heh, my new design for a clock (see other topics) will pwn yours in the brightness region, 18:58:58 will be about 1000W in brightness, blind me and half the street! think i will make it remote controlled... yours is still blue though.... have to get some blue plexi infront of it!
Cheap, simple RTC modules can be found here (for all you people who live out in the heathens, like america ) http://www.sparkfun.com/shop/index.php?shop=1&cart=261285&cat=1&itemid=301& I'm working on a clock that uses seven segment VFD tubes (the nuxie's successor, very rare). They're pretty cool. I'm avoiding all of the programming issues with things like multiplexing by just using a bunch of shift registers and driving the display directly (well, with output transistors. THe tubes nees about 50V at a few microamps per segment. As well as heater voltage.) I'll stick up some pictures later. ANother possibility for timing is to use GPS reciever modules, whch typically have a 1 Hz output pin, based off the cesium clocks un the GPS sattelites. Plus, if you're feeling up to it, you can interface with them serially, and find out where you are.
Hi Wolfe, Yes, those VFD 7-segment display tubes are cool! I acquired some sets of VFD 7-segment tubes myself (10x end viewed IV-22 and 6x side viewed IV-11 (both 20mm digits) and today arrived 6x side viewed IV-6 (10mm digits) VFD tubes. I lost two auctions for those magnificent transparent DT-1705e VFD tubes (6 and 20 pieces each...)... Still looking for a decent schematic. Found this one which is quite a good start (basic schematics and PIC code). I ordered two UCN5810 10bit VFD drivers (now replaced by the 6810) from Farnell to do some tests. Please let me know if you find some other interesting schematics. BTW: My Nixie clocks accept the 1Hz GPS signal as timebase signal too... CD
Yes, i just mentioned GPS units as an easy to get EXTREMELY accurate 1 Hz signal. WHy bother with a driver IC? just use any cheap NPN transistor, and 2 resistors. (a base resistor, and a pullup. THat way, when the logic signal is high, the segment is off (i'm using 2 82K resistors) THat, and a shift register, which is common enough, results in a very simple, microprocessor driven display. Give me a few minutes, and i'll draw up a schematic. *EDIT* Here ya go. Phear my L33t handrawn VFD symbol. *EDIT* I realised i forgot to draw the heater connections. Essentially, one is connected to ground, anf the other exactly 1V (My tubes have 1 volt heaters). Also, you DO NOT NEED TO BOTHER WITH OUTPUT ENABLE when you are not actually using them to interface parallel and serial circuitry, as the actual time involved in shifting the 8 bits of data into the register is so small you cannot see the change (something like ~1 ms, or less).
hey i came across this: http://www.selectric.org/nixie/ticlock.gif shematic for a nixie clock... just thought id share
Rugby MSF controlled LED clock project Hello all - my first post is this very interesting forum I am endeavouring to build an MSF (UK) radio controlled LED clock that just displays the time (HH:MM:SS) in 24hr format - nothing else required. I have a EM2 module/antenna & have been looking at the MSF-MCM232 / MSF-MCM RADIO modules from www.hkw-elektronik.de as a basis for extracting the serial data from the stream, the later will drive an LED display via shift registers/buffers. All very complicated (for me). With no PIC programming knowledge to speak of, apart from the fact that I have a couple of 16F84's to hand :-0, I would have to build a programmer, but that's not a problem. What I need is some code for a 16F*** PIC, to just decode the HH:MM:SS for the 60KhZ Rugby MSF transmission in the UK. Another option (was) a 12hr LED R/C clock I purchased on eBay.... the clock itself works great, but the weird multiplexed display is a nightmare, like nothing I have ever seen before, with stange combinations of connections between pairs of LED segments & split 'common' supplies across segments spanning differring digits (no pattern at all to wiring). I could use the clock for driving a custom LED display (using strings of series-connected LED's for each segment with seperate A's+K's), use buffers & hard-wire the lot, but not possible with standard commoned 7-segment displays - unless someone knows different ? Thanks
There was a good Rugby MSF synchronised clock in Elektor electronics magazine last year i think, anyway if you search the website or something i'm sure you'll find references to some really good articles on the subject.
