Since no one else seems to be saying it, I will. Dave is a friggin genius for figuring this out when he was a teenager, especially with no internet to look up things. I could barely work a transistor at that age.
I can hear the affection in your voice for this old project Dave. One of my first projects when I was into electronics as a teenager was a digital clock kit from Tandy in the UK. (I think.) I still remember the joy I had putting it together, and even though it no longer works, I haven't got the heart to throw it away.
Clock design is an excellent educational tool, not only for the self taught hobbyist but for engineering students as well. I remember having to design an alarm clock for a first year subject, having barely any knowledge of electronics or logic. The more I looked into the problem, the more I had to learn about electronics design and interactions between components improved. By the end of the task, I had designed the whole thing from scratch, and had an insatiable desire to understand the engineering behind everyday tech. The floodgates were opened, and all of a sudden I had a platform of knowledge to design anything I could imagine.
All ahmed did was do a tear-down on an old clock and shove it in a "pencil case" that looked more like a suitcase, plus whats the point of a clock if you have to open the case to look at it. It wasn't an assignment and even I knew at his age that what he "created" was nothing to be impressed or proud of, pretty sure Ahmed's father is also a politician and it was his plan. People should really research before taking what the media says at face value.
+GT WING This. Father is a millionaire, did a huge media run before the kid was even in trouble. Also started a massive lawsuit with top lawyers which will suck the school dry. Also funny seeing his official twitter account's first tweet is a RT by Anil "Dox up; journalism is activism" Dash who's a massive political shill.
+Duderobi Patschie it's completely fucked up that teenagers get _arrested_ by the _Police_ for doing silly stuff at best. The world really went apeshit crazy
+GT WING Ahmed took a perfectly good clock and made a toy bomb. A 3 day suspension for bringing that to school seems about right. The kid that got suspended for chewing a Poptart into the shape of a gun is who really got shafted. Not to mention the school that banned calling Wednesday 'hump day.'
Very much enjoyed that Dave. Thank you for all the time and effort you put into explainations. Its easy to grasp the logic stuff the way you bring it accross. I'm re-ignighting my love of digital electronics at the moment in my shed (the wife calls it the man cave). Currently refreshing the 80's child memories I have of messing with this stuff myself. It would be wonderful to see more video's like this. I dont know how many others are thinking the same but a show of likes might give some indication. Anyway, just want to tell you how much the community appreciates the hours you put into these vids. Very grateful Brit here.
I made a clock like yours in 1972. With circuits in TTL 74XX series of Texas Instruments. I used Nixies tubes powered with a voltage of 150 V and with suitable drivers (7441). All was powered by a transformer recovered in an old tube amplifier (EL84 / ECC88). To make my first circuit, I drew directly on copper the tracks with a pen and paint. I remember a bunch of holes drilled with a mini drill. My mother was very unhappy when she saw the small holes separated by 1/10 of inch in the living room table. I still have that table, and every time I see those little holes, I think of my mother and my clock. That was 43 years ... This clock has it been long in the refrigerator in the family home, but it disappeared with the removal ... damage. Bye Sorry for the google translation French ===> English
+rene lefebvre Ha, a table with non-metric holes in it; you should 3D sculpt a piece of wood to put in as inlay. Wow, nixie was harder to find in '72, nice.
I once designed a discrete logic clock myself and wow, I got off real easy since we use 24 hour clock here, so my counting sequence goes ...22, 23, *0*, 1, 2... No problems at all for the hours reset. I hadn't even considered how much harder 12 hour clock would be!
The next step? I don't know man... Jumping from taking a pre-built clock out of it's case without any modification whatsoever to building one from scratch is a pretty big step.
THINboxKING I wouldn't discourage learning from hacks. That's how I started learning programming etc. I just would't call pulling the guts out of a clock, then mounting them in another case... a hack.
+Ognjen Galić IIRC, this was not uncommon back in the 70's (though alive, I was not aware of such things in the 60's). I'm pretty sure I had an alarm clock that did this (@60Hz for the USA). I was still just a kid then, but my EE dad explained it to me.
+taiwanjohn Yeah, it's normal... is one reason you can't use a USAnian clock in Europeland, and vice versa, even with a transformer. (or couldn't, back in the day!)
Dave, can you do a series on the most common discrete chips ? I appreciate that they're retro but there's a lot to be learned from how these work and how you would use them in projects like this clock.
Dave being such a Back to the Future nerd, I'm a little miffed that he's missing the obvious joke about how Ahmed didn't create a bomb, but made "a shiny bomb casing filled with used digital clock parts".
Beautiful discrete design really. You made me nostalgic to think about my old 2 digit digital bicycle speedometer I designed using CD4518 and CD4511 and NE555 as timing controller. I was in school then and design was very unique using IR LED and Photo transistor made from SL100B by cutting its upper metal shield. It used the latching feature to stick to the measured value and everything was possible. It was probably in the year 1990. Thanks to all the silicon vendors for making the life so easier now.
Ahh, brings back memories of TTL and salvaged resistors (early '80s). So you were a young teen in '87 and now got your own wife and kid. How time flies when not trying to track it.
Back in the day, as in late 80’s, I was working with a group that was testing explosive detonators for safety. We were measuring the accuracy of the delay, using a multichannel Honeywell reel to reel recorder to test multiple detonators at once. The problem was that we needed an accurate time mark generator to measure the delays. I made a tiny, 9 volt battery powered box with a 10 MHz oscillator and counters. The trick was putting in hundredth millisecond, millisecond, and hundredth of a second code bursts which overlayed a count by modulating the width of the pulses. I used some of the same analog trickery you did to shift in portions of the counter output without losing the accuracy of the leading edge, but visually apparent and readable when the tape was replayed. I think that the mixing of analog and digital like that, with diodes, transistors, resistors caps, with CMOS gates is something that has all but vanished from design. It’s not a great loss, but there is a degree of understanding that is now abstracted away.
Even as a beginner hobby engineer who doesn't grasp concepts like ground and impedance well, I was able to follow along and work out some ideas, great video!
How have I not seen this one before? A great project and sure as hell nothing to be embarrassed about. This is smashing stuff and shows you what you can do without PCB prototype fabrication and what not. Brilliant stuff!
Nothing to be embarrassed of course. I could never think of something like this until mid term in my EE engineering course ..and Dave did this when he was a teenager.
