Hello List.
Today in our weekly Mumble Session we talked about a smaller chip, which could be realized before the Northpoint.
The Reason is mostly, that the Northpoint (as an small MCU) - beside the Standard Cells - also has some challenging Analog Stuff (like the ADC, Pad-Cells) which set the hurdle a little bit high.
We need some physical haptic fancy / shiny silicon before which gives us also a realistic way to verify our simulation after model measurement we do with PearlRiver. Something we could hand over to other people and which blinks..
Well, we came across the well-known "triple-5 chip". The original NE555 [0] was developed by the Swiss electronics engineer Hans Camenzind [1] and get cloned by nearly all silicon firms on the market in this era. Great starting points for our CMOS-based clone are the wikipedia site and the free eBook [2] of the original author. I strongly recommend everybody to read this great book!
Mostly the 555 chips are in the -55 to +125 C temperature range - the same we assume for our technology. And the voltage supply goes up to +18 Volt, I already saw 555 which worked with higher voltage. If we like, we could using our BiCMOS feature for the output driver - driving the same current with less power consumption.
The 555 chip has a legendary status between electronic engineers, radio amateurs and tinkers. And some of them writing obeisance on hackaday.com still in 2018 [3]. Many variations of the chip are still in production. Short - the 555 would be a great try-out for LibreSilicon! And I am proud to announce that Ferenc will be our Lead for the "LS 555" (LibreSilicon triple-5). Thanks Ferenc for your commitment!
Regards, Hagen.
[0] https://en.wikipedia.org/wiki/555_timer_IC [1] https://en.wikipedia.org/wiki/Hans_Camenzind [2] http://www.designinganalogchips.com/ [3] https://hackaday.com/2018/10/10/the-555-and-how-it-got-that-way/
Hi Hagen, we could make a combo chip: NE555, some LM741, and some CD4066 (transfer gate) UC3845, UC3843 (simple PWM chip for switch mode power supply) So geeks could combine the chips using transfer gates :-)
Or lets think about a simple analog geeky chip, that contains: saw tooth generator, comparator, rs latch, error-amps (UC3845) while the the saw tooth generator is replaced with a NE555 (transfer gate, or just fitting the ne 555 instead of the internal uc3845 saw tooth generator) The Op-Amp in the UC3845 may be replaced with a 741 or similar simple op amp.
So we have an universal circuit containing: -OPAMPS (741, LM324 or similar) -universal SAWTOOTH-Generators (NE555) -RS-Latch (NE555 also has a latch) for use in fake UC3845 -the rest used in UC3845
the transfer gates are used as pin mux so we can access internal stuff of the modified UC3845 and split it into, OP-AMPS, NE555 and so on.
So having a programable simple analog circuit would be cool. Also a TL431 programable z-diode woud be cool as well as TAA 761 open collector op-amp. So here we could combine TL431 with TAA761 open collector op amp.
We would have an analog "fpga"
What do you think?
Cheers,
Ludwig
On Sun, Mar 3, 2019 at 2:13 PM Hagen SANKOWSKI hsank@posteo.de wrote:
Hello List.
Today in our weekly Mumble Session we talked about a smaller chip, which could be realized before the Northpoint.
The Reason is mostly, that the Northpoint (as an small MCU) - beside the Standard Cells - also has some challenging Analog Stuff (like the ADC, Pad-Cells) which set the hurdle a little bit high.
We need some physical haptic fancy / shiny silicon before which gives us also a realistic way to verify our simulation after model measurement we do with PearlRiver. Something we could hand over to other people and which blinks..
Well, we came across the well-known "triple-5 chip". The original NE555 [0] was developed by the Swiss electronics engineer Hans Camenzind [1] and get cloned by nearly all silicon firms on the market in this era. Great starting points for our CMOS-based clone are the wikipedia site and the free eBook [2] of the original author. I strongly recommend everybody to read this great book!
Mostly the 555 chips are in the -55 to +125 C temperature range - the same we assume for our technology. And the voltage supply goes up to +18 Volt, I already saw 555 which worked with higher voltage. If we like, we could using our BiCMOS feature for the output driver - driving the same current with less power consumption.
The 555 chip has a legendary status between electronic engineers, radio amateurs and tinkers. And some of them writing obeisance on hackaday.com still in 2018 [3]. Many variations of the chip are still in production. Short - the 555 would be a great try-out for LibreSilicon! And I am proud to announce that Ferenc will be our Lead for the "LS 555" (LibreSilicon triple-5). Thanks Ferenc for your commitment!
Regards, Hagen.
