I'm available for projects

Give ubeam a call. Might not stretch to July, but they pay well.

He's got time to take a ride. I think you got the gist of my comment.

The classic: says on the side of the truck, "Warning: contains political promises."

Tim

You have a granddaughter with that circuit lab kit. Eventually you will explain things. Just think about how you will do that.

Or occupy your time by riding a bike over the Himalayas like Pease did.

You could tell LT how to design low voltage PSU input protection at a reasonable price. I'm looking for a part that's basically a combination of a few LT parts:

Surge stopper and reverse input protection.

Ideal diode.

Hot swap controller.

Basically I'm looking for a $2 14 pin TSSOP or similar that drives two external back-to-back N-ch MOSFETs to protect the input of a DC/DC converter from an input +12V (NOT automotive) or +24V supply.

It needs only a small number of external components - just the two FETs, some resistors to determine threshold voltages, and a cap or two.

It can protect against:

- reverse input to -50V without needing a lot of external components.

- input overvoltage to +50V, whilst clamping the output to some voltage that I specify using two resistors.

- overcurrent protection (with timer for circuit breaker (and retry)).

- Power (or perhaps SOAR) estimation for external hotswap FET, with shutdown (and retry).

- overtemperature

Features:

- accurate UVLO, programmed by a resistive divider.

- enable input

- input current during shutdown or UVLO < 10uA.

- ideal diode (allowing two DC inputs to be combined - there is no requirement for current sharing though).

- current limit (will need Rsense - extra marks for using FET Rds(on)).

- current sense has very low offset voltage, to avoid the need to waste power.

- programmable output slew control during startup.

- inrush limit (either by slew x output cap, or by current limit).

- power good output, to enable downstream DC/DC converter.

- charge pumps can drive 10V Vgs for Vin > 5V.

- zener clamps on Vgs.

- Draws only tiny current from output (uA) when Vout = 24V and Vin < Vout. (I'm looking at you, LM5050.)

If I make all that out of LT parts, it uses too much space and far too many $. If I make it out of discretes, it uses too much space. Other manufacturers have parts that come tantalisingly close, but don't quite do everything I want, or blow up (or only handle 0.3V of reverse input), or draw too much current.

Example circuit:

Vin FET1 FET2 Vout to load

----+------+---+----+---+ +-+---+---+-+ +---+----+---+------+----- | | | | | | | | | | | | | | | | | | | ----- | ----- | | | === big bidir | | | | | | | | | | cap TVS | | +---------------------------+ | | gnd | | | |sns g1 sns g2 sns| R | | | | | | | | gnd R +--|Vin power sns|--+ R | | | | +------|UVLO OVLO/clamp|------+ | | | | R | | R | | timer gnd en pg | | gnd +---------------------------+ gnd | | | | === gnd | | | | +-----------> gnd >-----+

I don't really mind whether the ideal diode FET or the hotswap FET comes first. Perhaps the ideal diode FET (with its source connected to Vin) would be easier as less of the circuit would see -50V.

Thanks, Allan

Grand_son_... 8 years old... the same one who's on the All Star Inline Hockey Team and was just awarded a trophy for best defensive player of the whole league ;-) ...Jim Thompson

Mike Engelhardt doesn't take well to suggestions. I sat next to him at lunch about a year ago at his seminar in Tempe, AZ. When I brought up flaws in LTspice he reacted negatively.

Right now there's an issue on the LTspice List where Mikey treats a .IC=0 on a capacitor as shorting it out for a .DC analysis... totally out of step with every other simulator except MicroCap. Mikey is resisting changing that.

Have you checked out my "OverAndReverseVoltageProtection.pdf" on the S.E.D/Schematics Page of my website, as a starting point for your needs? ...Jim Thompson

I would be miffed if I was addressed as the above paragraph suggests, but no matter.

I admit that I never do DC analysis and only rarely resort to using .IC directives. I always go straight to transient analysis. Why is initially shorting the capacitor different from forcing .IC=0?

Jeroen Belleman

Well- what kind of mental midget even includes a cap in a DC analysis circuit? Caps are open and inductors are shorts.

(1) Capacitors have their own built-in initial condition setting.

(2) .IC is only an _initial_condition_ for .TRAN, not .DC, for most simulators.

(3) Why don't you do any .DC analyses? It's useful for evaluating non-linear transfer functions. ...Jim Thompson

Yes, thanks. That's similar to the sort of things I've been doing, except I've been using N-ch instead of P-ch, for the better Ron x dollar product. Also, my designs don't destroy the first FET during an input transient. :) (voltage on source follows Vin, gate is grounded.)

By the time I've made it robust enough to expose to customers, it's no longer small nor cheap.

Regards, Allan

No, he has since sold the ice company, said it was one of the best days in his life! As a side note, the ice machine used ammonia as the refrigerant, apparently the meth makers can use ammonia in part of the process producing meth. He lost several bottles to theft. I helped put in cameras and a recording system, but he never caught anyone.

Back to illegal aliens, we had this happen in our city.

"Vehicles are towed after being abandoned on the Thomas Drive flyover by illegal aliens Thursday.PANAMA CITY BEACH

Seven cars full of illegal aliens abandoned their vehicles on the Thomas Drive flyover Thursday evening and ran from sheriff?s deputies.

Authorities don?t know how many people ran from deputies, but two men were arrested. Both were charged with resisting arrest with violence, said Ruth Sasser, spokeswoman for the Bay County Sheriff?s Office.

The seven cars were ditched as deputies were responding to an unrelated incident on the flyover.

?We feel that the people in the cars thought we were after them,? Sasser said. ?They just pulled over and ran.?

The confusion started when"...

This was back when we had a condo boom and had many illegals here. I have friend whose wife is a kindergarten teacher, one year she had

Mikek

Well then just pretend that he's going to be an engineer and not a goalie.

He's super with math, and loves the circuit lab upgrade I gave him at Christmas. But he's struggling with reading... a bit of dyslexia... time to hire a tutor (I would do it, but he needs an authoritarian type to keep him from deviating from work and controlling the conversation... he's a charmer... seeks out the girls in the crowd... and he's only 8 ;-) ...Jim Thompson

I first kissed a girl at 7, and was made fun of more for that than for doing experiments.

Today, you'd be expelled (for either).

The amps in old Tek scopes also go bad, and there's a market for replacing them too. I suggested years ago that somebody should make a DIP module to replace them. Today's standard op-amps could do it I bet.

A lot of Tek scopes use weirdly high supply voltages.

Cheers

Phil Hobbs

There's something very rewarding about creating a PCB. It may be similar to the feeling from creating an IC, but in a way it's even better: seeing your idea ready for *permanent* inexpensive replication by anybody.

I often create a design I'm proud of, and watch it made (breadboard-style) and put into use, but if I go the extra mile and create a PCB, then it invoke a new level of feeling entirely. E.g., the RIS-761 pulse-picker project I put up here yesterday. See my dropbox folder with photo, a design anyone can make, complete with a laser-cut panel and a PCB to slide into a Hammond box.

I think PCB layout is a skill that's well worth any circuit-design EE's time to learn.