I can do this programmatically, I can flowchart this, but translating this into cmos logic with supporting circuitry as Dave has - way beyond me. The fact he did this as a teenager makes me so envious. Hat's off to Dave and everyone else who got the EE bug earlier than I did.
I have a bunch of similarly dated parts laying around from my dad's old student projects. A frequency counter among other things using these same types of chips! They still work, too. Was fun to play around with when I got to the point of starting to understand logic gates and IC's.
Hi Dave, yes I have grown up as well with the 40xx-Series. And even today, if the job is too siple for an Arduino, such a chip does its job. The Vcc 3...15 Volts and the nearly intinite input impedance are just great features. I hope the series will live for ever.
At 9:56 Dave mentions that he tapped off 50hz to use as a reference clock. I did the same thing about the same time with our US 60hz power. Maybe 50 hz would be a bit easier to just divide by 5 rather than by 6 to get 10hz to get the 0.1 second digit ticking along. We really need to change our power cycles to some nice binary 256hz or something! Wouldn't that be sweet! At that time I only had a single channel Heathkit scope (1mhz?). As an accessory to do digital design I built a little mux to allow 5 (10?) digital inputs by summing the individual digital signals with individual stepped DC voltages. Worked a treat. Ahhhh, memory lane.
Ahh memories.. I made similar clock 15+ years ago using jk flip flops as counters and basi logic gates for setting and resetting them. it had pcb 2 boards - one with counter and the other with 7seg led decoder and the leds. You had to switch it into "high speed mode" and wait till the clock forwarded to real time, then set it to normal speed.
Sir, that clearly looks like a thermonucular warhand. Judging by the cable colors, it has the energy of 50 megabytes of TNT. Unfortunately I haven't been able to decode the Arabic hieroglyphs you wrote on the sheet of paper, but my team is working on it. You will pay for your crimes!
This reminds me of the clock I built in the late 70's, still going perfectly now. It was from a kit using a single chip clock, everything except the display drivers is on the chip. It's 6 digit with a single led as the am pm indicator. I spec'd mine with the Xtal oscillator time base upgrade, on a separate board, and the power fail battery back up also. Took a bit of fiddling to get the Xtal osc set for accurate tine, due to temperature drift. It swings around -15 to + 15 seconds summer to winter. Quite a neat kit with good F/glass boards, ribbon cable wiring etc. Would like to modify it to radio time signal setting, not sure how that would be done though.
That's nice, especially diode logic circuits like not(not(a and b and c)). They almost deserve their own vlog as they are handy if someone wants to trigger one pin on microcontroller by three different signals (possibly some of them inverted)
Daves mention of low power and 4000 series reminded me of a gas billing meter I briefly worked on back in the 1980s. It had a custom LCD display with individual segment drive (gated square waves, no LSI drive chips) on multiple circuit boards covered in 4000 series chips. The whole thing operated for a couple of years on D size batteries. It was almost a computer made of standard 4000 series chips. It recorded natural gas use at a car assembly plant here in Wellington NZ. I still occasionally work on a Japax EDM machine that is based on a series a Toshiba chips that are CMOS but have TTL pin-outs. Probably a very early CMOS series from the 70s. 74HC series chips fit and sometimes work as replacements. But the original machine designers relied in many places on the propagation delays and 74HC chips are too fast, causing the logic to fail.
I used a similar case for a multipurpose LED serial display I made, has a few PICAXE micros, one took in 8bytes of 4800b serial and controlled the LED displays via shift registers (ucn5841 eight of em to drive the displays, most expensive parts aside from the displays) the other took interrupts from an automotive analogue clock ticker coil drive output, rectified to get 1pps. It ran off a higher voltage so ran ok on 5 volts and was enough to trigger a micro's input without buffering. The code space was pretty small (256byte max) but I managed 24 hour, animated segment between hour and minute with 12 hour hourly beep with setting routines in the clock micro and on the display ~32 ASCII codes (bell,0-9 ":" A-U) controlling 64 segments and a beeper on 4 outputs and a serial input
Don't ever throw out those 4000 series, I recently got into some amateur electronics thanks to your videos and I'm playing with 4000 series stuff at the moment :-) I'll always take em off your hands, and happily pay you for them. I get mine from dodgy chinese websites and they take so long to arrive. They're really cheap though and a LOT of fun. Just got hold of a couple of those 4026 and some seven segment displays.
Wow. I didn't know I could have done this much with my 4000 series ICs. I have an old collection gathered from my teen years, desoldered from old equipment. Had to pause this video to check on them. I could never really throw away good working parts.
Hi Dave, I Love seeing your videos and stuff you repair and build. Much more interesting then general teardowns. I'm nothing more then a general hobbyist and sometimes have no clue what your saying but i have learnt so much from watching you. You inspire me to build projects they are nothing special and pointless and no way near what you can do but i try. Keep up the good work.
Dave, that's pretty amazing! I didn't even realize it was possible to do so much without a microcontroller. BTW, one of the things that confused me for half the video was that the reset pins on these chips are active-high instead of the more common active-low.
This is my favorite video you have done in a while. I would love to see more like this. Teardowns are fine. Mail bag is ok. This is just fun! It is like reading a Forest Mims book.
>implying that designing an analog clock from scratch using discrete components and digital logic is "the next step" after disassembling a commercial clock and putting it into a new case. That's like saying once you've modded a computer case, the next step is to design and build your own computer architecture from scratch...
EEVblog I was being facetious, but thanks for the reply! :) AntiProtonBoy Your standards for what belongs in /pol/ seem suspect. dellagd Im glad someone got it! :)
I was given a custom-programmed clock that originally was Vishay LCt-104 web tension transmitter. He reprogrammed the EEPROM to make it a clock (it also uses a TIMEKEEPER RTC/RAM chip), which a start-up, displays a greeting, fireworks, and a sailing boat. During operation, the upper alphanumeric displays show time, while the alphanumeric displays below occasionally scroll Bible verses.
As far as I know here in Australia our power line frequency 50 Hz is actually maintained to be true over long periods of time specifically for this reason. I was told it was a requirement, but I am not sure how true that is.
Can't we also use the LOAD pin of the BCD counter IC (if it has one) to load a 1 to the first hour counter and and just CLEAR pin of the tens hour counter to get 01 after 12 ,by kind of decoding 12 ?
Very nice, Dave! I've built many clocks like that in my past too (usually using 74xx series though). I even built a few alarm clocks that actually had two different clocks (one to keep the time, the other to store the alarm time) and then a magnitude comparator to set an RS latch when the clock matched the alarm times.
man what you did with the hours/12 digit cycle is remarkable. this is a real beautiful clock. its really cool to see a clear example of using that 50hz power clock. heres hoping nobody changes it to 60hz.