[0] https://en.wikipedia.org/wiki/555_timer_IC [1] https://en.wikipedia.org/wiki/Hans_Camenzind [2] http://www.designinganalogchips.com/ [3] https://hackaday.com/2018/10/10/the-555-and-how-it-got-that-way/ _______________________________________________ Libre-silicon-devel mailing list Libre-silicon-devel@list.libresilicon.com http://list.libresilicon.com/mailman/listinfo/libre-silicon-devel
TAA761 and other historic classic chips in german http://www.ic-ts-histo.de/fad/ics/taa761/taa761fa.htm
On Sun, Mar 3, 2019 at 11:10 PM ludwig jaffe ludwig.jaffe@gmail.com wrote:
Hi Hagen, we could make a combo chip: NE555, some LM741, and some CD4066 (transfer gate) UC3845, UC3843 (simple PWM chip for switch mode power supply) So geeks could combine the chips using transfer gates :-)
Or lets think about a simple analog geeky chip, that contains: saw tooth generator, comparator, rs latch, error-amps (UC3845) while the the saw tooth generator is replaced with a NE555 (transfer gate, or just fitting the ne 555 instead of the internal uc3845 saw tooth generator) The Op-Amp in the UC3845 may be replaced with a 741 or similar simple op amp.
So we have an universal circuit containing: -OPAMPS (741, LM324 or similar) -universal SAWTOOTH-Generators (NE555) -RS-Latch (NE555 also has a latch) for use in fake UC3845 -the rest used in UC3845
the transfer gates are used as pin mux so we can access internal stuff of the modified UC3845 and split it into, OP-AMPS, NE555 and so on.
So having a programable simple analog circuit would be cool. Also a TL431 programable z-diode woud be cool as well as TAA 761 open collector op-amp. So here we could combine TL431 with TAA761 open collector op amp.
We would have an analog "fpga"
What do you think?
Cheers,
Ludwig
On Sun, Mar 3, 2019 at 2:13 PM Hagen SANKOWSKI hsank@posteo.de wrote:
Hello List.
Today in our weekly Mumble Session we talked about a smaller chip, which could be realized before the Northpoint.
The Reason is mostly, that the Northpoint (as an small MCU) - beside the Standard Cells - also has some challenging Analog Stuff (like the ADC, Pad-Cells) which set the hurdle a little bit high.
We need some physical haptic fancy / shiny silicon before which gives us also a realistic way to verify our simulation after model measurement we do with PearlRiver. Something we could hand over to other people and which blinks..
Well, we came across the well-known "triple-5 chip". The original NE555 [0] was developed by the Swiss electronics engineer Hans Camenzind [1] and get cloned by nearly all silicon firms on the market in this era. Great starting points for our CMOS-based clone are the wikipedia site and the free eBook [2] of the original author. I strongly recommend everybody to read this great book!
Mostly the 555 chips are in the -55 to +125 C temperature range - the same we assume for our technology. And the voltage supply goes up to +18 Volt, I already saw 555 which worked with higher voltage. If we like, we could using our BiCMOS feature for the output driver - driving the same current with less power consumption.
The 555 chip has a legendary status between electronic engineers, radio amateurs and tinkers. And some of them writing obeisance on hackaday.com still in 2018 [3]. Many variations of the chip are still in production. Short - the 555 would be a great try-out for LibreSilicon! And I am proud to announce that Ferenc will be our Lead for the "LS 555" (LibreSilicon triple-5). Thanks Ferenc for your commitment!
Regards, Hagen.
[0] https://en.wikipedia.org/wiki/555_timer_IC [1] https://en.wikipedia.org/wiki/Hans_Camenzind [2] http://www.designinganalogchips.com/ [3] https://hackaday.com/2018/10/10/the-555-and-how-it-got-that-way/ _______________________________________________ Libre-silicon-devel mailing list Libre-silicon-devel@list.libresilicon.com http://list.libresilicon.com/mailman/listinfo/libre-silicon-devel
something about op amps and cascode internal circuits in german: http://www.ic-ts-histo.de/anim/lina/lina.htm#anf
On Sun, Mar 3, 2019 at 11:11 PM ludwig jaffe ludwig.jaffe@gmail.com wrote:
TAA761 and other historic classic chips in german http://www.ic-ts-histo.de/fad/ics/taa761/taa761fa.htm
On Sun, Mar 3, 2019 at 11:10 PM ludwig jaffe ludwig.jaffe@gmail.com wrote:
Hi Hagen, we could make a combo chip: NE555, some LM741, and some CD4066 (transfer gate) UC3845, UC3843 (simple PWM chip for switch mode power supply) So geeks could combine the chips using transfer gates :-)
Or lets think about a simple analog geeky chip, that contains: saw tooth generator, comparator, rs latch, error-amps (UC3845) while the the saw tooth generator is replaced with a NE555 (transfer gate, or just fitting the ne 555 instead of the internal uc3845 saw tooth generator) The Op-Amp in the UC3845 may be replaced with a 741 or similar simple op amp.