I finally received all my LED displays from LC-Led Inc. (one delivery was considered as lost but has finally been returned to LC-Led as "not claimed"... A little "group" photography...: I just wanted to share my joy with you guys (Except when I'm thinking of all the display PCB's to develop for the LED clocks...) And now I'm starting to collect some Nixie and VFD tubes to build clocks with... CD
It is BEAUTIFUL!!! Theshadow : that schematic is a simple microprocessor - free design the SN74141s are BCD to decimal converters, that are (i am guessing) open-collector, enabling them to get around the issue of driving high voltage rather easily. The SN7490 are 4 bit binary counters. The clock is driven by a 50Hz mains frequency. If anyone wants to build one, the SN7490 - SN74141 combination could be replaced with a CD4017, or the associated TTL chip.
WOW - that is a wonderful range of lovely displays ! Thanks very much to g0th for the information re the ELEKTOR project on building a MSF clock receiver - I found the article eventually, but the design is too 'over the top' for what I need (just a simple HH:MM:SS display). It uses an amazing 8 IC's, including a AT89S53-24PC u/p & a Xircon X25320P ?, all mounted on 3 PCB's ! So, I am still looking for some PIC code for MSF (HH:MM:SS only) I have been in touch with Frizie at PicBASIC.nl, but he cannot help If anyone can help, I would be really grateful In the meantime, I have built a simple 'missing pulse detector' based around a 555 timer, which is set to trigger at a pulse length of >300ms. Basically, the circuit ignores 58 out of the 59 inverted 100ms or 200ms pulses from the receiver module. On the 59th second, there is a long break (start of code minute marker), which is 500ms for MSF & much longer for DCF77. This long pulse triggers the 555 timer, which produces a negative output on pin 3. This can drive an LED/sounder directly, giving a pleasing 'beep' each minute This pulse could be used to resynchronize any digital clock with an 'adjust/set' button, that 'zeros' the seconds in set mode. You would still have to set the clock mannualy to start with, (HH:MM), but the seconds counter would be reset automatically to zero every minute. (Or use a decade counter to do this say once every 10 minutes !). I'll post the circuit later, if anyone is interested.... I suppose I COULD use DCF77 here in England, but reception is much poorer than MSF & I'm in a built up area with lot's of electrical noise - Ideas ??
I developed a complete PIC-routine (ASM) for DCF77 HH:MM:SS and DD/MM/YY (and also day of week and month in characters) decoding which is working great. I just downloaded the MSB coding documentation, will have a look at it. Don't worry too much about the DCF reception. The DCF77 transmitter in Mainflingen (near Stuttgart in Germany) has a power output of 50kW where MSF in Rugby has a 10kW transmitter. When I was on holiday in Corsica (about 1500km from Mainflingen?) a few years ago my DCF77 watch and DCF77 travel clock still continued to synchronize normally (the official range is about 1500-2000km). Anyway, if the DCF77 reception is disturbed by electrical noise in your area, your MSF reception will be too... the cheap AM-receivers and antennas used in those decoder modules are the culprit. Reception can be greatly improved by putting the antenna on the outside of the house through a 3-wire shielded cable. I don't know about prices for MSB modules, but DCF77 modules are available from Conrad for 9.95 Euro (about 6.68GBP). CD
Lots 'o clocks There is a lot of great info in this thread. Thanks to everyone who has posted. I'm working on a project that will enable a PC (or possibly uController) to contol a large number of HH:MM clocks (60+). The idea is to have a "wall of clocks" that are each set about 1 sec apart in time. So when the time changes, it does so in a pattern. I have a design in mind that I am confident will work. It uses four TI CD74HC4511E per clock to decode BCD, drive the leds, and latch them on. My last hold-up with the project are the 7-seg led digits. I need 240+ (4x60=240). That quantity in the 2.3" size led from LC-LED.com would cost $895.20 USD!!! Ouch! I dont need any of those fancy blue ones.... just plain-old red will work fine. Does anyone have an idea of a place to get super cheap ones? I'm looking for digits that are at-least 1". Places like digi-key and mouser have some, but only in small sizes. Thanks!
