Comparator VS Amplifier

Sort of. The comparator actually has a finite (but very high) gain like the opamp, so it really has the same static gain equation. The big difference is that the opamp has the right internal dyamics (frequency response, time lags) that it can be used in stable negative feedback loops, whereas comparators are just designed to be flat-out fast.

But for working purposes, your defs are OK.

There are some combo chips, but they tend to be slow.

The best-kept secret around is that LVDS-to-CMOS line receivers make very fast, cheap comparators.

What are you working on? Details help.

John

ps - ignore Fred. He's in a bad mood this morning.

Who said "Never use a comparator as an opamp, and never use an opamp as a comparator."? Somebody semi-famous, I think.

John

Yes.

I believe that National makes a part that contains one section of an LM339 and one section of LM324. Search their site.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

Right. They're usually spec'd for 0.2 to 2.2 or some such; most seem to work well from a tad below ground to maybe +2.5. The ones we've tried work nicely comparing against ground.

A nice external-clock-input circuit is a series R, a resonant LC to ground, and an LVDS receiver. That tolerates all sorts of ringing and crud on the input.

John

Can't be me... I'm INFAMOUS ;-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

Probably someone who sold both op-amps and comparators. ;-)

LM392. But of course it's more expensive than an LM324+LM339 (almost twice the price of the pair) and has 1/4 the capability.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Just in case, the commonmode range of these is rather small.

Rene

LM392 is second-sourced by a number of companies.

RL

You have to be careful about taking the input of an ECL circuit up to ground (assuming that you are usuing the standard power supply rails at

0V and -4.5V (or -5.2V, -3.3V or -2.5V or whatever the flavour of the decade is).

With some ECL IC's this messes up the internal biassing. IIRR one of the ECLinPS data books had a whole chapter on the subject with a one page list of parts classified according to what you could get away with, mainly in the context of what you did with unused inputs.

Needless to say, I can't find it on the web-site

----------- Bill Sloman, Nijmegen

I was actually suggesting something like

or maybe the 5-v version

These work from a bit below ground to +2.5 maybe. Output risetimes are around 500 ps and they're dirt cheap. Input offsets aren't as good as a bipolar comparator, but they seem very stable once that's nulled out. Lots of people make this stuff... TI, On, National, Fairchild.

John

What parts do you refer to? A DS90LV028A dual LVDS receiver has 2.4 ns max prop delay and costs $1.12, 56 cents per comparator.

We've just almost finished replacing about 2500 defective MAX9690 comparators in units in the field. They fail at high temperature and, over time, the failure temp drifts steadily down...

John

I used to think that too, but take a look at Maxim's site. They have comparators which are faster, better and cheaper (at least in small quantities) than LVDS receivers.

--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

The 9201 looks like a good deal; quad 7 ns for $2.25. But a bit slow for my taste. The LVDS is about as fast a "comparator" as there is if you still want CMOS output levels.

I bet the LVDS inputs of an FPGA would make interesting comparators.

This is fierce:

but ECL.

Yeah. They knew they had a problem and sampled us 2500 replacements. But we still had to do the work.

John

MAX9201 versus DS90C032. The main reason I choose the 'real' comparator for a project is that I'm not comfortable with the limited _specified_ input range and the rather high threshold.

Meanwhile I also noticed the faster LVDS parts are way cheaper than the slower ones. Buy two Ferrari's for the price of Honda Civic.

Did you get genuine parts? Maxim warns for counterfeit parts. We have had similar problems with Maxim clones from Sipex (branded as Sipex though).

--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl