LM211 internal hysteresis?

I'm just going by my own subconscience statistics but I'd say most comparators have less than 40mV of internal hysteresis and guess there's a good chance you'll need to add hysteresis either by pin control (special comparator) or by external resistors.

D from BC

There isn't any "internal hysteresis" in the LM311.

As far as I know my MC1650 design was the first comparator to feature internal hysteresis (which could widen to effect a latch function)...

...Jim Thompson

Of couse, the Am685 did it better - it's latch worked right at the input, with correspondingly smaller clock set-up and aperture times, whch is why the Am685 is the device that was upgraded into the various

Jim never does seem to get it quite right - he'd be the Linus Pauling of electronics.

-- Bill Sloman, Nijmegen

Yes, you'll need to apply some hysteresis. Not much in your case.

An LM311 is butt slow. So applying hysteresis via positive feedback from the output may cause grief depending on signal speed.

...Jim Thompson

Hey Jim, How's the hips doing ? Are you well enough to do some break dancing with the young guys yet? :)

Not quite yet. I'm restricted to 90° flex, no crossing legs, etc.

...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 | |

Hi,

In article , To-Email-Use-

Can you suggest another "jelly-bean" comparator that's sourced by more than one company? National, TI and ADI all make some modern fast comparators, but they're all proprietary and expensive.

I'm shooting for ~50-100 ns response time from the comparator, 0 to VCC input range, single-supply...

Rob.

Say, if they put in unusually compliant joints, could you get into contortionism? Or is that more limited by ligaments? Hmm...

Tim

-- Deep Fryer: A very philosophical monk. Website @

An LVDS-to-ttl receiver is a pretty good comparator with prop delays in the few-ns range, blinding fast (like 600 ps) active pullup/pulldown output. Cheap, too.

John

This is an interesting idea...

Is there a specific part you have used with good success? I've found single LVDS-to-3.3V CMOS output parts from TI and Fairchild but I'm concerned about:

1. Input voltage range
2. Hysteresis

One input to comparator will be fixed at 2.5V, the other can take anything from 0 to VCC. I'm concerned about reliable operation across that range given LVDS usually has a 1.2V offset with +/- 200 mV swing.

Also, do you have any comments on what kind of hysteresis you've seen? I can't seem to find any SPICE models, just IBIS .

Thanks Rob.

National DS90LV018, DS90C402, TI SN65LVDS2DVBR. Fairchild's lvds-lvds FIN1101 is good at 2.5 volts. Most work well from ground (maybe below a little?) to at least 2.5 volts input range. Spartan3 FPGA inputs seem to be good rail-to-rail-input comparators.

Most lvds things have a bit of deliberate input offset. I don't know if they have hysteresis.

John

The 1.2V "offset" is the center-point of the Differential OUTPUT.

Has nothing to do with the input.

...Jim Thompson

I know the "offset" is what appears on a traditional LVDS OUTPUT.

The point was, I was uncertain how the LVDS INPUT would behave if it did not get a signal with the typical "offset".

Rob.