So we have an universal circuit containing: -OPAMPS (741, LM324 or similar) -universal SAWTOOTH-Generators (NE555) -RS-Latch (NE555 also has a latch) for use in fake UC3845 -the rest used in UC3845
the transfer gates are used as pin mux so we can access internal stuff of the modified UC3845 and split it into, OP-AMPS, NE555 and so on.
So having a programable simple analog circuit would be cool. Also a TL431 programable z-diode woud be cool as well as TAA 761 open collector op-amp. So here we could combine TL431 with TAA761 open collector op amp.
We would have an analog "fpga"
What do you think?
Cheers,
Ludwig
On Sun, Mar 3, 2019 at 2:13 PM Hagen SANKOWSKI hsank@posteo.de wrote:
Hello List.
Today in our weekly Mumble Session we talked about a smaller chip, which could be realized before the Northpoint.
The Reason is mostly, that the Northpoint (as an small MCU) - beside the Standard Cells - also has some challenging Analog Stuff (like the ADC, Pad-Cells) which set the hurdle a little bit high.
We need some physical haptic fancy / shiny silicon before which gives us also a realistic way to verify our simulation after model measurement we do with PearlRiver. Something we could hand over to other people and which blinks..
Well, we came across the well-known "triple-5 chip". The original NE555 [0] was developed by the Swiss electronics engineer Hans Camenzind [1] and get cloned by nearly all silicon firms on the market in this era. Great starting points for our CMOS-based clone are the wikipedia site and the free eBook [2] of the original author. I strongly recommend everybody to read this great book!
Mostly the 555 chips are in the -55 to +125 C temperature range - the same we assume for our technology. And the voltage supply goes up to +18 Volt, I already saw 555 which worked with higher voltage. If we like, we could using our BiCMOS feature for the output driver - driving the same current with less power consumption.
The 555 chip has a legendary status between electronic engineers, radio amateurs and tinkers. And some of them writing obeisance on hackaday.com still in 2018 [3]. Many variations of the chip are still in production. Short - the 555 would be a great try-out for LibreSilicon! And I am proud to announce that Ferenc will be our Lead for the "LS 555" (LibreSilicon triple-5). Thanks Ferenc for your commitment!
Regards, Hagen.
[0] https://en.wikipedia.org/wiki/555_timer_IC [1] https://en.wikipedia.org/wiki/Hans_Camenzind [2] http://www.designinganalogchips.com/ [3] https://hackaday.com/2018/10/10/the-555-and-how-it-got-that-way/ _______________________________________________ Libre-silicon-devel mailing list Libre-silicon-devel@list.libresilicon.com http://list.libresilicon.com/mailman/listinfo/libre-silicon-devel
On Sun, Mar 3, 2019 at 8:08 PM ludwig jaffe ludwig.jaffe@gmail.com wrote:
Hi Hagen, we could make a combo chip:
[...]
What do you think?
WAY too ambitious. The [7]555/6 is a well-known standard design, so it's mostly a proof-of-concept for the tool chain used.
It's an excellent idea to start with small steps.
What tool chain does Ferenc want to use, anyhow?
tatzelbrumm <- hanging out in Gert's institute for another week
Cheers,
Ludwig
On Sun, Mar 3, 2019 at 2:13 PM Hagen SANKOWSKI hsank@posteo.de wrote:
Hello List.
Today in our weekly Mumble Session we talked about a smaller chip, which could be realized before the Northpoint.
Well, we came across the well-known "triple-5 chip". The original NE555 [0] was developed by the Swiss electronics engineer Hans Camenzind [1] and get cloned by nearly all silicon firms on the market in this era. Great starting points for our CMOS-based clone are the wikipedia site and the free eBook [2] of the original author. I strongly recommend everybody to read this great book!
Hello Ludwig.
Imagine you are like to build a "many-lane" bridge over a river..
1. You could start with all lanes at the same time. This will cost you many effort, you need support constructions for every lane and the risk is quite high to fail at all.
2. You could start with a small bridge for pedestrians. The support construction is not so difficult and you're able to open this lane soon.
Later, you'll take the support constructions and setting up the first street lane. After opening the pedestrian bridge you're now proud to open the first real lane, your picture hits the news again.
Go ahead with the second lane, and so on. If you'd fail on the way, the bridge is already there and (partly) functional.
IMHO the second approach is the better one.
The 555 is quite known. Our effort is just "to port" the design to the LibreSilicon Technology.
Of course you are free to use the technology to design and build your own chips. I like to see your secure smart cards soon! :-)
We need a proof, that we are able to deal with the technology. And this impressive try-out for all the haptic white-collar professionals has to be as simple as possible.
Regards, Hagen.
libresilicon-developers@list.libresilicon.com
-
Christoph Maier -
Hagen SANKOWSKI -
ludwig jaffe