Favorite precision rectifier

Yep, but it's becoming hard to get.

Cheers

Phil Hobbs

The big problem with the 1496 is that its output is a differential between two points fairly far from ground. If you're building a lock-in, one good method for getting round this is to use a single cap between the (collector) outputs, and a switched-capacitor block to transfer the difference voltage down to ground. Since there's only one cap, there's no opportunity for the two sides to have different bandwidths and phase shifts, which helps. I used three of those in a little non-contact head tracker about 15 years ago. Nice gizmo.

Cheers

Phil Hobbs

Cripes, if you need a few they are 3 cents each?

US$70+ship buys 2500 SOIC on tape/reel?

About the cost of a cheap BJT.

Jon

TO-5 or DIP? ;-)

I've used PDIPs and SO-14s. 2N3904s were invented a long time ago too.

Cheers

Use high speed comparitors instead of slow op-amps, for starters...

On Jun 18, 2:00=A0pm, "Tim Williams" wrote: > What's your favorite full wave precision rectifier circuit? >

See page 19, figure 12 of the Analog Devices AD8036 datasheet.

G=B2

AAMOF, I have a few LM1496S (TO-5 style!) in my parts box. Removed from some TV modulator equipment. Some CA3026s and TL442s, too.

Tim

Intristing.

Sounds horrendously expensive (well, of course it is, it's AD).

Tim

-- Deep Friar: a very philosophical monk. Website:

See page 19, figure 12 of the Analog Devices AD8036 datasheet.

G²

On a sunny day (Sat, 20 Jun 2009 00:20:42 -0700 (PDT)) it happened snipped-for-privacy@yahoo.com wrote in :

Nice chip :-)

I see it listed at 12 of the 21 distributors indexed by

It isn't too much, but when these old chips start to increase in price, you have to start being careful, IME. Of course it may be that the rest of the world discovered their extreme usefulness and drove up the price.

Cheers

Phil Hobbs

And single source, _top_poster_ with misplaced separator :-(

...Jim Thompson

If you had simulated that circuit you would have seen that "undershoot".

If you had simulated my submission, you'd still see some "undershoot", though significantly less.

There's a specific reason for that. Study both versions and try to discern the difference... Left as an exercise for the student ;-)

...Jim Thompson

o 100kHz

step at

ext -

t -

Thanks Phil, I=92ve no foreseeable project requiring a lockin, but was only asking out of interest.

=93>Since there's only one

I=92d never thought of the uniform bandwidth and phase shift advantage of switched capacitors. (Versus the paired RC=92s I=92ve used and always lamented the different time constants.)

George H.

Seems to be a mistake- shows 0.38 above as price in multiples of 2500 (sometimes). Digikey 0.383, Avnet 0.364 in 1-reel lots.

Best regards, Spehro Pefhany

ep at

=A0 =A0 ...Jim Thompson

=A0 | =A0 =A0mens =A0 =A0 |

=A0 | =A0 =A0 et =A0 =A0 =A0|

=A0|

=A0 =A0 =A0 |

oted text -

Thanks Jim! I'll include a few in the next order I place.

George H.

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Mmm?

stratus' post didn't autoindent for some reason. I can't be bothered to edit it. Short, general comments go at the top. Easy to understand. My sig was correctly placed, and correctly formatted. That leaves one thing, top poster prejudice. In that case, go f*ck off and die, I could care less.

Tim

Er well, that's fairly obvious. It goes through two op-amps.

Tim

-- Deep Friar: a very philosophical monk. Website:

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...