Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!eternal-september.org!.POSTED!not-for-mail
From: piglet <erichpwagner@hotmail.com>
Newsgroups: sci.electronics.design
Subject: Re: comparator with two analog switches
Date: Fri, 25 Apr 2025 21:38:48 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 36
Message-ID: <vugvd8$smnp$1@dont-email.me>
References: <87r01gqc76.fsf@librehacker.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Injection-Date: Fri, 25 Apr 2025 23:38:49 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="44e385d80862c49a1da44bf180baf4c4";
	logging-data="940793"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+lOP0csfpI5Abh75oepQw+"
User-Agent: NewsTap/5.5 (iPhone/iPod Touch)
Cancel-Lock: sha1:dabOV1h02NcEHXama2gcBOXiJKM=
	sha1:aYhbbf5B3s2sS8lqrV+bHqy3CL0=
Bytes: 3100

Christopher Howard <christopher@librehacker.com> wrote:
> Hi, I am building a little 70's style analog computer, mainly for
> learning purposes. For the most part I've been able to follow the
> Grappendorf project schematics, but that project does not include
> schematics for a comparator. I looked at the Anabrid THAT design
> schematics, and I was going to make something based off that; however,
> the Anabrid design uses Schmitt-Trigger inverters, which requires 5V
> supply, and I was hoping not to have to mess around with 5V, everything
> else in my system requiring only a +/- 15V. So, I was wondering if I
> maybe I could get rid of the Schmitt-Trigger inverters from the design
> and get by with just use op amps instead, like so:
> 
> http://gem.librehacker.com/gemlog/starlog/img/2025/20250425_comparator.jpg
> 
> But when I try to build this, I've seen various strange results from the
> switches, like it getting stuck with seemingly one switch stuck open and
> one stuck close. I've done various bits of troubleshooting, but haven't
> got to the bottom of it yet. Is there something fundamentally wrong with
> the design? Maybe the Schmitt-Triggers are critical for some reason...?
> (Hysteresis?) I wonder if maybe I might have damaged the switches, so
> that might be it, but I can't afford to order more right now.
> 

Those DG212 switches have cmos control inputs which need driving to 0v and
5V and not allowed to float to intermediate levels. Those diodes prevent
the inputs being driven negative but do not stop the input floating higher
so really need a pulldown resistor. Alternatively just omit them and ensure
the series resistor is high enough that when the op amps outputs hit either
supply rail the current is limited through the DG212 input protection
diodes. The device needs a 5V logic level supply but the current demand is
very low and a simple resistor dropper from the system 15v rail with 5.1v
zener should be enough. Comparators nearly always benefit from having
hysteresis even if only a tiny bit. Not all op amps work as comparators.

-- 
piglet