Nice! Just been playing with 74 series counters this week. So perfect timing for me. The diode logic gates is a fascinating way to implement combinatorial logic. Never new about diode logic gates before. Very neat.
I'm 50 years old and I have just discovered the 4026, just this last week-end I got 2x4026's connected to a double 7 seg display (a green one oh la la) and managed to get it to count up to 99 by sending it a signal from a 555 timer, I tried for ages to get the 555 to give me exactly 1 second, but could quite do it. I noticed you didn't use the mythical ungated "C" segment, which the data sheet hints can be used to for divide by 60 and divide by 12, but no one seems to know how to use it. Anyway as I said I fiddled with the 4026, 7 segment display and 555 for about 2 weeks before I got it to work correctly, mainly because at first I was trying to give the clock signal by using a tactile switch before using the 555, but ever time I had anything connected to the clock pin the 7 seg display would go haywire counting up very fast on it's own, which I guessed was like you said it was picking up noise, maybe something to do with the PSU I was using. Anyway It's very satisfying watching the green numbers count up to 99, in nearly 1 second increments.
The 4026 is what I have used for my VFD discreet logic clock. Due to the complexity of needing a multiplexer it turning into a design and build your own multiplexer project. So the multiplexer and display and PSU is the only part that has ever been completed. I should really Finnish the clock part and finally complete the project.
I did a few projects like this. I was more of a 74xx ttl and 555 timer fanboy. 4xxx series cmos chips were not all that robust ( perhaps in my hands). That mains strain relief is still a lot better than nothing.
Kinda cool. Thanks for sharing. I think I will try start trying to design my own setup! The only point in making this for me would be to practice my knowledge after a digital design class.
Never seen segment drivers "throttled" by a single common resistor. That should change your luminosity based on how many segments are lit by changing the voltage drop over the resistor with the changing current. Might not be perceptible though on account of how LEDs seemingly magically emit the exact same amount of light for anything between 1-5mA... Of course we were too lazy to hook up 7 resistors per digit too in our day, but we used a number of series diodes instead of that resistor - they dropped 0.6V each (until the drivers / LEDs were happy) and didn't care at all about how many segments are lit...
Love the old school resistor diode OR and AND gates. I used some of that myself back in the mid 70's when I designed and built myself an 1802 based CPU board. Off the top of my head, you could have saved a few parts by using a PNP transistor instead of the extra diode and pair of transistors for the 6 detector reset circuit. If you didn't have PNP's back in the day (since you were using what you had at hand), it makes sense you used a pair of NPN transistors.
"what can I do with them, I know - make a clock!" Heh, I'm in almost the same situation right now. Got a nice case from old APC UPS, initially wanted to put the step-down transformer here, to get 100v from 220v for the Japanese record player. Then I thought "Hm, that case seems a bit too big for the transformer alone, and I don't want other extra components sitting around and making a bit of a mess". So now I'm building 4-channel stereo multiplexer (74HC4052) with 1 phono input (and 3 line inputs), separate line out, and built-in headphone amp, all that with digital controls powered by AVR. Just because I've got parts salvaged and just because I can :)
Still have my trays of CD4000's and 74-series IC's around somewhere. I miss the days when TV's and VCR's were actual parts goldmines. BTW Dave, that clock is _nobut bodge_ :)-
+Duan van't Slot Not surprising actually, it would need reciprocal counting to get better precision due to the slow gating time. It shouldn't display all the digits though.
It was not a bomb, nor did he make a clock.. he just pulled the guts of an old consumer clock and put in a case.. Ahmed Mohamed is a liar. And if EEVblog can't see that, then I have no reason to subscribe.
i think it would be much simpler to use a 4026+4013 pair for the hours and reset from 13->1 using the kind of logic you used for the 6's reset, with a simple monostable pulse delay to bump the hours from 0->1.
O-GOD! This brings back memories. I design and built a similar clock out out 74 series logic. except mine didn't have the 10th of a second count and you could not set the time ether. to set it at the right time. you would plug it into the wall at 1:00, because it started at 1:00 when you plug it in. I was 12 years old back in 1983 when i built this thing and I no longer have it anymore sadly.
Love It.. I built a TTL clock just like this with a bunch of 7490's and Hewlett-Packard 5082-7340 displays. Lets add a 24 hour option to this clock. I would Love to see how you deal with that using Diode/transistor logic. More Videos Like this Please....
A 24 hour clock would be easier: you would just connect the two bcd counters with the carry as normal, creating the reset as soon as you reach 24 which is just an AND of q2b and q3a.
Agreed. Keeping the 12 hour clock and adding a 24 hour option will be more difficult. You would need to switch out (Some or all) of the 12 hour diode logic and switch in the more Simplified 24 hour circuit. It will be a very interesting how to video working all of that out. That's all i'm saying...
Dave, what you seem to miss is that little Ahmed is NOT at all interested in electronics. He is interested in getting attention. I find it very hard to believe Ahmed will ever get to the point where he can build a clock like this one himself. Time will tell, now that he is a celebrity it should be pretty easy to follow his progress through life.
This makes me want to roll a clock like this right now. I would actually use a watch crystal (32768Hz, paired to a 74LVC1GU04 and a 74LVC1G14 to emit a 32768Hz signal) and instead of a huge board of chips I would use a CPLD. My LED module requires multiplexed input, so the driver circuit is pretty much a CD4511B, two 74HC06's and a 74HC138. The entire thing would be powered by a single 18650 lithium-ion cell directly.
Steve Nordquist Should I use higher precision oscillator I would prefer something like DS32kHz DTCXO over the MEMS oscillator, or go with a microcontroller and a DS3231 I2C DTCXO RTC. DTCXO are more accurate than MEMS oscillators. My preferred CPLD is lower pin count Altera MAX II series like EPM240 or EPM570. Cheap and still supported by the latest free software.
+陈北宗 That DTCOX crystal is going to outlast the watch crystal, too. I was thinking of the oscillators that also control voice waking for phones...so Wolfson digital mic. parts (wolfsonmicro.com ...now part of Cirrus?) The mems rtc would permit sleeping 100ms or more. DS3231M is the i2c one...but 5ppm accuracy is off maybe .4s a day. SiT1552 is a plain oscillator w. that accuracy. 0.5ppm sounds better all right.