Thanks for the info Turbokeu - I think I'll give it a go with DCF77 then. The reception problem is a mystery... I have a Junghans MEGA1 DCF watch (a present from the wife many years ago) - it has always had difficulties with reception of a signal... but stand in a field, and there are no problems I have a couple of spare ULN2003 darlington drivers from when I was making some stepper-motor circuits... I will use these for the LED anode drivers. Here is that circuit I promised... not up to your excellent standard I'm affraid - I used 'Circuitmaker Student' for this - you can see why they called it 'student' The 555 is connected as a standard pulse generator, but Q1 is indirectly wired across C1 and is driven via trigger (pin 2) which is fed from the negated output of the DCF77/MSF receiver module. R1 (variable) & C1 set the monostable's period to about 300ms (outside the usual 100ms / 200ms breaks sent each second). Therefore the output of the 555 remains high until such a time that a period greater than 300ms is detected - ie the 59th second ! On this occasion, C1 has time to discharge and this triggers the monostable. The 555 can directly drive an LED or buzzer, but the added CD4093 schmitt trigger, (gates shown on the right), provide a pleasant 1-second / 1khz tone at the beginning of each time code frame. This makes for a very easy way to alternativley synchronize a clock, to DCF77/MSF each and every minute....
I don't think you will find 2.3" displays much cheaper than that, even in red... Farnell has a big choice, but not cheaper. Maybe you could find cheaper 2.3" displays from Kingbright. CD
You're welcome. BTW, I found this cheap MSF receiver at Maplin. Why don't you use a microcontroller to decode the MSF signal from the receiver? I found a link to a PIC MSF clock with LCD display: http://www.insideelectronics.net/projects/clock/ The sources are in CCS "C", the LCD display routines could be easily replaced by LED display routines (when I developed the DCF routines for my LED clock I debugged them on a LCD display too). CD
Is there anything similar to MSF or DCF in America? I know there is, but i have no idea what it is called. Also, are you multiplexing the Nixies? I have heard that multiplexing nixie tubes is an issue because nixie tubes take a few Ms to turn on (something to do with the neon gas having to ionise, etc...)
Yes, WWVB transmitter near Fort Collins, Colorado. Frank Techniek in the Netherlands sells WWVB receiver modules, so you will probably find suppliers in the USA. Yes, I know, and yes, the nixies in my clocks are multiplexed (2x3 mode). I don't know the MPX frequency because the PIC I use is sold preprogrammed with firmware 5.8 from Claus Urbach. Moreover, beneath the fact that the nixies are multiplexed, they are also PWM'ed for brightness control. CD
@ Turbokeu Thanks for the http://www.insideelectronics.net/projects/clock/ link - very interesting... something to work on. As a overdue beginner to PIC programming (been stuck with CMOS & linear designs for years), I really must make an effort to get familiarised with it. There seem to be so many programming options though : assembler, C, Pic Basic, 'PicAXE'.... what would you recommend for a beginner ? The Maplin receivers have been discontinued (last week) - just the MSF antenna available now... I found this in a draw today. It's an old Hitatchi LCD display using the HD44780A LCD driver, taken from a fax machine as I remember... looks like 16 x 2 lines, but I can't find any data on it. Model is LM104L. 14 wires : pin 1 GND, pin 2 +5V, pin 3 contrast etc... similar to the LM018L by the looks of it - all matrix dots display, scrolling in a random vertical pattern when powered up (inputs left floating). So hopefully it might work with the DCF77 www.picbasic.nl software for the 16F628A
Richard, I'm really sorry for the late response... Yes, you can use this LCD with the Picbasic project, since it's a standard HD44780 LCD display (search google for the connections). Use the schematic and SW as found on ww.picbasic.nl, once it works you can start using your own (or my) routines. I advise PIC-assembler to begin, you can always swith to 'C' or another language later if you prefer (I know ASM is a bitch to learn and understand but it is very powerfull, once you know all the insides of a PIC it's really simple to manage every bit of it...) If you want, I can make a schematic of the connections I used (I use the LCD in 4-bit mode), and send you my ASM routines. CD
In succession to my blue LED display collection, here comes my actual nixie tube collection: In detail: CD
