X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
X-Original-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=gmail.com; s=20120113;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        bh=GTjRET2y+iaeHw14c9Cp6UWiTUrZRm2G1/k98dAzNbw=;
        b=ylL8MD3XKrkD6Be+bbLIL+fN+73WcMkfMa+RW03k4l2y2Nk9vveeVs4vgCnw8E1GuE
         9lmMWc01bxmmaCndcCviLsIo3GZOIVJhpuOFYhUwpcmPzuoaZ8fXoTfZBjm0lEfLRezw
         8gIGxfSyhgsuBsxknq70RGeMhpKiuYsJbKiFj1MAqEIYEjQOdNCxAU0l9fdYK8eiH8by
         5++RFOrD4/pGkOKHeB5xYe/20WcSM3hP9nFaefjkUN+8VMD+NBva+c3IjJqvQ+jW3oF7
         33d4YeiXjyHPtYsvYepjyqCvTuN7tbT/mAXDsF1tg4wXViH+T28O9abNIZyk2DFdI6K5
         cLGQ==
MIME-Version: 1.0
X-Received: by 10.25.20.24 with SMTP id k24mr9393609lfi.117.1445728687689;
 Sat, 24 Oct 2015 16:18:07 -0700 (PDT)
In-Reply-To: <20151024225457.GA25400@visitor2.iram.es>
References: <CAM2RGhThg+5B6iGQ__Uh4=MxPeSVb9p1gS_9C+ik--E8AqS3GA AT mail DOT gmail DOT com>
	<CAOuGh89DAYWn49KnuqHO6NHcyB3GUoVsPD4n8FR=wcg8AY6wYw AT mail DOT gmail DOT com>
	<201510162117 DOT 12255 DOT ad252 AT freeelectron DOT net>
	<CAM2RGhSXgieJ+p5i_nQsAUMRQh1rp1PXHx_jB5p2XECY6jgjQQ AT mail DOT gmail DOT com>
	<20151023070213 DOT GA19358 AT visitor2 DOT iram DOT es>
	<CAM2RGhS4=PPXf1L5a_E9uABrLviY+L8n+0Opgi6_ZGnipi4grg AT mail DOT gmail DOT com>
	<CAM2RGhRYR61dADi=mRN5CuT_wNMZqc62CUJawBHCOmAaGBD+Qw AT mail DOT gmail DOT com>
	<20151023212815 DOT GA31581 AT visitor2 DOT iram DOT es>
	<CAM2RGhQvNr1B4um1y6DAeaWgTPugBCWiNt6Z4eOvdYCK1j1SrQ AT mail DOT gmail DOT com>
	<20151024225457 DOT GA25400 AT visitor2 DOT iram DOT es>
Date: Sat, 24 Oct 2015 23:18:07 +0000
Message-ID: <CAM2RGhRChUp6hAyHhKQjU26kJ=nL4HxF_WoYovBUku2bSsrMqA@mail.gmail.com>
Subject: Re: [geda-user] Stable and low noise references (was: The
 Chippocolypse is coming!)
From: "Evan Foss (evanfoss AT gmail DOT com) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
To: gEDA users mailing list <geda-user AT delorie DOT com>
Content-Type: text/plain; charset=UTF-8
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On Sat, Oct 24, 2015 at 10:54 PM, Gabriel Paubert (paubert AT iram DOT es)
[via geda-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:
>
>
>     Hi Evan,
>
> On Sat, Oct 24, 2015 at 06:15:22AM +0000, Evan Foss (evanfoss AT gmail DOT com) [via geda-user AT delorie DOT com] wrote:
>> On Fri, Oct 23, 2015 at 9:28 PM, Gabriel Paubert (paubert AT iram DOT es)
>> [via geda-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:
>> > On Fri, Oct 23, 2015 at 09:25:52AM -0400, Evan Foss (evanfoss AT gmail DOT com) [via geda-user AT delorie DOT com] wrote:
>> >> On Fri, Oct 23, 2015 at 9:16 AM, Evan Foss <evanfoss AT gmail DOT com> wrote:
>> >> > On Fri, Oct 23, 2015 at 3:02 AM, Gabriel Paubert (paubert AT iram DOT es)
>> >> > [via geda-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:
>> >> >> For example, and as far as I can tell, Linear Technology has eliminated
>> >> >> all TO-99 versions except the LTZ1000A, where the metal can is superior
>> >> >> for reasons related to thermal insulation and mechanical stress on the die.
>> >>
>> >> Sorry I miss read this.
>> >>
>> >
>> > Thanks for the second reply, I couldn't understand the first one.
>> >
>> >> > That is bizzar. I get that people hate paying for TO-99 stuff but a
>> >> > reference really should be in that package. I have to wonder if they
>> >> > are just switching people over to something newer that is available in
>> >> > a TO-99. LT makes a few different references.
>> >
>> > This is becoming offtopic, but the LTZ1000A is, to my knowledge, still
>> > the best reference available in terms of noise and long term stability,
>> > and until a few years ago, I only used buried Zener references when I
>> > needed high stability.
>>
>> It is. If you look at any of the tear downs people do of meters above
>> 6 digits they are all using that reference or one made by LT custom
>> for Fluke.
>
> I was not aware of that, but this means that this device has cornered a
> market from people accepting to pay for the performance and might not be
> EOLed too quickly.