Steve Nordquist Should I use an I2C RTC chip it would be the DS3231 non-M. Coupled to an ATmega328P running on its internal oscillator straight off a 18650 cell, it makes a decent clock. The CD4511/74HC06/74HC138 driver circuitry would still be used though.
I might be wrong but I seem to remember that the 4026 had an direct output active on 2 so you could AND that to the carry to reset on twelve ? I built one of these years ago and I seem to remember doing it that way
Hey Dave, you ever thought about doing a mini-series on basic discrete logic (TTLs, cmos, 7400s, etc) design and projects? I guess I can stick to microcontrollers until I go to university, but this seems really interesting. Or if you know some good books or anything for beginners that'd be great
Hehe :) Awesome! Is put me back one of My biggest mistery diy nightmare weekend. I did something with thermal controlling unit of an old russian heating central. and I used simple diodes to make or-nor gates. And one diode is printed reverse.... I spend 3 days on it (tons of diodes in a controller box) One diode can beat You down, and You get fully confused. And in this speciel controller box, nothing happened immediately. There is heating, cooling, ignition, opening, closing valves... With total random periods.. Ahhhhh, I almost cry :) Then... The last what I checked is the diodes. What a blama :))) Thank You dave :)
+jasonteknut I had several bookcases full of databooks! Every hobbyist did, it was the only way to find information. No Internet back then, no PDF's, just dead tree.
Seems like some work to get 12 hrs to work. Looks like some diode logic. In a similar year,, I did a similar clock based on 60Hz but I used 74xx series chips. Similar wiring style, lol. I still have the four digit display but I must have breadboarded the rest. I might have used a nand gate or two for the cases where it would not go to ten.
hey, thats how people used to prototype back in the day. If you were fortunate you would make this nest with wire wrap. Its just how things were done but the stuff worked.
The sister of the boy who brought a suspected hoax-bomb to his Texas high school said she was suspended from a school in a prior bomb scare. Her suspension occurred in 2009 while she was attending middle school in the same district.
No mailorder sure, but stuff was easier to salvage. I remember as a kid having loads of old hardware.. all through-hole stuff. Nowadays everything is tiny.
I have a question : you reset when it hits 12 o'clock then force it to display "12". Why not choose to reset when it hits 13 instead ? so it displays 12 no problem and when it hits 13 boom reset ? Have I missed something ?
+mrozu1337 Then instead of forcing a twelve on all zeros across both BCD counters force a one only when the tens counter was zero. Would clean up the circuit a bit I think.
+Andrew Querol Then you would get "1" twice. Resetting at 13 would be the obvious first thought, but then you'd need to generate quick extra clock pulse to get it from 0 to 1. Instead counting from 0 to 11 is just clever.
renxula well either way in the end would end up just as complex I would think. Maybe you could attach the carry signal to the set input of a rs-nor latch. And have the output signal of the rs-nor latch connect to the base of a transistor that would control the main hour clock from going into the tens counter and also connected to e and f signal lines of the tens segment display. Then connect the b output signal of the tens counter to the reset pin of the rs-nor latch, the tens counter, the ones counter, and the am pm indicator input. But that would be more discreet logic. That ended up more complex than I intended for writing on a phone :p
![BOWKYLION - วิงวอน (ex-change) [Official MV]](http://i.ytimg.com/vi/cLdnZWDaO-w/mqdefault.jpg)
this time, make a new schematic, fold up the paper and put it inside the case.
Since no one else seems to be saying it, I will. Dave is a friggin genius for figuring this out when he was a teenager, especially with no internet to look up things. I could barely work a transistor at that age.
I couldn't even think about making this one until like 2nd year into EE engineering.
I can hear the affection in your voice for this old project Dave.
One of my first projects when I was into electronics as a teenager was a digital clock kit from Tandy in the UK. (I think.) I still remember the joy I had putting it together, and even though it no longer works, I haven't got the heart to throw it away.
Repair it!!
Clock design is an excellent educational tool, not only for the self taught hobbyist but for engineering students as well. I remember having to design an alarm clock for a first year subject, having barely any knowledge of electronics or logic. The more I looked into the problem, the more I had to learn about electronics design and interactions between components improved. By the end of the task, I had designed the whole thing from scratch, and had an insatiable desire to understand the engineering behind everyday tech. The floodgates were opened, and all of a sudden I had a platform of knowledge to design anything I could imagine.
First year? That's cool. I had Digital logic course in 2nd year aka 4th sem
All ahmed did was do a tear-down on an old clock and shove it in a "pencil case" that looked more like a suitcase, plus whats the point of a clock if you have to open the case to look at it. It wasn't an assignment and even I knew at his age that what he "created" was nothing to be impressed or proud of, pretty sure Ahmed's father is also a politician and it was his plan. People should really research before taking what the media says at face value.
+GT WING
This. Father is a millionaire, did a huge media run before the kid was even in trouble. Also started a massive lawsuit with top lawyers which will suck the school dry.
Also funny seeing his official twitter account's first tweet is a RT by Anil "Dox up; journalism is activism" Dash who's a massive political shill.
+GT WING I agree with you 100%
+Duderobi Patschie it's completely fucked up that teenagers get _arrested_ by the _Police_ for doing silly stuff at best.
The world really went apeshit crazy
+GT WING It doesn't bloody matter. The clock clearly doesn't LOOK ANYTHING LIKE A BOMB... so, what the motive was doesn't matter
+GT WING Ahmed took a perfectly good clock and made a toy bomb. A 3 day suspension for bringing that to school seems about right. The kid that got suspended for chewing a Poptart into the shape of a gun is who really got shafted. Not to mention the school that banned calling Wednesday 'hump day.'
Very much enjoyed that Dave. Thank you for all the time and effort you put into explainations. Its easy to grasp the logic stuff the way you bring it accross. I'm re-ignighting my love of digital electronics at the moment in my shed (the wife calls it the man cave). Currently refreshing the 80's child memories I have of messing with this stuff myself. It would be wonderful to see more video's like this. I dont know how many others are thinking the same but a show of likes might give some indication. Anyway, just want to tell you how much the community appreciates the hours you put into these vids. Very grateful Brit here.
I made a clock like yours in 1972. With circuits in TTL 74XX series of Texas Instruments. I used Nixies tubes powered with a voltage of 150 V and with suitable drivers (7441).
All was powered by a transformer recovered in an old tube amplifier (EL84 / ECC88).
