Synchronar Update

[mrcalc]

Just a quick note. I don't have much to offer. I spoke to Roger Riehl's family who noted that they are still grieving but are starting to think about how to proceed with the Synchronar & Roger's legacy. One of the complicating issues is that since the corporation of Custom Circuits, Inc. had all the rights to the watch, with Roger's death, the various probate courts & such need to work out some of those issues before they can completely move ahead. If you know the US system, this could be many, many months all by itself.

In addition, there is no completed Mark V circuit (and I suspect no Mark IV material still available) so any future Synchronars will need a new chip with a completed design -- whether it be the completion of the Mark V work that Rogers was doing, or a reverse engineering of the Mark IV. This is a major expense of money and time. But the family is very interested in continuing Roger's legacy.

They ask that everyone be patient during this time. I would recommend that you do mail Custom Circuits with supporting information if you had sent in a watch or money. Please be friendly in your correspondence. Nothing will get resolved quickly and I suspect those who are nice in this situation will be treated better when things finally get resolved. (Or even partially resolved.)

As I've noted, I do not represent the family or have any interest in the Custom Circuits/Synchronar operation. They respect me for my coverage of Roger on my website -- www.ledwatches.net -- and have chosen to share some of their early thoughts with me in order for me to share with fellow collectors.

--Guy

[RaWatch]

You mentioned the watch chip might be created again by "reverse engineering." I am not sure how likely this is. I spoke with Mr. Riehl about how he designed the chips and he explained that he added usless, bogus, red herring portions to his circut designs to confuse anyone trying to reverse engineer. He seemed pleased that no one had done it.

[rewolf]

I'm quite sure Mark IV chip will never be "reverse-engineered". It's simply not worth the effort.
IMHO the easiest way to finish (actually: redesign) the Mark V is to take a modern low-power microcontroller chip, add LED driver transistors, a transistor to control battery charging, write a few hundred lines of software, et voil?: the Mark V. This would take only a few weeks, no need for a special chip.
I guess that's also the way the "Pimp" watches are made. Noone can afford to design a chip only for a few hundred or thousand watches.

[ledwatch]

.
With great respect I disagree, I developed a chip a few years ago using no software but standard cmos gates, I built it up on veroboard to check the operation and after lots of testing I took it to a specilaist chip manfacturer who shrunk it all down and produced a chip 5mm x 3mm in total size. The first batch was fairly expensive per unit item, I ordered an initial 100pcs. However on re-order the price was very much less due to the cad and software to produce them having already been done. I have the circuit diagram to most of the led watches including the Synchronar, albeit not the Mk4 chip, but going to manufacture would be rather inexpensive these days and well within the reach of the small inventor/hobbyist.

Infact I think the most expensive item would probably be the mold to make up the plastic cases. The trouble is that all the idiots in the led world, of which there are a few would then call them '''cheap knock offs'' and would probably use this forum to devalue all the work done and to ruin any prospective market. It was done a few weeks ago after someone had gone to alot of trouble and expense........... :wink:

Sorry, I'll step down from my soapbox.......

[h00hbt]

.
The trouble is that all the idiots in the led world, of which there are a few would then call them '''cheap knock offs'' and would probably use this forum to devalue all the work done and to ruin any prospective market. It was done a few weeks ago after someone had gone to alot of trouble and expense........... :wink:

Sorry, I'll step down from my soapbox.......

Hear, hear.

/H

[rewolf]

@ledwatch: You designed your own LED watch chip - quite impressive. Just curious: 5mm x 3mm, was this the die size or incl. packaging (seems a bit big for a die)?
I also remember times when I made breadboards with 10,20,30 CD40xx CMOS ICs, but I wouldn't do that nowadays.
Today one can get a complete development kit of a suitable microcontroller for US$50-100, including in-circuit debug interface, a prototype board (with space left to add your surrounding circuitry like LED drivers and the display), and software (C-compiler, assembler, linker, high-level debugger). The controller costs $1-3, even in very small quantities.
I'm software-biased, so for me it'd be easier to go this way than to have a chip made: I have the development kit, can get microcontroller samples within a few days (for free), so the only cost is my time . And I can always add new features to my watch. E.g. automatic DST switchover (configurable for different places), temperature-compensation (on-chip temperature sensor), etc.

Probably the display is a problem. AFAIK, noone manufactures the original 7-segment LED watch displays any more.

Do the molds and tools for the original Synchronar still exist?

[ledwatch]

Excellent ideas, a friend of mine once developed an led watch using this method on a pic chip but the problem was that the pic used massive amounts of current compared to cmos gate technology and so battery life was very poor, it could work but what would be the current draw on a microcontroller? And would you need to walk round with a battery pack strapped to your belt??

[rewolf]

I haven't ever used PIC controllers (I simply don't like the architecture), but AFAIK there are also versions with less than 10uA consumption @ 32kHz.

I'd take an MSP430 from TI: MSP430 Overview
It was designed for metering applications, where it has to run 10 years on a single Li battery.

You can let the quartz oscillator clock a timer that wakes the CPU core every second for timekeeping. Buttons use interrupt inputs to wake the CPU. This way you can leave the CPU core switched off most of the time, so current consumption is about 1uA (one microampere!), this is almost below the self-discharge rate of the batteries. It can run on 1.8V to 4V, ideal for 2 alkaline or AgO cells or a single Li. It cannot drive LEDs directly (segments yes, but not the digits), so you'd need at least 4 external digit driver transistors.
That watch would consume less current than other vintage and modern LED watches (my GP takes 5uA, the blue Pimp and the Saishin Diode both take 7uA, display off).
Definitely no need to carry a battery pack :wink:

Here is a very good Yahoo group dedicated to the MSP430, and a link to a thread about other microcontrollers with low current consumption: http://groups.yahoo.com/group/msp430/message/14125

[richard_uk]

This is an English speaking forum, could you please stick to speaking (typing) English

[fronzelneekburm]

hmmm, somehow I feel terribly stupid while reading this thread...

[Synchroserious]

.
The trouble is that all the idiots in the led world, of which there are a few would then call them '''cheap knock offs'' and would probably use this forum to devalue all the work done and to ruin any prospective market. It was done a few weeks ago after someone had gone to alot of trouble and expense........... :wink:

Sorry, I'll step down from my soapbox.......

How about 4o years of trouble and expence Phil?