Crowbar vs Transorb (TVS)

[aerhed]

Can anyone 'splain why a crowbar is used/better instead of just a transorb from the field breaker to ground? Won't a TVS reliably trip a 5A field breaker in an OV condition? What am I missing?

[Ted Johns]

Can you "splain" why you think a TransZorb across the field circuit of an alternator would be of any use?

[aerhed]

TVS from load side of field breaker to ground. If breakdown voltage is exceeded won't the field breaker trip instantly? Isn't this the same action you get from a crowbar circuit?

[Ted Johns]

And if the TransZorb breakdown voltage isn't exceeded? Errors in the control circuitry could result in legal field voltage and current, but excessive output voltage from the alternator under light to medium loads. Just throwing a clamp across the field circuit isn't any kind of properly designed OV solution.

A proper solution measures the output voltage, and opens or crowbars the field circuit when the output is illegal.

[aerhed]

Isn't it the bus output voltage you're regulating? Why would I care about voltage upstream of the bus? If breakdown voltage isn't exceeded at the bus, isn't everything okay? Won't tripping the field always take it off-line? Why clamp the output instead of the field? Don't just tell me I'm wrong, please explain.

[aerhed]

As I understand it, the denso type alternators have a transistor that can fail in the regulator output which full fields the alternator with no provision to actually turn off the field in this condition as the field + is derived from the B+ terminal internally. The "switch" wire turns on the regulator but the field positive is already there looking for a ground through the previously mentioned transistor. The modifiers, i.e. Plane Power, reroute the positive field brush from B+ to the "switch" wire giving positive control of the field. They also put a crowbar circuit (sensing output) to short the "switch" wire which trips the field breaker and takes the field out. Now, for like the 3rd time: Why wouldn't a 16V TVS diode from the dowstream side of the field breaker to ground instantly trip the field breaker the same as a crowbar circuit with a 16V zener. Isn't a crowbar exactly the same function as a TVS except it includes a cap to soak up transient spikes or high ripples?

[Ted Johns]

Don't just tell me I'm wrong, please explain.

The problem with not speaking directly to a schematic is that the communication will be constantly subject to errors of assumption and misinterpretation.

If the TransZorb was at the breaker, rather than at the field, a number of faults might be caught. But not all of them. And this design might only work once. Breakers take long enough to trip, even with a large overload, that the TransZorb may reach it's thermal limit and die. They are designed to catch very short glitches, and they don't have the ability to shed watts for long.

If the regulator fails in a way that shorts measured output to the field, the scheme above will be useless. The alternator will then be essentially self powered.

A more foolproof method of OV protection would have a circuit monitoring the output voltage that can directly block the field current, i.e. is downstream from the regulator. Of course, these circuits are properly designed to work with the regulator, as it can simplify the design, and makes sure one does not destroy the other. This circuit can also be tailored to ignore high speed glitches, and thus reduce nuisance trips.

Which is why I assumed that's where you wanted to put the TransZorb.

[elippse]

Originally Posted by aerhed

Can anyone 'splain why a crowbar is used/better instead of just a transorb from the field breaker to ground? Won't a TVS reliably trip a 5A field breaker in an OV condition? What am I missing?

The crowbar, which makes use of an SCR to short out the field suppy and take out its breaker, is triggered by an overvoltage from the alternator output, not an overvoltage on the field. If you used a 16V TVS at the field supply. all it would do is clamp the field voltage, but not reduce it. If the field voltage is already that high at high rpm, it is already an indication that the regulator has lost its ability to control the alternator output. In this situation, the alternator output is on its way to very high voltage, and even with a 1500W or 3000W TVS, the TVS will be driven beyond its capability and burn up immediately, and may go open, in which case the field breaker won't pop. In checking out circuits that had a uni-directional 1500W TVS on its power input, the Bench Power Supply was generating transients as its voltage was raised, due to some internal relay switching, and it was taking out the TVS. They're good devices when properly applied, but look at their little DO-18 or DO-201 diode-style package; they won't take much power for long! I made my own PWM regulator with a series mosfet that would open if the voltage went beyond 14.7V. It has a fold-back circuit that will restore the output when it drops back down through 12V.

[Ted Johns]

Isn't a crowbar exactly the same function as a TVS...

Nooooo.

A TVS clamps at the voltage it is rated at, it does not become essentially a short, as a crowbar does.

[aerhed]

Ted, okay sinking in. elippse, I was thinking that the field breaker sits at bus voltage so the TVS senses bus voltage ala Grumman tiger, not actual field voltage which on these regulators I believe is controlled by varying the ground.