To make my first circuit, I drew directly on copper the tracks with a pen and paint.
I remember a bunch of holes drilled with a mini drill. My mother was very unhappy when she saw the small holes separated by 1/10 of inch in the living room table. I still have that table, and every time I see those little holes, I think of my mother and my clock. That was 43 years ...
This clock has it been long in the refrigerator in the family home, but it disappeared with the removal ... damage.
Bye
Sorry for the google translation French ===> English
+rene lefebvre Ha, a table with non-metric holes in it; you should 3D sculpt a piece of wood to put in as inlay. Wow, nixie was harder to find in '72, nice.
it all feels undoable when someone like me , a 2019 digital electronics fresher reads . your dedication sure was rock hard . hats off sir
I once designed a discrete logic clock myself and wow, I got off real easy since we use 24 hour clock here, so my counting sequence goes ...22, 23, *0*, 1, 2... No problems at all for the hours reset. I hadn't even considered how much harder 12 hour clock would be!
The true spirit of electronics. I bet, even experienced electronics hobbyist will have a hard time replicating that box of epicness.
The next step? I don't know man... Jumping from taking a pre-built clock out of it's case without any modification whatsoever to building one from scratch is a pretty big step.
+ArtificialDuality Err, yeah, I didn't mean that literally.
+ArtificialDuality thats where i started now i build "robots", don't discourage learning from hacks.
+EEVblog Kia ora from across the ditch
THINboxKING I wouldn't discourage learning from hacks. That's how I started learning programming etc. I just would't call pulling the guts out of a clock, then mounting them in another case... a hack.
Using Mains sine wave to sync clock.
Can, can you believe that! That is GENIUS!
+Ognjen Galić Very usual in old digital clocks. What needed some good thinking is the 12 hour problem. Very impressive.
+Ognjen Galić IIRC, this was not uncommon back in the 70's (though alive, I was not aware of such things in the 60's). I'm pretty sure I had an alarm clock that did this (@60Hz for the USA). I was still just a kid then, but my EE dad explained it to me.
+taiwanjohn Yeah, it's normal... is one reason you can't use a USAnian clock in Europeland, and vice versa, even with a transformer.
(or couldn't, back in the day!)
Mains sin wave is NOT ACCURATELY 50hz!
Bradman175 I think they try to balance it so that the average is 50 Hz over some time.
Wow! Take that to show your Prime minister. In a suitcase.
Dave, can you do a series on the most common discrete chips ? I appreciate that they're retro but there's a lot to be learned from how these work and how you would use them in projects like this clock.
Dave being such a Back to the Future nerd, I'm a little miffed that he's missing the obvious joke about how Ahmed didn't create a bomb, but made "a shiny bomb casing filled with used digital clock parts".
+Dosbomber -and he didn't have time to build it to scale or to paint it....
Beautiful discrete design really. You made me nostalgic to think about my old 2 digit digital bicycle speedometer I designed using CD4518 and CD4511 and NE555 as timing controller. I was in school then and design was very unique using IR LED and Photo transistor made from SL100B by cutting its upper metal shield. It used the latching feature to stick to the measured value and everything was possible. It was probably in the year 1990. Thanks to all the silicon vendors for making the life so easier now.
Ahh, brings back memories of TTL and salvaged resistors (early '80s). So you were a young teen in '87 and now got your own wife and kid. How time flies when not trying to track it.
Back in the day, as in late 80’s, I was working with a group that was testing explosive detonators for safety. We were measuring the accuracy of the delay, using a multichannel Honeywell reel to reel recorder to test multiple detonators at once. The problem was that we needed an accurate time mark generator to measure the delays. I made a tiny, 9 volt battery powered box with a 10 MHz oscillator and counters. The trick was putting in hundredth millisecond, millisecond, and hundredth of a second code bursts which overlayed a count by modulating the width of the pulses. I used some of the same analog trickery you did to shift in portions of the counter output without losing the accuracy of the leading edge, but visually apparent and readable when the tape was replayed. I think that the mixing of analog and digital like that, with diodes, transistors, resistors caps, with CMOS gates is something that has all but vanished from design. It’s not a great loss, but there is a degree of understanding that is now abstracted away.
Interesting.
Yes micro controllers and processors have made a lot of these digital and analog beauty to vanish
Even as a beginner hobby engineer who doesn't grasp concepts like ground and impedance well, I was able to follow along and work out some ideas, great video!
Wow Dave, that's so genius of you to design and build that digital clock with CMOS logic
Dave, When you were building this clock, I was designing stuff for a major electronics company in USA. Love the diode gates!!
You must be in your 70s now?
@@MikeJohnMentzer True. I used the same cases as he used and the same displays in my designs.
@@williamday7132 Thank you for your service
Can't find words to thank you Dave!
10/10 bomb confirmed
**How to design a digital bomb 💥💥💥
+EvilCatNip *Logic bomb
+EvilCatNip i'm calling cops
+EvilCatNip At least it's a real clock, not that bullshit Ahmed (or whatever his name is) built
A clock? It looks more like a bomb.
+turntechgodhead100 Nice meme m8, i r8 8/8 cuz its gr8 m8
+John Cena i r8 8/8 gr8 b8 m8
+turntechgodhead100 Indeed. At 5:27 , one can clearly see the port for the detonator.
Ernst Stavro Blofeld You can also see the tnt.
+turntechgodhead100 Naw, It's looks 'shopped. I can tell by the pixels and from seeing quite a few 'shops in my time.
Thanks, Dave. Now we know what an actual hobbyist clock should look like. Very nice.
How have I not seen this one before? A great project and sure as hell nothing to be embarrassed about. This is smashing stuff and shows you what you can do without PCB prototype fabrication and what not. Brilliant stuff!
Nothing to be embarrassed of course. I could never think of something like this until mid term in my EE engineering course ..and Dave did this when he was a teenager.
As dated as that is, it still to this day has inspirational qualities to anyone learning electronics.
Dude you are already 8 years ago
@MikeJohnMentzer I'm glad to see you got the counter part of that design figured out.
@@InsideOfMyOwnMind hahaha
8 years..time flies!
Great design and teaching moment! I too am obsessed with scratch-built clocks.
I can do this programmatically, I can flowchart this, but translating this into cmos logic with supporting circuitry as Dave has - way beyond me. The fact he did this as a teenager makes me so envious. Hat's off to Dave and everyone else who got the EE bug earlier than I did.
reported this vid to the FBI For bomb building instructions
+Sycabara You needed have bothered, it has all the requisite keywords, they already know.