Yes. Although the benchtop meter market is changing because...
Danaher Corporation owns Tektronix, Fluke, and Keithley Instruments.
Open up a Keithley meter and you will find a Fluke reference. For
those of us who worry about being able to have diversity to cross
check things it is a bad scene.

>>
>> > But then, Linear started to produce a line of low noise bandgap
>> > references that are fairly close in performance to buried Zeners,
>> > especially the LTC6655, which is better than most buried Zeners.
>>
>> I will read up on those. That sounds cool.
>
> It is an excellent reference, and it is much easier to use than the
> LTZ1000A, which needs to be surrounded by a significant amount of
> carefully layed out analog circuitry.

True that and the massive price difference make it very attractive.

If you are using a reference like the LTZ1000A then you are probably
doing a lot of delicate analog layout anyway.

>>
>> > This said, I use more LT6654 than LTC6655, it's clearly worse in the low
>> > frequency noise range, but sufficient for most of my needs and much more
>> > robust on the input (38V max vs 13.2).
>> >
>> > Both the LTC6655 and LT6654 are available in SMD plastic and ceramic
>> > package, but I've only used the plastic ones. The ceramic package
>> > has stability advantages over long term and temperature cycling
>> > which are of no interest in my applications.
>>
>
> The main reason I've not systematically used the LTC6655 is that it
> would require an additional linear regulator because I don't have
> a power supply rail in the suitable range.

+1

> While it's limited to Linear products, http://www.linear.com/docs/29441
> clearly shows which reference to choose when 0.1 to 10Hz noise is
> important. With 0.25 ppm, the LTC6655 is about 1.5 times the value of
> 0.17ppm given for the LTZ1000A, all the others are at least 0.6ppm,
> 3.5 times higher than the LTZ100A.
>
> Before the 6654/6655, the best compromise for my applications was often
> the LT1021, and I only sweared by buried Zeners, thinking that bandgaps
> would never give the required noise. Well, I was wrong, but it took
> about 20 years.

I think it was Bob Pease who predicted this would happen. I remember
reading about how a green LED was actually a reasonable bandgap
reference for cheap stuff.

>> Nice
>>
>> >     Gabriel
>>
>> Thanks for this it actually is useful in something I am playing with now.
>>
>
> Glad to help, but we've steered way off-topic. To be honest, the
> original thread was not really geda specific either.

I know but I felt like people should know. It was bogus how TI
announced it to some customers ahead of others.

>     Gabriel

PS: With Le Grand K being retired in 2018 there will be 2 things that
will happen
1. I may need to get my meter recalibrated
2. I will be going to visit friends in Paris and trying to get the
first selfie with it. Why should the iphone goofballs get all the fun
of camping outside in urban environments?

Evan

-- 
Home
http://evanfoss.googlepages.com/
Work
http://forge.abcd.harvard.edu/gf/project/epl_engineering/wiki/