy of display in a prominent location in your home!
If anyone else would like to help, or has some other ideas for design features, please feel free to chime in!
The T-mobile "pay-as-you-go" cellular service is quite reasonable for this type of application. There is an initial "start-up" fee of about $8USD to buy the SIM card "kit", and then a 3 month service plan costs $10USD. In my area, the US Northeast, I'll probably activate my service in November, and then get six months worth of service for $20.
I've made a lot of progress on my Cellular Remote Control project in the past few weeks, and I thought I'd share some more details. At this point, I've fully debugged the hardware, and have written the first fully functional version of the firmware. I've got a little bit of clean-up to do, and some features to add, and that will put the wraps on this effort!
Overall, the original hardware design was almost perfect! There were, however, several small issues. In one case, I named the input signal to a transistor the same name as the output! Oops!!! That fix resulted in one white wire on the prototype PCB. The other two white wires bring two signals, Command_Ready, and Network_Ready from the ADH8066 module over to the PIC. Originally I didn't think I'd need these signals, but it turns out that life is easier if I can monitor them! The multicolor wire is a temporary serial debugging port. Normally, I add a small header to my boards for this purpose, but I forgot in this case! Doh!!
The top cover LED's and switches are mounted to a small 'Front Panel' PWB. This board connects to the main board via a short piece of flexible ribbon cable. It's the first time I've tried something like that , and it worked out flawlessly!!
[attachment=0]
[attachment=1]
The firmware is relatively straight-forward, but it is kind of busy! I use an interrupt driven circular buffer to capture all data from the ADH8066 module at 115,200 baud. The code is constantly looking for an incoming SMS test message, polling the 'local' control switches, and reading the temp. sensor. There are a number of programmable 'features' such as 'Power Restore' notification, low/high limit temperature reporting, and auto relay shut-off after a user-selected period of time. The user can also 'register' a particular phone number, or assign a 'password' to prevent unauthorized operation.
I still need to add the RJ connector for the Relay Board, and actually test this with real relays, but I expect this will be more of a formality than anything. I also need to get the bootloader working for firmware updating!
This has been a really fun project, and one I've enjoyed a great deal! How cool is it to be able to text a "black box" somewhere, and have it respond to your commands!! I've shown this prototype to a couple of my flying buddies, and it looks like I'll be making at least a dozen or so of these for other airplane owners at my local airport!