+EEVblog True.
+Sycabara Now the FBI will be all over Austria...
I have a bunch of similarly dated parts laying around from my dad's old student projects. A frequency counter among other things using these same types of chips! They still work, too. Was fun to play around with when I got to the point of starting to understand logic gates and IC's.
Hi Dave, yes I have grown up as well with the 40xx-Series. And even today, if the job is too siple for an Arduino, such a chip does its job. The Vcc 3...15 Volts and the nearly intinite input impedance are just great features. I hope the series will live for ever.
At 9:56 Dave mentions that he tapped off 50hz to use as a reference clock. I did the same thing about the same time with our US 60hz power. Maybe 50 hz would be a bit easier to just divide by 5 rather than by 6 to get 10hz to get the 0.1 second digit ticking along. We really need to change our power cycles to some nice binary 256hz or something! Wouldn't that be sweet!
At that time I only had a single channel Heathkit scope (1mhz?). As an accessory to do digital design I built a little mux to allow 5 (10?) digital inputs by summing the individual digital signals with individual stepped DC voltages. Worked a treat.
Ahhhh, memory lane.
Ahh memories.. I made similar clock 15+ years ago using jk flip flops as counters and basi logic gates for setting and resetting them. it had pcb 2 boards - one with counter and the other with 7seg led decoder and the leds. You had to switch it into "high speed mode" and wait till the clock forwarded to real time, then set it to normal speed.
This exact video inspired me to build my own. I used 4510 fed into 4511 and AND gates to reset at 60/24. It works great!
Sir, that clearly looks like a thermonucular warhand. Judging by the cable colors, it has the energy of 50 megabytes of TNT. Unfortunately I haven't been able to decode the Arabic hieroglyphs you wrote on the sheet of paper, but my team is working on it. You will pay for your crimes!
This comment made my day.
This reminds me of the clock I built in the late 70's, still going perfectly now. It was from a kit using a single chip clock, everything except the display drivers is on the chip. It's 6 digit with a single led as the am pm indicator. I spec'd mine with the Xtal oscillator time base upgrade, on a separate board, and the power fail battery back up also. Took a bit of fiddling to get the Xtal osc set for accurate tine, due to temperature drift. It swings around -15 to + 15 seconds summer to winter. Quite a neat kit with good F/glass boards, ribbon cable wiring etc.
Would like to modify it to radio time signal setting, not sure how that would be done though.
That's nice, especially diode logic circuits like not(not(a and b and c)). They almost deserve their own vlog as they are handy if someone wants to trigger one pin on microcontroller by three different signals (possibly some of them inverted)
When I wanted to do something "low-level" in the late '90s, I programmed Z80 assembly for my TI-83. What are the kids are doing these days?
Tiktok
Great piece of nostalgia. Very informative! I learn more and more each time I watch your videos.
Daves mention of low power and 4000 series reminded me of a gas billing meter I briefly worked on back in the 1980s. It had a custom LCD display with individual segment drive (gated square waves, no LSI drive chips) on multiple circuit boards covered in 4000 series chips. The whole thing operated for a couple of years on D size batteries. It was almost a computer made of standard 4000 series chips. It recorded natural gas use at a car assembly plant here in Wellington NZ.
I still occasionally work on a Japax EDM machine that is based on a series a Toshiba chips that are CMOS but have TTL pin-outs. Probably a very early CMOS series from the 70s. 74HC series chips fit and sometimes work as replacements. But the original machine designers relied in many places on the propagation delays and 74HC chips are too fast, causing the logic to fail.
I used a similar case for a multipurpose LED serial display I made, has a few PICAXE micros, one took in 8bytes of 4800b serial and controlled the LED displays via shift registers (ucn5841 eight of em to drive the displays, most expensive parts aside from the displays) the other took interrupts from an automotive analogue clock ticker coil drive output, rectified to get 1pps. It ran off a higher voltage so ran ok on 5 volts and was enough to trigger a micro's input without buffering. The code space was pretty small (256byte max) but I managed 24 hour, animated segment between hour and minute with 12 hour hourly beep with setting routines in the clock micro and on the display ~32 ASCII codes (bell,0-9 ":" A-U) controlling 64 segments and a beeper on 4 outputs and a serial input
Dave, simply one of your best teaching videos. Thanks! A perfect follow up video would be, how to do this today with let's say an arduino and a RTC.
Don't ever throw out those 4000 series, I recently got into some amateur electronics thanks to your videos and I'm playing with 4000 series stuff at the moment :-) I'll always take em off your hands, and happily pay you for them. I get mine from dodgy chinese websites and they take so long to arrive. They're really cheap though and a LOT of fun. Just got hold of a couple of those 4026 and some seven segment displays.
Dave, It's absolutely fantastic. You implemented so neat solutions that it is unimaginable. I must admit that people are thinking different today.
Wow. I didn't know I could have done this much with my 4000 series ICs. I have an old collection gathered from my teen years, desoldered from old equipment. Had to pause this video to check on them. I could never really throw away good working parts.
Hi Dave, I Love seeing your videos and stuff you repair and build. Much more interesting then general teardowns. I'm nothing more then a general hobbyist and sometimes have no clue what your saying but i have learnt so much from watching you. You inspire me to build projects they are nothing special and pointless and no way near what you can do but i try. Keep up the good work.
Dave, that's pretty amazing! I didn't even realize it was possible to do so much without a microcontroller.
BTW, one of the things that confused me for half the video was that the reset pins on these chips are active-high instead of the more common active-low.
Micro controllers weren't popular until the late 80s maybe? Before that almost every electronics was just pure electronics lol
Very comprehensive explanation. I has a love/hate relationship with digital logic in my first year at uni!
*had
This is my favorite video you have done in a while. I would love to see more like this. Teardowns are fine. Mail bag is ok. This is just fun! It is like reading a Forest Mims book.
Best video yet. Possibly the most educational thing I have watched. Ever.
>implying that designing an analog clock from scratch using discrete components and digital logic is "the next step" after disassembling a commercial clock and putting it into a new case.
That's like saying once you've modded a computer case, the next step is to design and build your own computer architecture from scratch...
+Amra I didn't mean it literally. But perhaps it's something he can work toward to build his skills.
+EEVblog He can use a screw driver. You expect him to design his own circuitry?
