Antique car load dump

My wild guess is that a generator has a lower output impedance than an alternator, so the load dump won't be as severe.

You don't say what era. If it's 50s, they used vibrating relay regulators, so a rather different setup. If 1920s, things were much cruder, and the gen typically wasn't regulated electronically or electromechanically. It was a nything from no regulation whatever to adjusting the gen commutator to pick off less or more overall charge rate, still with no instantaneous regulati on whatever.

If you can tell us what era or charge circuit, at least it becomes possible to address your question, though I personally can't answer it.

NT

This says "50 V and 50% of rated output":

I believe the diode arrangement for 12V is quite different than for 6V pos. gnd.

Mid thirties to mid fifties.

Cars from the vintage being discussed didn't have alternators with diodes, they had generators with brushed. This info just doesn't apply I think.

I'm wondering how you'd get much load dump on the oldest stuff, with such low charge current feeding a big battery. Charge current was single figure amps in the 20s, memory's rusty on 30s but still very low compared to anything modern.

NT

Part of what I was hoping for was to get a handle on how much load dump to worry about -- not only is the charge current low, but there's not that much load to dump. Most electrical systems of that era only supplied lighting, and maybe radios -- there weren't all sorts of gizmos attached to the thing.

In the 30s it was just ignition & lighting, no-one had car radios. Car radi o was in the experimental domain at that time, with longish wire aerials st retching the length of the car feeding reaction sets powered by vibrators - a setup too unstable to be practical, and maybe too noisy. Usable for mili tary comms where steep cost & the need for constant tweaking were livable w ith.

The 50s was different of course. As you say I don't see much of a recipe fo r load dump with the earlier cars. Can you not measure one? Or apply genero us protection?

NT

Only tangentially related, but I remember reading somewhere that two pairs of three 1N4007 diodes in parallel is a very reliable replacement for those "0Z4" gas rectifiers they sometimes used with vibrator power supplies...

By 1940 or so radios were a fairly common option -- but yes, in the early

FWIW, automotive tubes -- those which boast a heater with a wide voltage tolerance, to accommodate the wide range of low to overcharged battery voltages -- don't mention load dump, ever, so it might not've been a concern.

Tubes are basically light bulbs, and vibrator supplies probably would've arced over then recovered (don't they spark enough normally?). No one needed TVSs or GDTs in those days, and when they did, NE-2's were plentiful and effective (often, in direct coupled circuits, to assure DC bias and safely limited grid voltage, as Tektronix used many times).

For that matter, in the time I've been driving a car, or in one, I've never seen a load dump event; even if they were more frequent back in the '50s, I'd be fairly confident that I'd have to replace several tubes and the vibrator by now. In other words, tubes have shorter lifetime anyway, so that they are unlikely to see such an event. It's not even worth protecting against, even if they were vulnerable.

Also, imagine that: automotive electronics you could replace and repair!

Tim

The actual amount of energy delivered in a load dump event isn't much, and tubes are pretty tolerant of that.

Since I'm contemplating building a gizmo out of modern electronics to go into an antique car, I need to worry about load dump.

Well, energy relative to the... yeah...

"Semiconductors have ratings, tubes have guidelines." ;-)

(A late-model sweep tube can be operated "short circuit" (~8x rated dissipation) for 40 seconds, until everything else warms up and oscillators start and bias establishes. They even put that in the datasheet!)

On the upside, it sounds maybe like a modest size MOV would do a fine job (or TVS, but those are rather expensive in comparable sizes), and you could use a HV input LDO for ride-thru. Whereas a modern alternator might cook the MOV (I forget how many J is in a standard load dump; 60V * 120A * 300ms divided by load match and curve factor?), but the ride-thru would be fine if rated for a bit more.

Tim

For the fussy op-amp stuff, I was thinking of a series resistor and a zener in shunt (or a TVS). The resistor would lighten the load on the TVS.

I dunno if I'm actually going to design the thing -- my brother pointed out the need, but the market is small enough that I'll basically be throwing the work away, and I don't even own one of those cars.

How come Antique Electronic Supply are so stingy on the warranty for NOS sweep tubes, then?

Sounds like the question needn't be answered.

NT

Depends on how you define "need". I was curious -- isn't that enough?

Because they tend to get run hard in general, whether they're in a TV or an amateur radio transmitter.

Yeah, "lifetime" ranges from seconds (burning it up "short circuit") to a few thousand hours (more or less at ratings).

Not at all like semiconductors, where if kept modestly within ratings, lifetime can be exponential, hyperbolic even: millions of hours MTTF.

Tim