joule capacity of surface-mount resistors

You ought to TDR that thing :)

I'd

uoted text -

Hmm, I don't know the 500 uohm/square number... per square inch of Cu?

I figured you'd need ~5 square inches of 1 oz. Cu to make a gram.

So maybe longer and wider (2 oz Cu)? Well unless you can figure on the Cu dragging a bunch of the epoxy along with it.

George H.

ed text -

dea.

nt

reliable

Ahh, can you use something like that for the current problem? (too big?)

George H.

Hmmn... if it's what I'm remembering, those look really great but have a crappy thermal pathway from element to case. I've seen them fail as you describe.

That does look nice, at least for pulse power. For continuous you'd have to have some good heat path.

I prefer to use electrolytics as weapons of mass destruction:

Yup, and if your pulse is fast enough, the heat sink doesn't matter. It's all I^2*t, so it comes down to the volume of the current carrying portion, its thermal capacity, and the maximum temperature it can withstand. Perhaps a length of baling wire would work as well as anything else.

describe.

The Welwyn things are thickfilm on porcelanized steel. They are slightly cupped so that when you bolt them down, they hug the mounting surface. Neat.

text -

Per square anything of 1 oz copper.

Yeah, 35 joules into a gram of copper is about an 88 deg C temperature spike.

The epoxy would probably help, as would adjacent power or ground planes. So one gram is probably OK. But 5 square inches, maybe doubled for traces+spaces, is a lot.

Looks like it lost a pie throwing contest.

--

John Larkin Highland Technology Inc

Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators

With luck, thousands.

What looks practical is to stand a wirewound resistor, roughly 0.25 dia x 1 inch long, on edge. Inelegant.

Hey, copper is a pretty good RTD. Some clever circuit could dump the joules into the coil and measure the temperature before it cools off.

An assembly house had once changed Ohmite OY-series 100ohm resistors to carbon film resistors. The guy calibrating the device wondered why the device made a strange *phut* sound on every measurement.

Opening the device revealed that the resistors shot out 20mm flames on each measurement! Apparently the films did not like to 3kW peak / 20J load..

The Kanthal Globars you have quite a high maximum operating temperature!

inch

I remember those green "vitreous enamelled"" wirewounds from my youth, you could put them on car batteries and they would glow orange-hot, melting the glass. You could see the coil through the glass.

Didn't seem to do them much harm at the time, but hardly a scientific test!

t? I'd

ic

to

I'd

It

com

e quoted text -

Got it.. so length/width is your 6,000. (for three ohms) If area is 5 sq-in. I get a width of 30 mil and a length of ~180" (ouch!) For your long time scales you might guess you get twice the volume of fiber glass dragged along... So 1/3 the area...

George H.

ike.

So one

, is a

.highlandtechnology.com  jlarkin at highlandtechnology dot com

inch

Why not mount it horizontally, but elevated, and put circuitry underneath?

Cheers

Phil Hobbs

inch

This board is mainly paved over with tall stuff, caps and heatsinks and connectors. I'd prefer to not cover up the little opamps and things either, cause that makes them hard to probe. I do have a niche in the corner, about 0.3 diameter, that's unused, and I could stand up a wirewound there.

Packaging is 50% of electronics. Actual circuit design, the funnest part, is maybe 10% if that.

inch

Yup. On the plus side, I just finished specifying a $100k digitizing scope as the back end for a demo instrument. The immediate customer is enthusiastic, partly I think because they'll get to keep the scope when the demo is finished. ;)

Cheers

Phil Hobbs

and

open

inch

Hmmm, comes out of a different departmental budget. Maybe all they want is the scope!

We recently got a LeCroy (now Teledyne) 4-channel 7 GHz scope for about $50K, Windows based. It does some cool things, when willing.

I guess you can run scope apps on it; haven't looked into that.

Really, it's more logical to get a knobless/screenless digitizer box and connect it to a PC.

I'm

surface-mount

long.

the

that

wirewound

seeing