+Amra this is not /pol/
+EEVblog I thought the joke was on point ;)
EEVblog I was being facetious, but thanks for the reply! :)
AntiProtonBoy Your standards for what belongs in /pol/ seem suspect.
dellagd Im glad someone got it! :)
I was given a custom-programmed clock that originally was Vishay LCt-104 web tension transmitter. He reprogrammed the EEPROM to make it a clock (it also uses a TIMEKEEPER RTC/RAM chip), which a start-up, displays a greeting, fireworks, and a sailing boat. During operation, the upper alphanumeric displays show time, while the alphanumeric displays below occasionally scroll Bible verses.
Really cool mod
Ripper use for an old device, that would probably otherwise just been chucked into the bin. (I ended up Googling LCt-104 for a looksee)
As far as I know here in Australia our power line frequency 50 Hz is actually maintained to be true over long periods of time specifically for this reason. I was told it was a requirement, but I am not sure how true that is.
Can't we also use the LOAD pin of the BCD counter IC (if it has one) to load a 1 to the first hour counter and and just CLEAR pin of the tens hour counter to get 01 after 12 ,by kind of decoding 12 ?
Very nice, Dave! I've built many clocks like that in my past too (usually using 74xx series though). I even built a few alarm clocks that actually had two different clocks (one to keep the time, the other to store the alarm time) and then a magnitude comparator to set an RS latch when the clock matched the alarm times.
man what you did with the hours/12 digit cycle is remarkable. this is a real beautiful clock. its really cool to see a clear example of using that 50hz power clock. heres hoping nobody changes it to 60hz.
Builds digital clock when he was a teenager
Grew up to invent a 32 channel logic analyser
Nice! Just been playing with 74 series counters this week. So perfect timing for me. The diode logic gates is a fascinating way to implement combinatorial logic. Never new about diode logic gates before. Very neat.
The next step for Ahmed Mohamed to learn the intrinsic of a torx screwdriver. He already mastered the philips screwdriver
.
I'm 50 years old and I have just discovered the 4026, just this last week-end I got 2x4026's connected to a double 7 seg display (a green one oh la la) and managed to get it to count up to 99 by sending it a signal from a 555 timer, I tried for ages to get the 555 to give me exactly 1 second, but could quite do it. I noticed you didn't use the mythical ungated "C" segment, which the data sheet hints can be used to for divide by 60 and divide by 12, but no one seems to know how to use it. Anyway as I said I fiddled with the 4026, 7 segment display and 555 for about 2 weeks before I got it to work correctly, mainly because at first I was trying to give the clock signal by using a tactile switch before using the 555, but ever time I had anything connected to the clock pin the 7 seg display would go haywire counting up very fast on it's own, which I guessed was like you said it was picking up noise, maybe something to do with the PSU I was using. Anyway It's very satisfying watching the green numbers count up to 99, in nearly 1 second increments.
The 4026 is what I have used for my VFD discreet logic clock. Due to the complexity of needing a multiplexer it turning into a design and build your own multiplexer project. So the multiplexer and display and PSU is the only part that has ever been completed. I should really Finnish the clock part and finally complete the project.
I did a few projects like this. I was more of a 74xx ttl and 555 timer fanboy. 4xxx series cmos chips were not all that robust ( perhaps in my hands). That mains strain relief is still a lot better than nothing.
I forgot about 555 , thanks for reminding. Nostalgia just hit hard.
24h clocks are just so much easier :)
Kinda cool. Thanks for sharing. I think I will try start trying to design my own setup! The only point in making this for me would be to practice my knowledge after a digital design class.
Never seen segment drivers "throttled" by a single common resistor. That should change your luminosity based on how many segments are lit by changing the voltage drop over the resistor with the changing current. Might not be perceptible though on account of how LEDs seemingly magically emit the exact same amount of light for anything between 1-5mA...
Of course we were too lazy to hook up 7 resistors per digit too in our day, but we used a number of series diodes instead of that resistor - they dropped 0.6V each (until the drivers / LEDs were happy) and didn't care at all about how many segments are lit...
Brilliant video, just the sort of thing to get people interested in electronics.
*grandpa voice* "back in my days we didn't have no fancy microcontrollers" :D
Brilliant Dave. I have been waiting for this video since you showed us the clock long ago.
My best clock is a broken clock. It is twice a day 100% exact. No other clock is that good!
Nice joke.
Love the old school resistor diode OR and AND gates. I used some of that myself back in the mid 70's when I designed and built myself an 1802 based CPU board. Off the top of my head, you could have saved a few parts by using a PNP transistor instead of the extra diode and pair of transistors for the 6 detector reset circuit. If you didn't have PNP's back in the day (since you were using what you had at hand), it makes sense you used a pair of NPN transistors.
Hay Dave, Nice I love all the old school stuff, excellent walk through of the circuit...... :-)
You need to have this running on the bench behind you. This clock deserves to run. :)
"what can I do with them, I know - make a clock!"
Heh, I'm in almost the same situation right now. Got a nice case from old APC UPS, initially wanted to put the step-down transformer here, to get 100v from 220v for the Japanese record player. Then I thought "Hm, that case seems a bit too big for the transformer alone, and I don't want other extra components sitting around and making a bit of a mess".
So now I'm building 4-channel stereo multiplexer (74HC4052) with 1 phono input (and 3 line inputs), separate line out, and built-in headphone amp, all that with digital controls powered by AVR. Just because I've got parts salvaged and just because I can :)
Still have my trays of CD4000's and 74-series IC's around somewhere. I miss the days when TV's and VCR's were actual parts goldmines.
BTW Dave, that clock is _nobut bodge_ :)-
Great video Dave! Thank you. Sigilent FAIL there at 15:10.
+Duan van't Slot Not surprising actually, it would need reciprocal counting to get better precision due to the slow gating time. It shouldn't display all the digits though.
It was not a bomb, nor did he make a clock.. he just pulled the guts of an old consumer clock and put in a case..
Ahmed Mohamed is a liar. And if EEVblog can't see that, then I have no reason to subscribe.
i think it would be much simpler to use a 4026+4013 pair for the hours and reset from 13->1 using the kind of logic you used for the 6's reset, with a simple monostable pulse delay to bump the hours from 0->1.
I REALLY want more videos like this
That is a nice design! Good work with the diodes there. Thanks for sharing Dave!! :)
O-GOD! This brings back memories. I design and built a similar clock out out 74 series logic. except mine didn't have the 10th of a second count and you could not set the time ether. to set it at the right time. you would plug it into the wall at 1:00, because it started at 1:00 when you plug it in. I was 12 years old back in 1983 when i built this thing and I no longer have it anymore sadly.
Love It.. I built a TTL clock just like this with a bunch of 7490's and Hewlett-Packard 5082-7340 displays. Lets add a 24 hour option to this clock. I would Love to see how you deal with that using Diode/transistor logic.
More Videos Like this Please....
A 24 hour clock would be easier: you would just connect the two bcd counters with the carry as normal, creating the reset as soon as you reach 24 which is just an AND of q2b and q3a.
Agreed. Keeping the 12 hour clock and adding a 24 hour option will be more difficult. You would need to switch out (Some or all) of the 12 hour diode logic and switch in the more Simplified 24 hour circuit. It will be a very interesting how to video working all of that out. That's all i'm saying...
Dave, what you seem to miss is that little Ahmed is NOT at all interested in electronics. He is interested in getting attention. I find it very hard to believe Ahmed will ever get to the point where he can build a clock like this one himself. Time will tell, now that he is a celebrity it should be pretty easy to follow his progress through life.
Best video in a while, more of this please!!
Absolutely loved this, thank Dave !
+Ian Clarke No worries, glad you enjoyed it.
801 awww takes me back love this vid love it
Started building one today using 4017's. 4000 series doesn't mind it bare bones.
This makes me want to roll a clock like this right now. I would actually use a watch crystal (32768Hz, paired to a 74LVC1GU04 and a 74LVC1G14 to emit a 32768Hz signal) and instead of a huge board of chips I would use a CPLD. My LED module requires multiplexed input, so the driver circuit is pretty much a CD4511B, two 74HC06's and a 74HC138. The entire thing would be powered by a single 18650 lithium-ion cell directly.
+陈北宗 Cool, but throw a MEMS oscillator in there, so you can have error compensation. Toshiba CPLD or Lattice?
Steve Nordquist Should I use higher precision oscillator I would prefer something like DS32kHz DTCXO over the MEMS oscillator, or go with a microcontroller and a DS3231 I2C DTCXO RTC. DTCXO are more accurate than MEMS oscillators.
My preferred CPLD is lower pin count Altera MAX II series like EPM240 or EPM570. Cheap and still supported by the latest free software.
+陈北宗 That DTCOX crystal is going to outlast the watch crystal, too. I was thinking of the oscillators that also control voice waking for phones...so Wolfson digital mic. parts (wolfsonmicro.com ...now part of Cirrus?) The mems rtc would permit sleeping 100ms or more. DS3231M is the i2c one...but 5ppm accuracy is off maybe .4s a day. SiT1552 is a plain oscillator w. that accuracy. 0.5ppm sounds better all right.
Steve Nordquist Should I use an I2C RTC chip it would be the DS3231 non-M. Coupled to an ATmega328P running on its internal oscillator straight off a 18650 cell, it makes a decent clock. The CD4511/74HC06/74HC138 driver circuitry would still be used though.
I might be wrong but I seem to remember that the 4026 had an direct output active on 2 so you could AND that to the carry to reset on twelve ? I built one of these years ago and I seem to remember doing it that way
Hey Dave, you ever thought about doing a mini-series on basic discrete logic (TTLs, cmos, 7400s, etc) design and projects? I guess I can stick to microcontrollers until I go to university, but this seems really interesting.
Or if you know some good books or anything for beginners that'd be great
Hehe :) Awesome! Is put me back one of My biggest mistery diy nightmare weekend. I did something with thermal controlling unit of an old russian heating central. and I used simple diodes to make or-nor gates. And one diode is printed reverse.... I spend 3 days on it (tons of diodes in a controller box) One diode can beat You down, and You get fully confused. And in this speciel controller box, nothing happened immediately. There is heating, cooling, ignition, opening, closing valves... With total random periods.. Ahhhhh, I almost cry :) Then... The last what I checked is the diodes. What a blama :))) Thank You dave :)
When you were a kid, did you have a book of chip datasheets? How did you know what the ICs did and how to use them?
+jasonteknut I had several bookcases full of databooks! Every hobbyist did, it was the only way to find information. No Internet back then, no PDF's, just dead tree.
Seems like some work to get 12 hrs to work. Looks like some diode logic.
In a similar year,, I did a similar clock based on 60Hz but I used 74xx series chips. Similar wiring style, lol. I still have the four digit display but I must have breadboarded the rest. I might have used a nand gate or two for the cases where it would not go to ten.
hey, thats how people used to prototype back in the day. If you were fortunate you would make this nest with wire wrap. Its just how things were done but the stuff worked.
Now this is one hell of a homemade clock!
Nice really. Would even look better with cmos gates on the "12" decoder instead of those diod logic (except 7 segment tap-off diodes of course)
NO!!! Its not embarrassing. Its something to be proud of!!!
Ah, I miss the days where I would take junk TVs and VCRs, desolder the parts, and organize them into my bins for reuse.
The sister of the boy who brought a suspected hoax-bomb to his Texas high school said she was suspended from a school in a prior bomb scare. Her suspension occurred in 2009 while she was attending middle school in the same district.
No mailorder sure, but stuff was easier to salvage. I remember as a kid having loads of old hardware.. all through-hole stuff. Nowadays everything is tiny.
I have a question : you reset when it hits 12 o'clock then force it to display "12". Why not choose to reset when it hits 13 instead ? so it displays 12 no problem and when it hits 13 boom reset ? Have I missed something ?
+Louis
Then there's problem to reset it to 1, not to 0.
+mrozu1337 Then instead of forcing a twelve on all zeros across both BCD counters force a one only when the tens counter was zero. Would clean up the circuit a bit I think.
+Andrew Querol well you would need to still have the nor gate for the ones counter :/ I guess it makes no difference
+Andrew Querol Then you would get "1" twice. Resetting at 13 would be the obvious first thought, but then you'd need to generate quick extra clock pulse to get it from 0 to 1. Instead counting from 0 to 11 is just clever.
renxula well either way in the end would end up just as complex I would think. Maybe you could attach the carry signal to the set input of a rs-nor latch. And have the output signal of the rs-nor latch connect to the base of a transistor that would control the main hour clock from going into the tens counter and also connected to e and f signal lines of the tens segment display. Then connect the b output signal of the tens counter to the reset pin of the rs-nor latch, the tens counter, the ones counter, and the am pm indicator input. But that would be more discreet logic. That ended up more complex than I intended for writing on a phone :p